Hi Dreyfuss-Hudson,
Following my visit to the NRM last month, I have managed to obtain the two volumes of The Science of Railways: The Steam Locomotive by Dr Lee Towers and am working my way through these. These books formed the basis of a course given at the University of Leeds School of Continuing Education in the late 1990s. I spoke to the author who kindly sent me the books. He is a physicist and as that is also my own background I find his approach more systematic and meaningful than what I've read before, which has been written mostly by engineers,
Neil
Dear Neil,
As your data show, I don’t think that within limits, mechanical considerations such as wheel diameter:stroke (D:S) ratio had much influence on ability to reach 100mph. The D:S ratios of British locos that reached 100mph lie in the range 4.15 (Britannia, 74” drivers, 28” stroke)-4.89 ( Scot, 81” drivers, 26” stroke). The MN and LNER Pacifics are at the top of this range, about 4.83, also the GW Castle. Only the LNER Pacifics and GW Castles were ever scheduled to run at 90mph on a daily basis. In this D:S range however, the mean and maximum piston speeds vary by only 5 ft/sec at 90mph, about 15%. All types that succeeded in reaching 100mph had 74+” drivers, though none with smaller drivers could develop enough power to make 100mph except down very steep grades. The N&W J has a D:S ratio of 3.44 only, 70” drivers, 32” stroke, and one of these famously reached 110mph, though this was more a stunt than demonstration of daily capability. The speediest locomotives on a daily basis were the MILW F7s, with a ratio of 4.4, 84” drivers. It is said that the ATSF 2900s ‘regularly’ reached 100mph and their ratio is 3.93. So I don’t think D:S ratio in the range 4.2-4.9 is a critical consideration.
There was no apparent correlation between stroke and clearance volume on UK locomotives, CV being more of a design choice. The GW believed low CV was good, hence went for 6% with 26 and 28” stroke. The LMS had 12% CV on the 28” stroke Duchess, BR standards with 28” stroke were also generally 10+%. LNER/Bulleid went for about 8%, with 24-26” stroke. CV data on US locomotives is hard to come by.
The idea that Low CV is good for efficiency is true at low speeds; at high speeds it is a positive handicap because of compression loops as the GW found on the King. At high speeds the effect of increased CV on engine efficiency is pretty small, certainly in the range 8-12%; remember also the HP cylinders of the French Compounds had CVs of 24+%! I think the myth of CV effects on engine efficiency at speed is one of those that will outlive us all.
I have no information about the riding of 60163, though on all the head on videos of it I have seen heading through curves at high speed it looks as steady as a rock, likewise 60532. I will ask a friend who might know. I did read somewhere that whilst the Peppercorn Pacifics were worse riding than the Gresley ones, the real culprit for the reputation was the state of ER track in the immediate post war period- it had deteriorated badly during the war. This highlighted the differences. There was 60 mph limit out to Hatfield for this reason well into the 1950s.
Dear Dreyfuss-Hudson,
The comparison A2/MN/BR8 is interesting. Where they differed was in their piston strokes. The short stroke (24 in) MNs had the same piston speeds as an A4. This allowed a genuine 100+ mph engine in a relatively small package. The compromise was probably lower cylinder efficiency. With a short stroke, other things being equal, clearance volume is higher, reducing cylinder efficiency. In Britain and Germany, 26 in stroke was most frequently chosen. Golsdorf was an early European advocate of relatively long stroke (28.34 in) and later Austrian designs retained this. Probably Golsdorf saw this as a weight sving measure and he was working within a 14.5 mt axle load.
Laterly, there was some move toward longer stroke The German class 10 had 28.34 in. Cylinder efficiency was probably better but machine resistance increased. However, some American engines with 32 in stroke still easily reached 100 mph.
The main fault of the A1/A2s was their indifferent riding (A2s the poorest). Do you know if Tornado is any better in this respect? My understanding is that the A4/MN layout was better in having a lower rotational moment of inertia (MOI) about the central vertical axis. The A1/A2s had higher MOIs through the forward location of their middle cylinder, so the energy of the swaying motion that needed to be damped out was higher. Also their short rigid wheelbase would have likely made them less steady. This was inherited from Thompson on whom it was forced by his ackward cylinder layout.
Thanks for this. BR’s sole 8P Pacific 71000 was of course largely designed by Harrison of the LNER and Cocks from the SR, and so not surprisingly it has three cylinders! I always see this as a fight back of these factions against the dominance of the LMS mafia that ran the BR ‘Standard’ programme, seizing the opportunity presented by the loss of 46202 at Harrow in 1952. In essence basic boiler dimensions are those of an SR MN and LNER A2, and it shares their 74” drivers. The cylinders are 28” stroke in line with other Standards, and doubtless many other components were ‘Standard’- not that this guaranteed best practice was being followed! I can’t recollect why they went for Caprotti gear, but this was certainly seen as the future at the time. So I think of the Duke as basically a Caprotti A2. What the BR designers would not tolerate of course was the fancy continental exhausts espoused by the LNER and SR, so it finished up with a plain BR DC. At Swindon, it could, as a consequence not be steamed satisfactorily above 26000lbs/hr on the test plant, though with better firebed airflow on the road, they did make 30000lbs/hr, still very poor for a 48.6 sqft grate. The Kylchap now fitted has transformed things of course.
The defenders of the LMS/BR faith still think that the plain BR DC is wonderful of course, and that the Duke’s poor steaming was due to excessive spark loss due to the Caprotti gear.
To try to get to the bottom of this, I have had a go at applying turbulent jet theory to locomotive exhausts in the past, unsuccessful because it was not possible to estimate the momentum loss in the chimney (which determines draught) from simple dimensions. I’m having another go at this, having found that the University Illinois report on draughting their ¼ scale model boiler with a large variety of draughting devices is available in full on the internet.
https://www.ideals.illinois.edu/handle/2142/4435
www.ideals.illinois.edu/.../engineeringexperv00000i00274.pdf?...3
Whilst turbulent jet theory is very straightforward for an orifice discharging into a uniform, stationary pressure field, as soon as you go to more complex systems, you need to do experimental work to modify it appropriately, and the Illinois data may provide the information needed. However, this really is way outside my area of expertise, and I’m not sanguine I can come up with anything useful. Success would mean putting to bed the plain DC vs Kylchap argument once and for all.
The rebuilt MN did not appear until 1956, and by that time, BR had abandoned the idea of steam (a decade behind the times!), so could not have been the basis for a BR Class 8. Detailed engineering considerations apart, it would have been scarcely possible to improve on the LNER A1/A2. Really good to see A1 Tornado and A2 Blue Peter proving the point still.
Dear DreyfussHudson,
Thank you for the interesting information about the Merchant Navy tests.
I have just acquired an old book (1974) about the Bullied pacifics. In it there is an account of how the BR engineers initially tried to improve the Merchant Navy class. The Lemaitre blastpipe was considered non-optimal and it was felt a standard BR type single blastpipe would be better! They found that whatever they did that a single blastpipe limited the engine to about 31000 lb/hr. This is not really surprising as the slightly smaller V2 also had a single blastpipe limit of 31000 lb/hr at a relatively high coal consumption of 5760 lb/hr (Clay and Cliffe). The 'unsatisfactory' Lemaitre gave 39000 lb/hr for 1 hr and 42000 lb/hr for 20 minutes, which was not the absolute limit but thought to be near it. Perhaps rather surprisingly, given their preconceived ideas, they decided to retain the Lemaitre system even in the rebuilds. Far from showing the advantage of a single blastpipe, their tests showed the considerable increase in power available from a more advanced exhaust. Probably the superheat was a little higher too at the higher outputs.
Winkworth, who gives many timed runs of both original and rebuilt engines, says the coal consumption was not reduced by the rebuilding although the oil consumption was. But the performance, in terms of power and speed, of both was entirely up to class 8 standards, while keeping to quite modest size and weight limits, especially in the originals.
BR might have done worse than to make the rebuilt MNs the new standard 8P, rather than embarking on a completely new design, which would initially inevitably have teething problems, and meant needing to keep an additional set of spare parts.
I have a friend who has created a first principles model of the machinery resistance (those between the pistons and the wheel rims) which would include components which can be fitted with roller bearings. The model is in good agreement with the ‘good‘ Rugby data. For an MN, the value is about 150 HP at 70 mph. (The model needs some adaptation for US designs). Only 40% of this is due to Journal friction and Coupling rod losses, which says that you are working with about 60 HP with roller bearings. Total resistance between the cylinders and drawbar is about 540HP at this speed, so any benefit for roller bearings will be relatively small.
The Rugby MN bulletin is the only one from that testing station that I believe is in significant error. You can see this by looking at difference between the reported IHPs and those measured at the wheel rim by the Amsler dynamometer. The values are generally negative! If you add a sensible value of machinery resistance to the wheel rim powers, what you find is that the difference between the measured IHP and the IHP that was probably being developed, as estimated by this method, is dependent on cut off. At long cut offs, and low speeds, the values are in reasonable agreement. At high speeds, the measured IHPs are up to 15% too low, a problem I believe with measuring areas on ‘thin’ indicator diagrams. Other locomotives tested at this time also show this problem, though the bulletin results seem to have been massaged and are basically correct. Rugby beefed up its expertise on indicators in 1953 and this resolved the issue.
If the original MN cylinder efficiency was as bad as you say, then its drawbar efficiency would be poor. In fact,on full regulator, it is just the same as the Duchess, confirming the hypothesis above. There was already a strong body of opinion that the class should be rebuilt, and the Rugby tests supported this. The results were however wrong! In good condition, the original and rebuilt locomotives have similar efficiency. The efficiency of the class was not as good as it might be a) because the thermic syphons fitted reduced superheat b) they did not have exhaust injectors and c) they were generally far too powerful for the work required (especially at the original 280 psi boiler pressure), which meant that they were habitually driven on the regulator/throttle often 100-120 psi only in the steam chest, with the valve gear set in a very long, inefficient cut off.
Spent a number of happy holidays in Ringgenberg watching the steam paddle steamers on Lake Brienz, and riding them to the Reichenbach falls, where, as we know, Sherlock Holmes met his end fighting Moriarty- also listening to the exhaust chatter on the Brienzer Rothornbahn. One summer had a steam service on the main line between Interlaken and Brienz too! Hard to beat.
Sorry for the delayed reply but I was on holiday in Switzerland. There, there is still some steam on the public timetable. I was on a paddle steamer, Lotschberg, on Lake Brienz. This was built in 1914 and still has the original 2 cyl compound condensing engine: 550/950 x 1050 mm, 14 atm, 300 F 450 IPS. Not much but the boat can carry up to 800 passengers. And it is smoother with less vibration than any diesel. One steps off the quay and then almost on to the valley station of the Rothornbahn. This converted back to steam in the 1990s with new steam engines. It was open only to the half-way mark, but spectacular is almost an understatement about this railway. The engine was No 12 built in 1992. More powerful than the older ones and able to go up smoothly without much of the familiar jolt-jolt of steam cog. An unusual feature was the location of the water gauges. Both on the right hand firebox side outside the cab. At Interlaken there are two small operational metre gauge engines, one a 2 cyl and the other a 4 cyl with both adhesion and rack drives. There is also a standard gauge project: the frames and cylinders are now complete for a 2-10-0 4 cyl compound engine of 1917 known as an Elephant. These lasted till the 1950s and some I believe were lent to the Germans in the War.
Regarding the Class 10s the main way they were state of the art was in having all roller bearings, including for the connecting and coupling rods. This would not have increased their IHP but might have reduced the difference between IHP and DBHP.
Were all test results optimistic? Probably not. Clay and Cliffe also give figures for the original unfancied Merchant Navy class: 2480 IHP at 42000 lbs/hr sustained (16.9 lbs/IHP hr). This is probably an unvarnished figure given the enthusiam to rebuilt them. Another well known low rating was when an S3/6 was tested at Grunewald in the 1930s. The steam chest pressure was about 20 lbs/in2 below the rated boiler pressure, and these engines could sustain significantly more then the Wagner boiler rating of 57 kg/m2 hr (as could the 38s). The official raating of the S3/6 of 1830 PSi appears to give 13.55 lbs/PSi hr, but this figure was neither the lowest consumption nor the maximum sustained.
I'm a bit sceptical about the 12.1lbs/ihp-hr claim for the 01-10, not least because I think the 12.2 of the Duke is significantly overstated. The raw data from the Swindon tests is at the NRM, and if you look at this then a) the test bulletin takes a rosy view of what they actually measured and b) the cylinder powers they estimated were in any case probably about 5% high, meaning that I doubt what they actually got on test was better than 13lbs/ihp-hr, slightly but not significantly better than the Britannia - Caprotti brings very little in the way of efficiency itself.
The minimum specific steam consumption is determined by the fact that superheat, hence efficiency increases with steam rate; working against this is the fact that as steam rate increases, backpressure increases, and longer less efficient cut offs are needed to consume the steam. So the 01-10 would require very high superheat at lower steam rates than UK types for the claim to be true.
Alternatively, conventional designs lose a minimum of 5% efficiency through leakage of various kinds, often more, through valve and piston rings, stays etc. Now if the Germans were better in these respects, the claim would be more credible. Vorsprung durch Technik, as they say.
I have done some first estimates of the steam rates at which the GW jumper cap would lift, and it seems that this is less than the economic limit i.e. not in accord with my original postulate. I'm hoping to get an independent verificaiton of this; if correct, the question is how come the drop in inlet momentum due to the lifting of the jumper does not reduce draughting.
Yes, the A4 was by reputation the British pacific with the highest superheat annd this probably allowed it to nearly equal the output of the larger Pappercorn A1 and also make the fast non-stop runs from Edinburgh to London without the need to refuel.
Another European simple pacific which had apparently high superheat was the German class 10. Clay and Cliffe (West Coast Pacifics) in an international comparison of pacific tests put that maximum sustained output of the class 10 as: evaporation 39700 lbs/hr; IHP 3000; steam rate 13.2 lb/ihp-hr. The official rating of the class 10 was 2500 PS. The 3000 figure is in no way exceptional for an engine as large as the 10. The corresponding figures for a Duchess were 42000 lb/hr, 2910 ihp and 14.5 lb/ihp-hr. Where the 10 apparently scored was in having a distinctly low steam rate at maximum output. Clay and Cliffe say its minimum steam rate was 12.07 lb/ihp-hr, marginally lower than the BR8 at 12.2 lb/ihp-hr, but probably in the margin of error. In practice the 10s were overweight and restricted in the routes they could operate on but German limitations on axle load were greater than British. Clay and Cliffe say the 10s exceeded the 100000 mile per year target. That was probably true to begin with but 10 001 preserved at Neuenmarkt-Wirsberg ran a lifetime average of just under 60000 miles a year.
Clay and Cliffe also give test results for the US K4S. Piston valve in 1937: steam rate at maximum output 19.8 lbs/ihp-hr (3530 ihp @ 700000 lb/hr). And with Franklin poppet valves: 16.4 lb/ihp-hr (4267 ihp @ 70000 lb/hr) with a minimum rate of 15.0 lb/ihp-hr (3190 ihp @ 47770 lb/hr). Of course, these were old engines and not representative of the best US types.
I agree there is a sweet spot for efficiency, since as steam rate increases, boiler efficiency decreases and engine efficiency increases due to superheat. Boiler efficiency starts to deteriorate significantly due to increased coal loss above about 600lbs/sqft/hr. The reason for initiating this thread in part was for me to understand what steam rates were typically required to maintain schedules in the US, and it seems it was about 500-600lbs/sqft/hr. In the UK the smaller 4-6-0s were steamed at 600-700lbs/sqft/hr. Whether this ‘glass ceiling’ was a consequence of crews not wanting to waste coal, or a desire to keep boiler maintenance costs down I do not know. Maximum outputs above this level, as quoted for US, UK and French types, were in practice irrelevant.
This says that superheat is primarily what matters for overall efficiency. Reed was a very knowledgeable writer, but I think his estimate of 320 deg C as a typical maximum in the UK is low, for some types at least- though you wouldn’t get above that with a Duchess or Princess on a run to Scotland. 60007 has a pyrometer on the footplate, top right of the backplate. There are some footplate runs on You Tube (put in 60007 footplate). One clip shows 60007 leaving Retford from a water stop, climbing the 5 miles to Markham with 500 tons working quite hard. Going more easily down the other side the pyrometer shows 340-350 deg C. Another clip has it climbing the 1% from Settle Junction to Helwith Bridge, and it shows 360-370 deg C after about 10 minutes I guess. These figures are about 20 deg C below what one might expect the equilibrium value to be, looking at the V2 test plant results (very similar boiler). I was talking to a friend yesterday who has spent a lot of time on the Duke footplate on the main line. This also has a pyrometer. He said it took about 15 minutes for it to reach a maximum equilibrium value of 400degC, when working hard. Having said this, we do treat our Pacifics in quite ungentlemanly fashion these days, often whipping these old iron horses more than old Jockey club rules allowed.
I agree some the UK power classifications were overstated.
I guess at 60 I come into the category you describe. After 20 years in Austria I find there is much more steam here. Passengers on diesel trains are mostly in the 20 to 40 age bracket, while on steam turns it's nearly all 60 plus year old men. Comfort is also likely a factor: the old 2nd class Mark 1s were more spacious. The current standard class is nearly all high density commuter-type stock. I believe the HST 2nds started off with 72 seats but are now 84 seaters. So anyone over 5 ft 7 in is cramped. In the Mark 1s everyone had a good view out both sides, now anyone in an aisle seat has nowhere to look. Few travel in 1st class. On steam few travel in 2nd, Premier Dining Class is the most popular (£200) and there is sometimes Pullman above that.
Coming back to more technical matters, Brian Reed wrote in his book commemorating 150 years of British steam that superheat attained in daily work was never high in Britain. Anglo-Scottish turns rarely reached 620 F band the average throughout a run was usually only 550 F. Maybe this is why the low superheat GWR types did not stand out as very inefficient in relation to the others.
For efficiency, the boiler effeiciency and superheat seem the most important factors. And unfortunately the work against each other. The boiler efficiency peaked at a fraction of the maximum output and fell in a straight line thereafter. Superheat on the other hand was low at low outputs and then rose significantly. Cylinder efficiency peaked at a fairly low average cylinder pressure and then rose a little more than linearly with average cylinder pressure.
The Kychap was about the only unambiguous improvement: better draughting and increased power.
BR perhaps made a mistake with some of the power ratings: the King was more 7MT (on account of its adhesive weight), the Castle 6MT and the Britannia nearer 8P6F.
The reputation of the GWR types was made more on short steep banks. Last Sunday I saw an ancient Castle 4-6-0 (the 76 year old Earl of Mount Edgecombe) come through Bristol Parkway. It had at least maintained the 14 minute allowance from Temple Meads with a 12 coach train on a very wet morning when adhesion on the 4 miles of 1 in 75 might have been a problem for some more powerful Pacifics. But this does not shed any light on the jumper ring as this is one of the engines with a double blastpipe. I recall the heaviest turn in the 1980s was the 12 coach Plymouth to Edinburgh, which was always down to about 30 mph on the curved part of the bank. But it was baulked to 20 mph at the foot of the bank by the weak bridge over Stapleton Road, closed for many years. The Castle was probably doing 45/50 mph at that point.
Yes, I agree that all sorts of economic factors other than thermal efficiency killed the steam locomotive. This discussion about efficiency is really about the era when steam was the only game in town, and you had to make the best of it.
It’s maybe worth bearing in mind that in this country, the decision to get rid of steam was only taken in 1955. One plan was to import diesels with proven technology from the US; politics killed this idea, so local manufacturers produced a lot of underpowered and initially not very reliable diesels. So, in the summer of 1960, the world still was pretty much in order, most everything steam hauled, shiny passenger locomotives. Even in 1961, with a bit of myopia and optimism you could believe things were ok, 15 years after the corresponding era in the US. The consequence of this is that there is a whole generation of folk in their 60s, 70s and 80s over here with time and money that remember how it used to be. The consequence of this is that between now and the end of September, there will be more or less a steam express train every day on our main lines, with 60-80 year old 4-6-0s and Pacifics with heavy loads mixing it with high frequency modern trains, often travelling at 125mph. In addition there are at least a dozen tourist lines that need 3-4 locomotives in steam to run their daily services, and a host of smaller ones. Scarcely believable, certainly not foreseen.
So, as you can imagine there are numerous small businesses and a veritable army of volunteers to whom steam traction is still a live issue. How does it feel living in this time warp? Well, I can think of a lot worse things! Imagine if your daily commute from Glenview, Il was enlivened a dozen times a year by a Hudson plus a reasonable facsimile of the Afternoon Hiawatha racing through at track speed. Sure you’d get covered in coal char, but is it that bad?
The Problem with Steam Locos was not so much the cost of fuel (After all N&W and the Laccawanna owned huge coal mines) but the cost of a army of craftsmen to fix and maintain the locomotives
Ell's comments on Castle superheat refer to what he thought would happen, not actual test results. 2, 3 and 4 row superheater Castles were tested subsequently. Only the results on the 3 row Castle were published. The original report contained some nonsensical efficiency diagrams which Ell later corrected, although the efficiency claims for this Castle remained unchanged, even though I am pretty certain they are significantly overstated, for reasons I think I understand. Part of the benefit of raising the superheat was lost by narrowing the blastpipe.
I think from a driver's perspective, a 3 row would scarcely be any better than a 2 row. Worked in the same way i.e. same cut off, power would not increase. What would happen is that steam hence coal consumption would go down by several percent. Enough for a crew to throw their caps in the air? I'm not sure.
With a four row superheater and double chimney, you do get a step change in performance at a given cut off, and the better efficiency and higher darughting limit (no jumper cap!) do allow outputs in a different league to the original two row single chimney types.
I am beginning to think that, whilst increasing superheat is surely by far the best way to improve engine efficiency, the benefit in practice may not be as large as test plant results suggest, for the simple reason that it takes a significant amount of time for equilibrium superheat to be reached, and if the road is undulating, or with frequent speed restrictions, the equilibrium value may not be reached. It's one thing climbing 5000' from Topock to Wampai Az in 120 miles, another working the Midland main line from St Pancras to Leicester, constant undulations with necessary variations in steam rate, four separate summits in 100 miles.
My suppostion is that lower superheat classes get closer to their equilibrium value sooner than higher superheat types do. See Wardale for some interesting on the road test results on superheat.
Thank you for your informative reply. The idea that the jumper ring was mainly about economy may make some sense. The South Devon banks are very steep but short and Dainton and Hemerdon can be climbed in less than 4 minutes while the very steep part of Rattery takes less than 7 minutes. The jumper ring opening at the bottom increases power temporarily, while keeping ejected coal down (Reed says LMS types ejected 30% of their coal on the last four miles of the climb to Shap). By the top, boiler pressure is down, which means that the safety valves do not blow when the regulator is shut for then descent, another economy.
Rutherford gives details of tests with various levels of superheat on Castles in 1946. At 240 rpm with the standard superheat, IHP was 1500. The highest superheat boilers gave 1720 IHP (essentially the same as a low superheat King) it says). Summarising Ell says:
'... an increase of 14.5%, and it may be noted that this extra power is obtained with a normal rate of combustion and with a normal boiler efficiency. Hence it is obtained at a far less cost than for a similar increase by methods requiring a higher draught, where the extra power is obtained always at the expense of lower boiler efficiency by reason of excessive loss of unburnt fuel.'
Many thanks. In a separate correspondence I am engaged in, I have been trying to work out the mysteries of the GW jumper cap. I have the GW original drawing and the details of the size of blastpipe cap and weights of jumper ring fitted to different classes- I hadn’t appreciated almost every original GW type had them, except their higher superheat designs from the late 1940s on. My understanding is this change led to carbonisation problems, hence they were abandoned. This will have also been a significant issue on the high superheat A4s.
Thinking about it, these are basically yes/no devices, which lift when the exhaust pressure on the inner cap surface exceeds the weight of the cap (20-25 lb). I think of draught as being the consequence of a momentum differential across the chimney choke.
Dp = Mi – Mo/A
Where Dp is smokebox vacuum.
Mi is chimney inlet momentum.
Mo is chimney outlet momentum.
A is the choke area
Obviously, when the cap rises, there will be a sharp drop in inlet momentum, and special circumstances apart, the outlet momentum will drop also, so there will be a sharp decrease in smokebox vacuum, hence draught will be less. Not a recipe for going faster.
I think the logic of Rutherford and others, which up to now I accepted, is that the lower backpressure when the cap rises will allow a more efficient engine, hence power will rise. Well, maybe initially, but you can only use steam more efficiently when you’ve got some, and if the draughting disappears, pretty soon you won’t have.
A person whose knowledge of UK history is enormous, who I trust, says that the rationale for jumper caps, certainly on the LNER, was economy. If I’m right, this is not about more efficient engine working, but by putting an enforced draughting limit on crews, beyond which they couldn’t burn coal any faster. Devious, huh? Supporting this, earlier GW types do seem to hit a brick wall in terms of sustained power. The original Castles, for example, this is about 1450IHP, though you can get a bit more short term. The claims, especially by Nock for power outputs of the original Kings climbing Saunderton bank border on the absurd. The highest powers I can find are three runs delivering 1850-1900ihp for 5 minutes. Sustained powers for the original Kings seem to hit a brick wall about 1600IHP; the three best runs I can find give 1680-1720IHP; trouble analysing old logs is you don’t know which way the wind was blowing which can have a significant impact on the estimate.
The best sustained powers I can find for jumper cap A4s are about 2150IHP. This will be a not dissimilar steam rate to the best sustained powers from original Kings, which were hopelessly inefficient on account of their Churchwardian superheat. The Kylchap raises the sustained A4 limit to about 2600IHP, in common with other UK Pacifics. The limit here is the fireman. With the kind of firing the French Compounds were given on test, this could be pushed up quite a bit more, since the draughting now allows it.
All I need to do now is work out at what steam rate the jumper cap actually rises. It may be then that it will be back to the drawing board!
The Churchward jumper ring was an early variable blastpipe that catered for a wider range of outputs than a simple blastpipe. An old book I have just got by Rutherford says the blastpipe area on the Kings was 23.8 in2 increased to 33.7 in2 with the ring open. The author says this allowed the Kings to produce transitory outputs far over their continuous rating, for example on Saunderton bank.
Another type to have this device was the single chimney A4. These reached up to 2500 IHP and my feeling is this would have not been possible without the increase in blastpipe area.
After the War, poorer maintenance led to these rings being deactivated/removed in both the A4s and Kings. Perhaps this was one reason for both types failing to reproduce their preWar performance.
To restore matters the Kings eventually received higher superheat and a double chimney with a relatively modest blastpipe area of 25.14 in2.
Folk on this forum pointed me to RP Johnson's (Baldwin) comprehensive book 'The steam Locomotive' (1942) (available on library loan in this country and at the NRM, a must read). This has a chapter on Front Ends, which begins 'Probably no part of the locomotive has been subject to more experiments and investigations than the 'front end', which is the right perespective, I think. He goes on to observe that the results were contradictory and difficult to explain. The chapter includes results on the Kiesel nozzle.
It does seem to me that Chapelon discovered something important in this area with his Kylchap set up, though his ideas as to the basis of his invention, and those theorising on the subject to this day seem to me wide of the mark.
The reason Chapelon's Compounds achieved such high outputs is that the Kylchap set up allowed specific evaporation rates in excess of 1000lbs/sqft/hr to be obtained i.e. very high front end limit; similar results can be obtained with simpler devices but at higher back pressure. That the hand fired coal rates these efforts required were quite impracticable, and hopelessly uneconomic was overlooked.
Giesel in his Anatomy of the Steam Engine does describe a couple of American blastpipes that are little known in Europe. There was the Sweeny blastpipe, of which he gives a photograph of the newer variant of 1899. It was star shaped with six elongated openings. It allowed a good mixing of steam and boiler gases already at a low height above the blastpipe. Its effective cross-section could be changed through a central plate (a bit like the Lemaitre). He says it was occasionally used. Only the Norfolk & Western used it consistently on their large Mallet compound engines for slow mountain service into the 1950s.
The Kiesel blastpipe was rather similar. A drawing is given. Here there were six narrow radially placed openings. He says from tests on the Wabash in July 1933 that it can be calculated that the Kiesel blastpipe gave the same backpressure as a conventional blastpipe of 19% smaller diameter. The main disadvantages were that the narrow openings soon became partially blocked with Olkohle (probably tar) and the exhaust beat down on the driver's sight at high speed (bit like Lemaitre again). The NYC tested the Kiesel blastpipe in 1937 with a chimney of 686 mm diameter, as used in Pennsylvanian pacifics but rejected it because of the visibility problem.
He describes the extensible stacks used by the ATSF as a curiosity. Operated by compressed air they could be raised 915 mm. On open stretches the chimney top was 5.8 m above the rails and improved both the drivers sight and the draught for the fire. One wonders if drivers always remembered to retract them at the right time. In the War, a few Austrian engines, which had a generous height clearance, lost their chimneys when they moved off their home territory.
These examples go some way to dispel the idea that American engineers were always conservative in their choice of blastpipe.
Many thanks- so much to catch up on!- the liestungsdiagramm of the 03-10 of particular interest.
Hoping some of this stuff will turn up in the UK libraries.
Dear Hudhon-Dreyfoss,
I find this question easier to answer for Austria than for Germany. The 'standard' works for Austria, covering the complete steam era are:
'Locomotivebau in Alt-Osterreich 1837-1918' by Karl Golsdorf
and
'Die Ara nach Golsdorf' by Adolph Giesel-Gieslingen
Publisher Verlag Slezak (don't know if still in print).
The first has many wonderful photos of ancient locos with outside Stephenson gear, which was standard before being replaced by Heusinger gear about 1900 (Heusinger the same or very similar to Walchaerts and invented slightly later).
I don't have any corresponding German books' only more recent stuff. Both the German turbine locos were war casualties. Much data that may have come from Grunewald is given in current publications. I have one about Class 03 10. There is given in this the Merkbuch for the class, with all the weights and measures. More interesting is the Leistungsdiagramm. This gives for a boiler evaporation rate of 57 kg/(m2 h) for the speed range 40 to 130 kph curves for drawbar PS, drawbar pull, specific steam consumption and specific coal consumption (both drawbar values) and cut-off used. PS and drawbar pull alone are given in the 0 to 40 kph range. For example, cut-off falls from about 48% at 40 kph to 20% at 130 kph. Maximum drawbar pull was about 1460 PS at 64 kph. Minimum steam consumption was about 8 kg/(IPSh) from 40 to 60 kph, and minimum coal consumption about 1.1 kg/(IPSh) from 40 to 70 kph, with a footnote to the latter that is not given.There was also a train weight, gradient table, gradients from 1 in 1000 to 1 in 40. For example, a 3.10 was rated for 540 tonnes at 60 kph on a 1 in 140 and for 300 tonnes at 130 kph on a 1 in 1000. So, plenty documentation exists.
Most notable, I thought, were the high mileages in the 1950s. For example, at Dortmund shed in 1953, montly mileages were often over 20000 km. In July 1953 No. 03 1043 ran 28460 km. And this was with the original boilers of 1940, made without certain alloy elements which were reserved for war use.
Thanks-the story of devices and systems aimed at improving economy or reliability not living up to their potential or being more trouble than they were worth seems to be a recurring theme.
I know very little about German steam developments other than part of a book on their attempts at turbines which I translated a few years back. Do you know if, how much and where good data is available? I found a reference to what was said to be the magnum opus on the subject, but it's not in libraries in this country. I am rather assuming that the Grunewald test plant and all its data succumbed to the RAF/USAF? Interestingly, Himmler lived in this suburb, as did Planck and Bonhoeffer in the same street about a mile away. Didn't even know there was asteam museum in Neuenmarkt- how remote can you get?
Dear Dreyfuss Hudson,
Thank you for the recording of Princess Elizabeth on the Lickey Incline, I enjoyed this the more so because I saw it earlier that day at Bristol Temple Meads. I agree the sound is indicative of a constipated exhaust. The outside steam pipes struck me as small in relation to those on Castles and Kings. And the mechanical layout is somewhat similar that of the Thompson Pacifics with a related tendancy to loose cylinders and leaking exhaust steam pipes. They appeared to be an obvious candidate for at least a double blastpipe and if they had belonged to any other Region would probably have received one (the LMS tests of 1934 already showed the advantage).
British Pacifics appear to have had an advantage in percentage of weight on drivers compared with both Continental pacifics and American 4-8-4s. But the US engines had the advantage of bar frames and roller bearings.
The Germans, in the post-war (Witte) era belatedly adopted combustion chambers, reduced the grate areas somewhat, and moved in some cases to roller bearings. Giesel says chimney diameters became a bit less extreme (in the Class 50 the chimney was 590 mm diameter under Wagner and 520 mm after Witte's retuning). The reason for the reduced grate was to reduce radiative losses (German expresses usually had more stops than British ones). The roller bearings, sometimes also on the rods, were not a complete success. Several locomotives, including one or both the Class 10 Pacifics, were condemned because of broken rod roller bearings. Axle roller bearings sometimes overheated (especially on 03 10s), with the need for immediate further lubrication and delay to the train concerned. The mixed type of feedwater heater also was not a success, nor the American type superheater regulator (although the Class 10s also had an auxiliary regulator in the normal position in the dome that avoided the kind of disaster that befell Blue Peter at Durham about twenty years ago). Performance seems to have been a bit below the British level, although the mileages achieved were higher especially in the mid to late 1950s. There was a relative shortage of pacifics.The lifetime mileages of the Pacifics in the German steam engine museum is upper 60000 to low 70000 miles per annum for simples and 55000 miles per annum for the 18.6, although these include ususally a zero to low mileage at the end of the War and for several years afterwards.
Many thanks- I agree that at speed the limitation on maximum cut off is of no consequence, and at 0-20mph is a good idea if you need to put down maximum NTE on starting, as was the case. The only situation where it might have helped a bit is climbing to e.g. to Cajon at 20-30 mph. At this speed, even in full gear the engine can't deliver more than 70000lbs TE, so the 'rule of 4' is not broken, and you could get an extra 500HP, albeit at a pretty astronomical steam consumption.
I read ATSF 2926 is being restored in Albuquerque. To see it running north to La Junta would be well worth the price of an air ticket, hopefully it will run before old Santa Fe route is closed, as many seem to think it will be.
Santa Fe test results showed that the 3765 class with limited cutoff could maintain full boiler pressure at all speeds and cutoffs. Test runs with 3766 showed 5450 MIH at 55 to 70 mph and over 5000 IH from 35 to 90 mph. Given the Santa Fe penchant for long runs at high reliability, this high HP over a very wide speed range makes for excellent performance with heavy, fast services where maximum output is required over wide speed ranges. The concept of limited cutoff was recommended for implementation on the 3460, 3776, and 5001 classes. In passing, the test of 3766, as equipped with roller bearings on all axles, gave a mechanical efficiency of about 90% up to 30 MPH and then dropped to about 82% at 60 MPH and 62% at 90 MPH.
The Nord Paris- Calais route is generally very well engineered. Normally gradients are 0.3-0.5%, (1/333- 1/200), steep enough to require high sustained power outputs at speed, which the 35 sqft grate Nord Pacifics certainly achieved. However, leaving Calais ‘cold’ there is about 7 miles of 0.8% (1/125) to Caffiers, before dropping back to sea level at Boulogne, then a few miles of 0.7% out of Boulogne. Serious adhesion tests in a maritime environment, I would think. With a simple, even slogging away at 40 mph up this grade, you would be back in something like 35% cut off, so the more even torque from the longer Compound cut offs might come into play.
If you look at UK practice, by my estimates it was unusual for locomotives to use much more than about 16000-18000lbs TE when starting , 50-60% NTE only, even with the heaviest passenger trains of around 650 (long) tons gross (loco +train) weight. In the US, gross weights were easily double this, but since you can only increase TE in proportion to driver load for a given number of axles (Max UK load about 22.5 long tons, US 30-32 tons), a much higher proportion of starting TE would probably be needed on starting- not analysed enough logs to know for certain. Hence the use of boosters. This may explain why US designers were so hot on the ‘rule of 4’ for the ratio of driver load to NTE. Hence the limitation of the maximum cut off on the ATSF 4-8-4s, which surely could have used a bit more TE on their biggest challenge, climbing Cajon Pass. Maybe the rule of 4 could be relaxed in a less demanding environment?
So, if high power was not used on starting, why did UK Pacifics, often slip? The text book says full gear and part regulator on starting, which should give the same torque as a Compound? Maybe other factors, e.g. slow release of brakes, uneven track, sharp turnouts, leaking lubricant contributed. At speed drivers would often work in 25-30% cut off, part regulator, again maybe to get a more even ride, but not what the text books say. In fact the loss of efficiency due to lengthen cut off between 20% and 30% is pretty small, so this was not really an issue, except with the Bulleid Pacifics, which, especially with their original 280 psi boilers produced so much power relative to the schedule requirements that they were often worked in 40% cut off, ca100 psi, which is very inefficient, hence the class’s coal eating reputation.
Lots of mays, mights and maybes in this, which is a shame, because the subject of adhesion is clearly crucial to operation, but something it’s difficult to get hard data on.
Back to the subject of blastpipes and superheat. I thought this clip might be of interest. The MR main line from Bristol to Birmingham has a stretch of 2 miles at 2.6%, thanks to the intransigence of a local landowner in Victorian times, an operational nightmare to this day, banking imperative in steam days. This spectacular recreation last weekend involved a Pacific with 3rd Division superheat, primitive blastpipe- the least efficient UK Pacific by far- listen to the racket/music it makes- banked by two 0-6-0s with 4th Division superheat (none), basically an 1890s design. (from about 1’40” on) Which just goes to show that efficiency isn’t what gets people’s emotions.
http://www.youtube.com/watch?v=7nCakwix6pQ&feature=youtu.be
Thank you for information about efficiencies of British pacifics and your thoughts about machine resistance.
I too was struck in rereading old Locomotive Profile No. 13 on Nord Pacifics by Brian Reed about the high loadings allowed compound engines. On Calais to Paris boat trains Nos 3.1252-90 hauled 600/675 tonne trains despite having only 56.8 tonne adhesive weights. Nominal factor of adhesion was a very low 3.33. I thought then that this was because this was a fairly flat route, but in the case of the S3/6 they were intended for the hilly countryside of Bavaria. The Norfolk & Western also was a believer in low factors of adhesion. Perhaps an advantage in adhesion was one reason for the great success of the Class Y6.
Giesel, without giving any examples of particular locomotives, also says that compounds could start and accelerate better than simples. The reason he gives is that, although they worked at a shorter cut-off overall, the cut-off in the individual cylinders was longer than for simples. This gave them a more uniform torque, which translated into less likehood of slipping. I don't know anything about modern traction, but it may be here too that there is more uniform torque than with steam and lower factors of adhesion.
Thanks for the efficiency data on the 3/6s- at about 12.8lbs steam/ihp-hr, that's up with the very best simples. ( Best measured in the UK was about 13.2; the Caprotti Pacific Duke of Gloucester was claimed to be better, but if you look at the raw data, this is another example of the data being stretched to prove a point; I think it was no better than 13.2. The Kylchap A4, having similar superheat but better exhaust pressure was possibly in the high 12s). From what little I know of the 3/6s they had a large superheater area for a 17.25' tubes, so superheat was probably very good.
The comment about low adhesion weight is very interesting. When analysing logs of French Compounds, I was very struck by a run of a P-O 4-6-2 on the Bordeaux line, when the normal 235 ton Sud Express was made up to 600 tons for the fastest stretch, 1 hour for 70 miles with a 75 mph speed limit. This was for the benefit of A I Lipetz visting from Alco. The key to keeping the schedule was getting this huge load started quickly, which they did, with no more than 54 tons of adhesion weight, far faster than any Pacific in the UK would have done I think; these slithered even with 67 tons adhesion weight. The smoothness of the Compound drive has often been commented on, not sufficient reason to persuade the accountants it would seem.
Dear Pete,
Thanks for the clarification- a treasure trove that I need to get my hands on. Particularly interested in the De Glehn, as it ties two continents' data together.
Dear Paul,
Thanks for these comments and ideas.
With respect to fluctuations in cylinder temperatures and the reason for the negative impact of condensation on LP efficiency, let me give you my understanding of the situation.
There’s no problem with saturated steam being formed in a cylinder during expansion, as you suggest. However, I’m talking about condensation not of steam in the cylinder, but of the steam onto cylinder walls, something which does impact efficiency. If the cylinder wall temperature is below the saturation temperature of the inlet steam, then a film of water forms very rapidly on the cylinder surface at admission. Since water is a poor conductor, it very quickly reaches a limiting thickness - up to 0.1mm. (Thickness depends on P1/2, and the cylinder temperature). Now during expansion, the steam in the cylinder falls below the temperature of the cylinder wall and its film, and the water film re-evaporates, and disappears up the exhaust without having done any useful work- a kind of leakage. The amount lost also depends on the surface are of the cylinder; on smallish UK cylinders, the amount can reach 1000lbs/cylinder/hr, not a large amount, but sufficient to take the edge of high efficiencies. I’m maybe making the LP losses through condensation more dramatic than they likely are, but they could eat away several % of any expansion ratio benefit compounding brings.
Question then is, what is, and what determines cylinder wall temperatures. These questions were addressed at Altoona in 1912; obviously the temperature lies between inlet and exhaust temperatures, and depends on the temperature of the inlet steam, and cut off- the greater the exposure to live steam, the higher the temperature. Knitting these data together with a theoretical model which calculates the size of the film under various operating conditions allows you estimate how much steam is wasted by this process. It is this kind of analysis that says that, in the HP cylinders of Compounds, where the exit temperature is always higher than the saturation temperature of the inlet steam, there will be no wall condensation. In the LP cylinders, where wall temperatures will be lower, there will be, despite the lower saturation temperature of the inlet steam.
This condensation process is the main source of heat transfer losses in cylinders. The Altoona data do not mention temperature fluctuation during the piston cycle, and Professor Hall, who developed the quantitative cylinder condensation model mentioned above built an apparatus to measure cylinder wall temperatures over the cycle. What he showed was that the block as a whole reached an equilibrium temperature, and if no condensation occurred, as with high superheat, there was no fluctuation in the temperature of the wall. When condensation did occur, there was a fluctuation in the very top of the surface, as heat flowed in from the condensation, then flowed out as the water film re-evaporated.
http://5at.co.uk/uploads/Bill%20Hall%20software%20and%20papers/The%20Effect%20of%20Superheat%20on%20Cylinder%20Condensation.pdf
So, clearly, the condensation process apart, there is little heat transfer to and from the cylinder walls during the piston cycle, once the equilibrium temperature of the cylinder block has been reached. Chapelon feared heat losses along these lines, but his fears were I think unfounded.
As to what would happen to cylinder wall temperatures if you went to very high superheat, obviously there’s no data, but like you I suspect any problems are surmountable. However, there is another effect worth considering which is that as you increase superheat, so the temperature of the exhaust goes up, and with it the back pressure at a given exhaust flow. With conventional exhaust systems, this means that once inlet steam gets up to 8000F, you are already getting into diminishing returns, and by 1200oF the benefits disappear and things get worse, according to my calculations . I suppose you would be able to optimise the exhaust to deal with this, haven’t thought about this though.
With respect to your comments on lessons from modern power plants, perhaps the biggest lesson is that if you want to use steam efficiently in a heat engine, turbines are the way to go? Come back Jawn Henry? Seems to me you would only go down this road if there were only coal left to burn, and the economics of electrification didn’t add up.
Thanks for this info from Giesl. HIs conclusion about typical steam rates with NYC Hudsons conflicts with my own analyses, which suggest that no more than 3500IHP was required to gain significant amount s of time of the fastest schedule, the 20th Century with about 1000 (US) tons, no more than 50000lbs/hr cylinder flow, say 45000lbs/hr blastpipe flow with 10% feedwater recycle. I seem to have found some data that the J1s blastpipe as having increased from 6.75" to 7 3/8" over time, but with 6 sqins lost due to 1/2" bars. At the lower area, you would indeed get blastpipe pressures around 1 atmosphere, at that kind of steam flow.
You will find me sceptical to the point of tedium on most claims from the steam age (maximum speeds, powers, efficiencies etc) especially where there is a personal or company interest involved. Many don't stand up to detailed scrutiny, or are data massaged in such a way as to present the interest group in the best possible light. (Nothing new here, then). So, I feel like a bit of a curmudgeon, but getting to the bottom of matters is what interests me.
I was interested in your comments about maximum German evaporation rates. I haven't looked at German practice much, but I did do some analyses of running at the end of steam from Bremen to Osnabruck, and ws surprised to find that the running of a heavy 600tonne train which gained 10 minutes on the schedule was done with little more than 2000IHP, this from an oil fired Pacific ( no fireman limit) with a 43 sqft grate, no more than about 600lbs/sqft/hr depending on boiler feed system Given this effort was significantly in excess of what was normally needed, I formed the impression that German boilers were not steamed hard.
I corresponded extensively with John Knowles during my researches into Locomotive resistance, and the basic approach we adopt is the same, seeing overall locomotive resistance as the sum of its resistance as a leading vehicle, given by the Davis equation, and the resistance of the machinery between the pistons and the wheel rims (MR). This, I later discovered is the Baldwin/RP Johnson approach, much more realsitic then the German Strahl equation. We were actually quite close in our estimates. Where we differ is in the value of MR; he uses an empirical approach which gives values somewhat higher than given by the first principles anaylsis done by a mutual friend mentioned in another post. I think these first principles figures are in much better agreement with UK test plant data for MR, and overall Locomotive resistance as measured in road tests- but we are talking relatively minor differences on a topic where there are very widely divergent published views, that do not really affect the outcomes mentioned here. Interestingly, the value of MR proposed by Johnson is I think way too high. It is based on Altoona testing, but these figures can be shown to be erroneous, a minor blot on an otherwise priceless set of data.
Paul Milenkovic The real limit on superheat temperature is the superheater tubes. Otto and Diesel cycles can have very high cycle temperatures because the applied heat is intermittent, and there is also water or air cooling on the outside of the cylinders and cylinder heads. Gas turbines cannot have quite as high turbine inlet temps, but again, there is a hot side and a not-hot side of the turbine blade, and various cooling schemes can be applied. With a steam superheater, the material has to withstand the full superheat temperature on both sides of the tube, which I see as a limit to steam cycle temps, even if you go to exotic "superalloys" for the tubes as you do for turbine blades.
The real limit on superheat temperature is the superheater tubes. Otto and Diesel cycles can have very high cycle temperatures because the applied heat is intermittent, and there is also water or air cooling on the outside of the cylinders and cylinder heads. Gas turbines cannot have quite as high turbine inlet temps, but again, there is a hot side and a not-hot side of the turbine blade, and various cooling schemes can be applied.
With a steam superheater, the material has to withstand the full superheat temperature on both sides of the tube, which I see as a limit to steam cycle temps, even if you go to exotic "superalloys" for the tubes as you do for turbine blades.
Paul,
My understanding is that the amount of superheat is limited by water disassociating into OH- and H+ at much higher rates than at room temperature. Both ions can be very corrosive and make life miserable for any metal in their presence.
- Erik
One of the better early compounds was the Bavarian 3/6 built between 1908 to 1931. Even in the 1950s they were preferred by many crews to the larger 01s and 03s on the hilly curved route from Munich to Lindau. Coal consumption for a round trip was about 2 tonnes less and oil consumption was similar despite the newer engines having only two cylinders. Their better balancing allowed them to start more easily than a two cylinder engine and acceleration up to 62 mph was better. Above this their machine resistance and corresponding wear rose rapidly. Maximum speed permitted was 74 mph (coupled wheel diameter was 1870 mm). Compared with the 3 cylinder 01 and 03 10s they were easier to lubricate (all had bar frames). The workshop personnel preferred the simpler 01s and 03s but overhaul costs were of the same order because damage to frames was unusual in an S3/6 on account of their smaller and better balanced piston forces.
Minimum coal consumption was 5.8 kg/ IPS, an excellent figure corresponding to about 795 C steam temperature in a simple expansion engine. With tube length limited to 17.25 ft evaporation rates up to at least 70 kg/(m2 h) were attained (2300 IPS), well over the official 1770 to 1830 IPS.
There was a small series with 2000 mm coupled wheels, which were used also outside Bavaria, including on the Rheingold in the 1920s and 1930s.
Despite their modest adhesive weight of 52.7 tonnes they handled trains of over 600 t. 18 478 is again operational.
They were a well balanced type with strong frames and a good boiler, as well as the good balance of their 4 cylinder layout.These other factors make it is hard to say how much their being compounds contributed to their success.
Thank you for your reply. I must correct an error in my post though - only 6 compounds.
The 8 tested were:
1. simple PRR 2-8-0 type H6a
2. simple LS&MSR Brooks 2-8-0 class B-1
3. cross compound MCR Alco 2-8-0 class W
4. tandem compound AT&SF Baldwin 2-10-2 class 900
5. DeGlehn PRR SA de CM essentially an exact duplicate of those supplied to the Northern Rlwy of France
6. Vauclain balanced compound AT&SF Baldwin 4-4-2 class 507
7.balanced compound w/ superheat Royal Prussian Rlwy Admin HMAG 4-4-2 class S8
8. balanced compound NYC&HRR Alco 4-4-2 class I1
Pete
I think we are pretty much in agreement that steam locomotive 2-expansion piston compounds can have losses between the HP exhaust to the receiver and the LP intake from the receiver. As I mentioned earlier, this shows up in compound indicator diagrams where there is the HP pressure-volume loop, the LP pressure-volume loop, and this big gap between the two loops representing what you call entropy loss (i.e. lowering in pressure without doing any mechanical work).
With respect to the thermal losses I had suggested, one of the knocks on steam-locomotive piston engines is that you have alternating hot inlet steam and relatively cooler exhaust steam flowing in and out of the the ends of the pistons, and such things as the Skinner Uniflow engines, used on the Lake Michigan ferry boats but never in railroad applications are supposed to reduce thermal losses from this effect. Separate into HP and LP cylinders, in the words of a gimmick sandwich package from a fast-food restaurant, are supposed to be better at "keeping the hot side hot, the cold side cold." What I am thinking of are conductive losses, of heating a chunk of metal in the cylinder with inlet steam and then cooling that same chunk of metal with exhaust steam.
With respect to conductive losses, why is condensation, especially in the LP cylinder so bad? I can understand you don't want too much fluid in the piston that you burst something with hydro-lock, but why the worry about saturated steam? The only thing I can think of is that if steam is condensing on the cylinder walls, you are transfering enormous amounts of heat as losses to those walls whereas if the steam stays as steam, you get much less heat transfer to the walls?
With respect to high superheat temperature, I don't think that the upper limit on superheat should be lubrication of the cylinder. You probably don't want high superheat during "hard working" -- 85% cutoff -- but with respect to high superheat under 15% cutoff where you are getting a lot of expansion, the high piston temperatures are only intermittent, much as a Diesel engine that they know to keep oiled in the face of very high intermittent temperatures. On the other hand, oil is typically injected into the steam, which has the effect of exposing the oil to full superheat temperature even though the steam cools down upon expansion in the cylinders. Perhaps one would have to inject lube into the cylinders more directly.
In light of this, the most efficient steam cycle, within the limits of boiler pressure brought on by scale formation in a cycle that is not using highly distilled water, and within the limits of not using a condenser as you don't have access to river water in a moving steam engine and you do in a stationary power plant, that steam cycle should use a high degree of reheat between HP and LP stages, yeah, yeah, the reheat tubes and headers also add to pressure drop.
I am thinking that not only do you want the exhaust to be above condensing, you want the exhaust to approach the boiler temperature. That way you could use exhaust steam to not only boost the feedwater temp up to atmospheric boiling of 212F, you could boost the feedwater temp up towards the boiler water temperature of around 400 F? One would have to work all the details of the thermodynamics, but I believe highly-optimized stationary powerplants use reheat between compound stages along with transfer of heat from LP exhaust to boost the feedwater to a high temperature (the feedwater heater needs to apply heat to the feedwater after injection to boiler pressure for this to work). And if automagically you could use this excess exhaust heat to pre-heat the combustion air, so much the better.
Before you think I am crazy, the British 5AT project (they have a Web site) has some kind of metal partition and use of the front part of the boiler at the smoke-box end as a kind of preheat-the-injected-water-to-boiler-temperature setup, but they are using the leftover heat from combustion in the flues just prior to the smokebox. I am proposing using exhaust steam heat for this purpose by applying reheat in a compound to boost exhaust steam temps, and I am thinking the power plants already do this.
But yeah, yeah, in the steam locomotive application, these schemes to wring out more efficiency involved more hardware (I am not giving up on compound expansion and I am adding a reheater in addition to the superheater along with a high-pressure feedwater heater), and in the end, it was the maintenance costs not the fuel costs that were the dominant economic effect.
If GM "killed the electric car", what am I doing standing next to an EV-1, a half a block from the WSOR tracks?
The $64K of question, of course! I drafted my thoughts on French Compounding into an article for a general audience, deliberately skating over some of the hard questions. I circulated to local experts who all came back and said that the fundamental questions needed tackling head on, so a complete rewrite needed. I now have four pages that attempt to answer your question, enough to send even the most dedicated reader to sleep. So I have had a go at trying to simplify it in response- still over a page- so see what you think. (I'll get to your specifics at the end).
Why Compounding?
There are two ideas to consider.
Firstly, improve expansion ratio, the volume of steam in the cylinder when exhaust ports open divided by volume of steam in cylinder at cut off, including clearance volume. The higher this is the better, providing the expansion is isentropic, as it is reasonable to assume in a simple engine. If you calculate expansion ratio for a compound, even flat out in 70% cut off in both cylinders, and allowing for 25% CV in the HP cylinders, this is in excess of 3, whereas for simples, the range varies from about 1.5 to 2.5 as cut off reduced from 50% to 20%. There is however a problem, not always recognised in steam age literature, namely that the expansion from HP inlet to LP exhaust in a compound is not isentropic, because of what happens between HP exhaust and LP cut off, so the above ratios are misleading, and the expansion ratio benefit is not actually known from these simple sums.
So the second, underlying idea is to maximise isentropic efficiency, in other words reduce irreversible losses associated with increases in entropy. The maximum amount of work possible from a heat engine-the isentropic (constant entropy) ideal is, for a given inlet temperature and pressure and exhaust pressure, fixed by the second law of thermodynamics. There are in fact four major sources of irreversible losses:
So how do Compounds fare on each of these criteria?
What’s important to recognise in this is that if you take a simple engine working in short cut off, and high superheat, i.e. no condensation, at realistic backpressures you can in principle get isentropic efficiencies of ca 85%- the cylinders are doing a pretty good job of getting the most that is possible from a given inlet temperature and pressure. The ‘in principle’ refers to the fact that, from UK test data, there is usually at least 5% leakage to take into account, which means that isentropic efficiencies in the low 80s were in practice the best achieved.
So, if a good simple can get 85% isentropic efficiency, there’s only 15% to go for, whether through compounding or valve gear design. The Chapelon 4-8-0s did, by my as yet provisional calculations achieve isentropic efficiencies in the low 90s, 10% better than actual simples. However this figure only applies if a) the superheat is high enough into eliminate LP condensation, and in most other cases it wasn’t and b) the right combination of LP and HP cut offs is used, which depends on both the design of the engine and the actions of the crew and c) there is no pressure drop between HP exhaust and LP inlet.
The design constraints on a steam locomotive are very considerable, which means, I think, that it is difficult to get the few % improvement in thermodynamic efficiency that is possible, which is why most abandoned the attempts to do so. Compounds were successful in a marine environment. The Titanic had triple expansion Compound engines rated at 46000HP. However, there are no hills or stations between Southampton Water and the Hudson, nor could the operating department add extra coaches for peak periods, nor was adhesion an issue, nor were there constraints on cylinder size- a much simpler environment.
With respect to you specifics;
a) It's clearly an objective in any engine to reduce wiredrawing, hence attempts to improve valve size etc. Whether, because Compounds were working at high HP cut offs against a higher backpressure, hence with higher pressure at cut off there was a net benefit is hard to say, but bear in mind if there's only 15% to go at, and a lot of this is due to incomplete expansion, it's maybe not as big a deal as people thought. Also bear in mind that you will have a second set of wiredrawing losses in the LP cylinders.
b) Not sure what you mean by 'reduced thermal losses'- do you mean e.g. convective and radiative losses? These seem to be small
c) Cylinder friction losses are not that high. I have worked with a friend to produce a good quantitative model of Locomotive resistance. A key element of this is to estimate 'machine friction' losses from first principles, which includes of course losses associated with cylinder friction. The friend's estimate comes out as rising from 3HP at 20mph to 14 HP at 90mph for a large UK 4 cylinder Pacific. Even if this is out by a factor of two, it's not such a big deal.
d) On the earlier French Compounds, the LP inlet pressure was below the HP exhaust pressure, so there was an unhelpful pressure drop of the kind you indicate (see above). Chapelon's development work elimiated this- so it is possible with the right valving.
Thanks- most interesting- more evidence that early 20th Centrury Compounds weren't fulfilling their potential. I did read Warden's book on the NW Ys in the unfulfilled hope of getting some test and technical performance data on them. He clearly was a big fan of the Ys, but never talks about efficiency benefits, and himself wonders if it would not have been better for the NW to have built more versatile As instead of the last Ys.
What is the compound supposed to do?
Is one thing to get a greater expansion ratio of the steam within the limits of a particular type of valve gear (i.e. Walschaerts)? That is, if you had a simple with the right valve gear without the "wired drawing" (throttling) limitation at short cutoffs, it could achieve higher expansion ratios?
Is another thing thermal, that by expanding in separate stages you keep the HP stage hot, the LP stage not quite as hot, and reduced thermal losses?
Is a third, possible one of cylinder friction, that cylinder optimized for a certain inlet pressure you have a certain amount of friction that there is a diminishing return for short cutoff and high expansion ratio? That a LP cylinder can have lower friction than a HP cylinder of the same volume on account of the lower steam pressure to seal against.
Working against any efficiency gains of compounds appears to be indicator diagrams I have seen. It seems there is considerable white space on the indicator diagram separating the HP indicator loop from the LP indicator loop, suggesting there are losses in the HP exhaust, LP intake and the receiver in between to contend with. Does this make sense, that compounds incur losses in transfering steam from the HP to the LP cylinders?
Thank you for your interesting reply. I agree that some historical data seem inconsistent or wrong, even from Swindon and Rugby. For an interesting but very demanding review of locomotive resistance, people might be interested in a piece 'Steam Locomotive Resistance' by John Knowles that has been posted on the web.
Regarding the West Country, a steam rate of 25000 lb/h seems typical of everyday use. Looking at curves of blastpipe pressure against steam flow rate for various degrees of superheat given by Giesel, for a pressure of 0.13 atu, flow rate is around 40 kg/(cm2 h). For 25000 lb/h this needs an area of 283 cm2 or 44 in2 (7.5 in diameter single blastpipe). This seems too high given that the Kylchap in the A4 was 39.23 in2. The only example I have found so far for a Giesel is for a Class 78. This was a medium power fast tank engine. When new in the 1930s, Giesel says blastpipe area was 128 cm2 but by the early 1950s some had only 95 cm2. Perhaps they had steaming problems similar to some British types postwar. From 1956 they received Giesel ejectors and higher superheat (through throttling gas flow through the large tubes while easing it in the small tubes). Giesel gives the ejector area as 232 cm2. This corresponds to a single blastpipe of 6.75 in diameter. Nominal steaming rate was 12200 kg/h or 26900 lb/h. The 7.5 in blastpipe would only look consistent if Giesel thought the nominal steaming rate of a West Country was 32800 lb/h with an all-out maximum of at least 36000 lb/h.
Regarding his time in the US he says he rode on a NYC 4-6-4 during tests in the early 1930s. The engine was one of the 1927/30 batch. With coal was of 7000 calories he says and normal evaporation rate was 60000 lb/h, maximum 70000 lb/h. This corresponded to an evaporation rate per sq m of heating surface of 77 kg/(m2 h), 90kg/(m2 h) maximum. This he says was far over the prewar German limit of 57 kg/(m2 h) and even their 1950s 'high performance boilers', which were 75 kg/(m2 h). Blastpipe pressures were a high 1.2 resp. 1.6 atu. He says the exhaust sounded like gunfire. He comments that the gas free area was relatively good. Further, in connection with the high specific and especially absolute boiler tube resistance, together with the self-cleaning smokebox, and the high performance, the overall design was good. Certainly better than the German.
Dreyfusshudson ...All of which says that when experiments were being done on Compounding in the early 1900s, it is quite possible that there would be little or no efficiency benefit for Compounding vs a well optimised simple.
This is an interesting statement. The B&O EL Class were originally built around 1916-1920 as compound articulated locomotives. In the late 1920's into the 1930's the B&O did a rebuild of the ELs in their own shops, converting them into simple articulated locomotives. Among the many changes, the boilers were redesigned to increase the draft and increase the superheat temperature. The test results not only showed a large increase in horsepower at speed, but an overall increase in efficiency. The final result was they took a locomotive that was designed as a slow speed drag locomotive, and ended up with a locomotive that was quite good at pulling manifest freight at moderate speeds, with a decided HP advantage over their S Class 2-10-2 (which were highly regarded freight haulers in their own right). The ELs ended up in service on the B&O right up into the mid 1950's before being scrapped.
Many thanks- I was unaware of the references you quote, and they are of great interest. Having completed the ‘American Project’ and written it up into an article for publication, I turned my attention to the feats of Chapelon’s Compounds, which in the 1930s were shown to be able to deliver more than twice the power at the drawbar of contemporary simples of similar size in this country (UK). There are two possible kinds of explanation of this; firstly the Compounds were markedly more efficient, so produced more power per unit of coal fired; secondly, the enhanced draughting allowed evaporation rates/sqft grate that were much higher than achievable with the simple draughting devices used in the UK at that time. This has been an enormous endeavour, wading through hundreds of pages of French text, and simulating the behaviour of Compounds using the computer programme.
Apart from possibly the Chapelon 4-8-0 (there are apparently conflicting sets of data in the same report), there is no evidence that French Compounds had any significant boiler or engine efficiency advantage over simples with the same superheat and exhaust pressure. The French Compounds were way ahead of most UK designs in these two respects in the 1930s, so will have had economy benefits (Explanation 1), maybe 20%, meaning overall, explanation 2 is the more important one, and if you work out the evaporation rates/sqft required for the highest claimed outputs, they could only be attained with superior draughting. However, if you then look at the coal rates required to achieve these steam rates, they are generally way beyond the amount it is reasonable to expect any fireman to achieve (most Compounds were hand fired). So, whilst the drawbar power claims are factually correct, they are simply demonstrations of potential, not of what might be expected in service, which, at similar efficiency would be no more than a simple could achieve if fitted with the same draughting and raised superheat. Many of the cylinder powers claimed are several hundred HP too high- Chapelon had a very unrealistic view of Locomotive Resistance- hence cylinder efficiencies also.
The Computer modelling of Compounds suggests that, as has often been stated, Compounding can provide up to 10% benefit in efficiency. This depends on exactly what you choose as the simple to compare with the Compound. If you fed live steam to both the HP and LP cylinders of the Compound i.e. operated it as a simple, this would have similar engine efficiency to the Compound, but you would have a brute with enormous tractive effort relative to adhesion weight, which would have to be worked in extremely short cut offs at speed- in practice an non-starter. It is against practicable simples that the benefits begin to show.
However, a number of factors can eat away at the 10%. Some of these issues Chapelon rectified, meaning older Compounds would not achieve 10%, and 10% also requires that you use the right combination of HP and LP cut offs. Most importantly, my calculations suggest that unless the inlet steam temperature is above 750oF (as on the 4-8-0, but not the other Compounds), there will be significant condensation on the walls of the LP cylinders, reducing efficiency; on simples, cylinder wall condensation ceases above about 6300F.
All of which says that when experiments were being done on Compounding in the early 1900s, it is quite possible that there would be little or no efficiency benefit for Compounding vs a well optimised simple. The Great Western Railway bought four De Glehn Compounds in this period and came exactly to this conclusion. That the de Glehn system was actually tested on the PRR plant is very exciting news, for one should be able to analyse the results to find out exactly what was going on in machines of this era, which by and large sealed the fate of Compounding. If the De Glehn reports do not form part of the numbered sequence of Altoona reports, then they are probably not at the library in London. I have found a source in Google books, The Pennsylvania railroad system at the Louisiana purchase exposition: locomotive tests an ... but unfortunately this exceeds the 700 page limit, so will not reveal its contents. I may be able to source this in the UK.
Just in case the PRR test results for 8 compound engines tested in St Louis in 1904 come within your sphere of interest and are not in the 2 bound volumes, and if no-one has already mentioned this, here are details:
The test reports are not numbered, as with other test bulletin numbers but identified by title only, eg 'Tests of De Glehn Atlantic Type locomotive, PRR'.
The test results are in the 1905 publication 'The Pennsylvania Railroad System at the Louisiana Purchase Exposition'. The volume includes a detailed description of the Altoona test plant as erected in St Louis, together with chapters on calibration of instruments and methods of recording and conducting tests.
Cheap, print-to-order copies are available from ABEBOOKS. I was thrilled to pick up a 1905 copy in a PRR memorabilia fleamarket stall near Altoona before the advent of the web made such things so easy, albeit now at a price ($hundreds).
Regards,
Pete1950
This responds to your three recent posts.
Firstly, with respect to the inlet steam temperature on the Jubilee, I had it rising from 298 to 3260C as steam rate increases in line with the test data on 45722. Steam chest pressure was assumed to be 215psig. I should say that, after ten years of using this fluid dynamics engine programme, it is in my view in a different league to anything available in steam days, providing of course, you know what you’re doing! Giesl’s approach is now superfluous.
On the Giesl results, I’m somewhat sceptical of Giesl’s claim. In UK test plant data, there are plenty of examples of misleading backpressures, due to problems with where the backpressure was measured. The programme mentioned does compute blastpipe pressure, in line with standard discharge theory. If you look at most Test results from Rugby, the agreement is very good between experiment and theory. There are a number of case where, for a given free nozzle area, measured blastpipe pressures are higher, but this is on locomotives where an orifice plate was fitted to the blastpipe. (An American idea, I believe). This reduces the discharge coefficient from 0.99 to about 0.90, so the effective area (working back to a common 0.99 coefficient) is about 10% less- so not as good as you might believe from a backpressure point of view.
There are then locomotives which show lower back pressure than they ought. Amongst those tested at Swindon, there are instances where the reported pressure is about half what it ought to be (Duke, DC, King DC, MN Lemaitre, to a lesser extent King Single chimney). Indeed the MN results at Swindon are about half of what Rugby reported (and the Rugby results are slightly lower than you would expect). At Rugby, the Duchess DC is about half of what it should be, and half of what the LMS reported. I have worked through the Duchess case carefully, and the problem is due to the fact that the pressure was measured at the base of the blastpipe, and there is very slight narrowing to the orifice, hence pressure increase at the tip. I suspect, but haven’t proved that this is the problem at Swindon.
Rugby did test a Giesl ejector vs a Standard orifice plate DC on a BR9. The free nozzle area for the Giesl was given as 30.2 sqins, though the reported back pressures are lower than expected, more consistent with 33.2sqins free nozzle area. I am assuming, but have not proved that this is also a measurement problem. Looking at the data you quote, the backpressure for the Lemaitre is consistent with a steam rate of about 25000lbs/hr. For the Giesl to be as good as Giesl claims, the free nozzle area would have to be over 40 sqins, equivalent to a 7 ¼ “ single pipe. I do not know what 34064 had, but I’d bet it wasn’t this. So, non -comparable measurements I think. I don’t think 34064 was indicated, so Giesl’s power calculations are on the basis of what would happen if the backpressure drop was as large as claimed. The Giesl on the BR9 didn’t dramatically change perceived performance. The reason, I think is simple. The steam rates required on Newport duties for BR9s were so low as for the reduction in backpressure achieved to not be significant. Giesl couldn’t believe how little and how lightly the BR9s were used.
It’s so long since I wrote in the thread, I’m not sure which boiler programme you are referring to. I have one of my own, rough and ready but works, but it cannot answer questions of the kind you ask. I have one from Professor Hall (He was chairman of the British Heat Transfer Group, so knew what he was doing!), which he abandoned uncompleted. He was concerned about some aspects of the theory; it is also obvious that it is a heat transfer model that does not take into account all factors involved in boiler efficiency, most crucially coal loss. I had rather ignored it, but recently discovered that it makes a quite remarkable (to me at least) prediction, namely that if there were no radiant heat transfer to the superheater, superheat would be very poor indeed. Obviously there is a radiant section at the beginning of the firetubes, and it is here where the majority of superheating occurs. What is critical than is the surface area of the superheater in this radiant zone, which will be greatly influenced, as you suggest by how near the superheater gets to the firebox. Chapelon consciously got his tubes as close as he could; because of the risk of burning the elements, most others preferred to keep them some distance away to reduce maintenance, but at the cost of superheat.
On this logic, if you make the firebox bigger, you could well reduce the radiant area in the firetubes which would not be good for superheat. Looking at US designs with large combustion chambers, even with A type superheaters, the level of superheat achieved was good but not top of class, although the PRR T1 with an E type superheater was excellent. It was short between the tubeplates, and I wonder if this meant the ends of the superheater were allowed nearer the firebox. The Chapelon 4-8-0 also had a very short boiler, and I wonder if, when tubes were short, designers, who were trying to maximise superheater areas, didn’t finish up with elements nearer the firebox.
The superheat measured on the V2 at Swindon was very good, also on the BR7 boiler of similar length, even better on the even shorter BR9.
So, I think the question you put is really interesting, but Bill’s boiler model is not good enough, nor is it validated against data, to allow conclusions at the level of detail your query requires.
On your corrected comment, I’m not sure why less water in the boiler would increase evaporation? Combustion is determined by amount of coal and draught, heat transfer is independent of boiler level surely? The amount of steam/hr/ gallon water would go up with a smaller boiler, but surely not the gross amount?
Sorry there is an error in my last post.
The second sentence of the second paragraph should read: So, compared with A4 there is less water in the boiler so for a given amount of heat produced in the firebox one might expect a higher evaporation rate.
Coming to Kylchaps and A4s I have a question that your computer program might help to answer. Coster says in his book on the A4s that the development of the LNER boiler might, with the benefit of hindsight, have been developed somewhat differently. That the 41.25 sq ft was quite adequate given the Kylchap but that firebox volume might have been increased.
So have a combustion chamber 1 ft longer than in an A4 and tube length 1 ft shorter (about 17 ft) as in the V2, A1 and A2. So, compared with A4 there is less water in the boiler so for a given amount of heat produced in the firebox one might expect a higher combustion rate. Less water in boiler reduces the overall weight so boiler plates can be a little thicker and boiler pressure increased a little as Gresley planned. The possible problem is that with the rear tubeplate 1 ft further from the firegrate so also are the rear ends of the superheater tubes. And lower superheat would be very negative from the efficiency point of view, perhaps negating the point of this thought exercise. Does your program make any prediction for the superheat in such a boiler, i.e. as an A4 but with about a 2 ft long combustion chamber and a barrel as in an A1?
Thanks for info, which I was unaware of. The blastpipe nozzle of 4.5 in would give a steam rate of below the critical value of about 93 kg/(cm2 h) for 20000 lb/h, but not by much.The minimum ssc you give for this single blastpipe (15.3 lb/ihp h) seems to indicate a steam temperature of at least 330 C (626 F) for 225 lb/in2 (following a steam table in Giesel's book Locomotive Athletes for a cylinder volume of 200 L), which would need to be corrected upwards to allow for the relatively small cylinders.
Giesel also gives the results of testing the West Country class pacifics with both the Lemaitre and Giesel exhausts. In going from the Lemaitre to the Giesel, blastpipe energy fell from 165 to 78 HP lowering pressure from 0.30 to 0.13 atu, increasing pumping work from 13.1 to 15 HP, which increased smokebox vacuum from 118 to 126 mm WS, lowering shock loss from 126 to 15 HP, and increasing other losses from 25.9 to 48 HP, which was enough to lift the exhaust clear of the engine. Pumping efficiency rose from 7.8 to 19.3%.
Thanks. I think the tests you describe are very important in that they had a crucial impact of UK locomotive design ever after. I have still to locate the full report, but what ex LMS senior engineer AJ Powell reported was that (surprisingly) there was no improvement in coal or water consumption for the Kylchap set up vs an extremely restrictive 4 ½” single blastpipe. I am at a loss to explain this. He notes that the Kylchap device used was probably far too powerful for the quite small Jubilee boiler, and that there was a continuous shower of sparks from the chimney, which may explain the lack of reduction in coal.
I have done some calculations ( see below) on what the benefit of the Kylchap would be, assuming it was of the dimensions used by the LNER, which were similar to the French Compounds. There is, even at the relatively low steam rate of 20000lbs/hr, an improvement in IHP of about 8%, so you would expect an 8% reduction in water at the same power. ‘At the same power’ may be the root of the problem. Because the cylinder backpressure is lower, the steam consumption at a given cut off goes up by about 10%, so power goes up by 10%. So it may be that on these road trials, the Kylchap locomotive was producing more power, and this interfered with the economy assessment. Hopefully the report will throw light on this.
Kylchap
Standard
Steam rate, lbs/hr
Cut off, %
Blast
pipe pressure psig
IHP
SSC, lbs/ihp-hr
vacuum, ins
13700
13.6
1.0
850
16.1
3.8
14.8
3.7
820
16.7
2.7
19100
19.6
2.0
1300
14.7
5.9
21.5
7.6
1210
15.8
6.0
23100
23.7
3.0
1590
14.6
7.2
26.4
12
1420
15.3
-
26400
26.8
4.0
1780
14.9
8.0
30
16.2
1530
17.2
The LMS later tried a Kylchap on a Duchess Class Pacific, and found no benefit. Here, I think the result is explicable. The Duchesses were fitted with double plain blastpipes with a similar free nozzle area to the Kylchap; no backpressure benefit expected then. A Duchess was tested on the Rugby test plant where it was concluded that the draughting was adequate, though not brilliant. Now what better draughting such as a Kylchap does is increase the specific evaporation rate (lbs steam/sqft/hr) possible. The Duchesses had grates of 50 sqft and in service were never required to steam at more than 30000lbs/hr (600lbs/sqft/hr) because they were hand fired. On test, they got 800lbs/sqft/hr; a Kylchap can deliver 1000lbs/sqft/hr- both these higher rates are quite beyond any fireman. So because only modest specific evaporation rates were possible in service, these could be satisfactorily delivered with a very ordinary plain double blastpipe.
So, in two sets of tests, the LMS concluded that Kylchaps weren’t worth it. The LMS mafia were in charge of BR locomotive design after nationalisation and guess what- no Kylchaps. BR standard designs were hobbled with very primitive exhausts. The mafia even tried to stop the ex LNER region from fitting Kylchaps to their Pacifics, where there was a clear need.
Just shows what happens when a) you have dodgy data and b) no sound quantitative theoretical model. Politics wins.
Thanks, I'll try again. Giesel in his Anatomy of the Steam Engine (in German) mentions tests the LMS did in 1934 comparing different blastpipes on Jubilee 4-6-0s, which initially were not great steamers. In all cases steaming rate was 20000 lbs/hr. For each arrangement he gives the total power in the exhaust at the blastpipe, the pumping power (that used for draughting the boiler), the shock losses (caused when the steam impacts the flue gases in the smokebox), other losses (not totally wasted as it lifts steam out of driver's line of sight), the smokebox vacuum, pumping efficiency (ratio of pumping power to total power) and blastpipe pressure. For the standard single blastpipe the results were: 238 HP, 9.5 HP, 154 HP, 74.5 HP, 91 mm WS, 4.0% and 0.45 atu. For an improved single blastpipe: 196 HP, 11.6 HP, 126 HP, 58.4 HP, 103 mm WS, 5.9% and 0.36 atu. And for a double blastpipe: 143 HP, 13.1 HP, 94 HP, 35.9 HP, 118 mm WS, 9.1% and 0.26 atu.
One sees a progressive fall in exhaust power, improvement in draughting and also implied a fall in cylinder backpressure, giving a further gain in useful power.
Neil, if you are using Internet Explorer 9, and didn't just mistakenly forget to type a message above, you will have to look for the small grey 'compatability view' icon that looks like a torn square, meaning a sheet of paper. It is just to the right of the top URL bar, along with a magnification glass icon, the curved 'refresh view' arrow, and the close-out X. Just click on the torn sheet of paper, at which it will conform to the view intended for this server and then it will turn blue.
Welcome to the forum.
Crandell
It would be interesting ot compare these various designs differences based on actual performance. A lot of data would be needed beyond most reports of drawbar horsepower, coal & consumption, and design parameters. Things like flame, firebox, flue & tubes temperatures measured at various points; percentage of imcomplete combustion & weight of air required for combustion; stack gas temperature, weight of flow & composition analysis; steam chest pressures and superheat temperatures; continuous pressure readings throughout the cylinder cycles; etc.
Too bad a lot this testing would require a time machine...or generous funds and cooperation by someone like the UP steam program or similar group that operates a late design steam locomotive.
Given that only about 12% of the energy of the steam is converted to mechanical energy at the wheels, the focus should be to try to design a system where the total heat of the fuel fed into the firebox, the total heat produced by combustion, and the total heat absorbed by the boiler producing steam is as close to unity as possible.
"If a nation expects to be ignorant and free, it expects what never was and never will be." Thomas Jefferson
Burgard540 If you look at the chart, notice that the proportions of heating surface are nearly identical regardless of the type of locomotive. Of the total heating surface: 63% is tubes & flues, 7-8% is firebox, and 29-30% is superheater. Notice that the ALCO built UP locos had the highest proportion of firebox heating surface at 8.4% for the articulated's and 7.8% for the 4-8-4's.
If you look at the chart, notice that the proportions of heating surface are nearly identical regardless of the type of locomotive. Of the total heating surface: 63% is tubes & flues, 7-8% is firebox, and 29-30% is superheater. Notice that the ALCO built UP locos had the highest proportion of firebox heating surface at 8.4% for the articulated's and 7.8% for the 4-8-4's.
Just to add some additional data, the 2-8-8-4 EM-1, built by Baldwin in 1944-45 had its boiler comprised of 61.2% (4,540 sq.ft.) Indirect Heating Surface, 28.6% (2,118 sq.ft. Type E) Superheater Surface, and 10.2% (758 sq.ft.) Direct Heating Surface. Baldwin clearly changed the boiler design on the EM-1 from the earlier AC10 and M3/M4 designs by enlarging the size of the firebox area vs. the fire tube area.
Thanks for these additional inputs- I'm heading off for two weeks vacation now, and will add my thoughts when I'm back
GP40-2 Comparing a locomotive's total heating surface doesn't tell the full story about the steam production capacity of a boiler and must be used with caution when comparing steam locomotives.
Comparing a locomotive's total heating surface doesn't tell the full story about the steam production capacity of a boiler and must be used with caution when comparing steam locomotives.
Thanks - that's pretty much my point in plain language.
Regarding tube length, the example Ralph Johnson uses of a USRA loco where increasing the flue length from 19 to 25 ft long increased the heating surface 29% (sounds good) but only increased the heat absorption by 2%.
Again, my point is that tubes and flues are more effective heat absorbers per unit if they are of moderate length and the velocity of the combustion gases is high. However, to achieve the better convective heat absorption requires decreasing the gas flow area, thus increasing the draft and back pressure. The trade-off of slightly less engine efficieny for greater boiler efficiency. Although to me, the focus should be increasing combustion efficiency.
Burgard540 I could have written my previous post more clearly. It was more born out of the questions regarding locomotive boiler capacities and performance given their design parameters such as direct heating surface (firebox & combustion chamber), heating surface of tubes & flues, gas flow area, etc. The ALCO built steam locomotives for UP (FEF's, Challengers, and Big Boys) had significantly less heating surface compared to locomotives of similar size. However, their performance was comparable. Going just by total amount of HS doesn't explain the boiler's capacity, assuming similar heat energy inputs.
I could have written my previous post more clearly. It was more born out of the questions regarding locomotive boiler capacities and performance given their design parameters such as direct heating surface (firebox & combustion chamber), heating surface of tubes & flues, gas flow area, etc. The ALCO built steam locomotives for UP (FEF's, Challengers, and Big Boys) had significantly less heating surface compared to locomotives of similar size. However, their performance was comparable. Going just by total amount of HS doesn't explain the boiler's capacity, assuming similar heat energy inputs.
But, were they really "undersized" compared to other steam locomotives? Or, was it just a case of a different ratio of direct heating surface to indirect heating surface compared to other locomotives?
As Dreyfusshudson correctly pointed out, a locomotive with many small diameter, long tubes did not get any additional heating benefit out of the last 4 or 5 feet of those tubes. However, what the additional length of the smaller tubes did was to give the locomotive the appearance of having a larger heating surface than a locomotive with shorter, larger diameter fire tubes.
Case in point. The N&W Class A, designed in 1935-36, used 24 feet long, 3.5" diameter tubes and had an evaporative heating surface of 6,639 sq.ft. However, its firebox area (which by far produces the majority of the steam) was 587 sq.ft.
Contrast that to the B&O EM-1, which was designed in 1944. The EM-1 used 4" diameter tubes just 20 feet long for a total evaporative heating surface of 5,300 sq.ft. What that doesn't tell you is that the EM-1 had a firebox area of 758 sq.ft. which was equal to the H8 Allegheny in size. The firebox and combustion chamber on an EM-1 was 27 feet long, 7 feet longer than its fire tubes. The UP Big Boy had similar construction with a 704 sq.ft. firebox area.
Theoretically, the greater diameter of tube require less draft & less back pressure, but as the diameter increases the less heat is transferred. The length of the tube should not be beyond 100-120 internal diameters. Most of the temperature difference in the combustion gases is needed to overcome the stationary gas film along the tube's walls. The best way to reduce the films effects is to increase the gas velocity. The greater velocity of flow also causes the gases to be exposed to more tube surface in a given time, thus increasing the heat transfer of the gas. Stationary gases in the tube can only give heat via radiation which as you mentioned only occurs for a brief initial length of tubes.
So, the ideal tubes and flues should be as small and long as possible with high velocity of gas flow. Of course the there is diminishing returns, where smaller diameters and longer flues lead to greater soot build-up and increasing back pressures.
The question then becomes for a given quantity of coal burned and combustion gas produced, does a the boiler perform better with relatively equal gas flow area and higher gas velocities through the tubes of less area vs. larger amounts of tube heating surface and slower gas velocity? And at what point does increasing velocity for a given tube length lead to too high exit temperatures, greater draft & back pressure, and thus loss of efficiency? Not to forget the front end design....
Clearly, an undersized locomotive boiler could be forced and produce the same quantity of steam as a larger boiler at the sacrifice of fuel efficiency. However, I don't think steam loco designers did this on purpose, nor would a railroad like UP (which had an excellent engineering department) accept this flaw (i.e. losing money on extra needed fuel for the same train hauling capacity).
No additional comments needed regarding the lack of heat transfer the last few feet of tube length, that's been well documented through various tests.
I agree that radiative heat transfer occurs in the tubes & flues briefly. That idea had occurred to me a while ago after studying the radiation of luminous and non-luminous gases and flames. The combustion gases entering the flues and tubes would have significant amounts of CO2 and H2O (good radiators) and be at high temperatures approaching 2000 degrees F.
Regarding superheater elements too close to the firebox, the Big Boys originally had the superheater elements end 15” from the firebox tube plate. However, that was increased to 24” in March, 1943, and later increased again to 36” in August, 1950. It would be interesting to see if there were any noted performance differences or tests done to measure the effect on the final superheat temperature (or possible less pressure drop) from the shortened elements.
Sadly a lot of these questions would require full scale testing, and each locomotive would be its own special case. Needless to say, some good parameters would be found.
Cheers
Addendum - Of course not to be neglected in the design is the type of coal or fuel used.
Dear Joe,
Thanks- not sure if I understand the beginning of what you write properly. Are you saying that if you increase boiler resistance by either narrowing the tubes or making the boiler longer to improve heat transfer, you have to improve draughting, which means that you have to narrow the blastpipe, which means back pressure goes up, so in effect you are trading boiler efficiency for engine efficiency?
If so, I would say the following. Firstly, back then, they did have to make this trade off, because the understanding of front end design we have now was not available. However, Chapelon said, only half in jest I think, that you should actually start locomotive design with the front end exhaust system; with the kind of systems he was developing, you could up the draught if need be without increasing backpressure, so the trade off was not needed.
More fundamentally I would say this. According to the computer model I have, heat transfer in the last few feet of the boiler tubes is negligible; every little bit of extra evaporation/efficiency is worth having if there is no cost associated with it of course, but longer tubes do cost both in materials, maintenance, and quite probably reduced superheat- a far greater loss to engine efficiency than the very small gain in boiler efficiency the longer tubes give. So, the proper solution if faced with this would be to cut the tubes down- 18’ is more than enough I would guess. (Though it wouldn’t do much for aesthetics if the Big Boy/Challenger boilers were 5’ shorter!)
So, for this reason, I don’t think the variation in heating surface areas you quote will be of much consequence to boiler efficiency, though they may indirectly influence superheat.
I have just rediscovered some stuff that adds weight to the idea that optimising boiler tube heat transfer is not that big a target. One of our testing stations looked at their data and tried to calculate what firebox temperature ought to have been from first principles- Stefan Bolztmann and the like. The approach really only needs you to know radiant surface area (which they assumed was just the firebox), and the tube gas flow, and the amount of coal fully burned (as opposed to fired, given some is lost unburned). You get these latter two via Orsat smokebox gas analyses.
Here are some figures they quote for where the burned coal’s heat goes; firebox radiation 38.3%, firebox convection 5.5%, superheating steam 9.5%, evaporation via convection in tubes 28.1%, 18.7% lost in flue gases. Trouble is that these calculations indicate that firebox temperatures were about 300F higher than they actually measured. Now they knew enough to know that what the pyrometer said was not necessarily accurate; even so, 300F is a big difference. Since it is the T4 Stefan Bolztmann radiant contribution that has the biggest effect on the estimate of T, I turned the calculations on their head and asked ‘What is the radiant surface area if the measured temperatures are correct?’ The answer is roughly twice the actual firebox area, which implies there is a radiant section in the firetubes just less than 10% of their length- about 1.5’ on UK designs close to the firebox. Now with this kind of approach there is a huge uncertainty on what this radiant length would actually be, but I think the data are good enough to confirm that the tubes do have a radiant section, hence the radiant heat transfer is significantly greater than the 38.3% above, hence the convective tube contribution significantly less than 28.1%, and since gas temperature decreases exponentially along the tubes, no surprise that losing this small piece of convection is a minor contributor. This also supports the idea I floated earlier based on computer models, that, were it not for a radiant section of the superheater close to the firebox, superheaters would be pretty ineffective, and there is a trade off between having elements too close to the firebox and getting burned, and amount of superheat you get.
If only there was, or we could do, proper experimentation to sort all these things out!
With regard to blastpipe pressures, then at equal steam rate, assuming (not necessarily correctly) that all blastpipe discharge coefficients were the same and close to 1, and that the earlier analysis of Big Boy blastpipe diameters over time is correct, then it would be 1942 Big Boy= H-8<A< 1946 Big Boy, with the Blast going sonic (> 12.3 psig) at about 110000 lbs/hr cylinder rate on the first two, around 90000lbs/hr or less on the latter.
Some other ideas:
The gas area through the tubes and flues should be as high as possible to avoid the draft loss (especially at higher firing rates). Samller area (and longer tubes) requires greater back pressure in the cylinders, due to higher velocities in the tubes. The higher velocity has the advantage that since for convection the heat transfer is largely dependpent on the amount of gas flow, doubling the gas flow per tube increases the evaporation 90%. Total gas flow area is limited by the diameter of the boiler due to loading gauge.
So intuitively it seems that there's a trade off for greater heat transfer & evaporation due to higher gas velocities in the tubes, at the cost of higher back pressure in the cylinders. This could explain the vast disparity between the heating surface for the locomotives below.
I calculated the gas flow are for the high powered articulated loco's: Big Boys, Allegheny's, and N&W "A"s. My calculated answer for Big Boy (first series #4000-4019) was within less than 1% than the given figure of 1631 sq in or 11.33 sq ft. I assumed that the 2.25" tubes were all 11 BWG (0.120") wall thickness.
Big Boys (4000-4019) - 11.43 sq ft (Flues & Tubes 22 ft long); Total HS 8,355 sq ft
Allegheny (1600-1644) - 12.49 sq ft (Flues & Tubes 23 ft long); Total HS 10,426 sq ft
Allegheny (1645-1659) - 13.00 sq ft (Flues & Tubes 23 ft long); Total HS 9,717 sq ft
Class "A" - 11.04 sq ft (Flues & Tubes 24'1" long); Total HS 9,353 sq ft
Curious what the back pressures were for these locomotives under full power, especially the "A" since it had only 3.5" flues at over 24 ft long. Big Boys had a back pressure of 18 psi.
Going back to Big Boys lbs of water evaporated per of coal, the figures given for the April, 1943 tests were actual numbers not equivalent evaporation. The 4.44 lbs water evaporated would be increase to 6.40 lbs equivalent evaporation per lb of coal fired. Equivalent evaporation factor about 1.44. I thought of this after seeing a few old reference books rating the boilers' evaporation per pound of fuel this way.
Other thoughts?
Joe
Have now got more data from Joe and other sources on UP front ends. The question of front end dimensions is not trivial, since the UP was constantly experimenting with and developing their locomotives. What is clear is that by 1951 the nozzles on the Big Boy and FEF were 3", on the Challenger 3 1/8". There is a 1942 drawing showing clearly 3.5" nozzles on the Big Boy, but this is marked 'not used' and was replaced by another one less than a month afterwards, when other features of the front end had been changed. A reference to nozzle spec is given, but this drawing is not to hand. In these and other diagrams up to 1946, the profile of the nozzles is similar, about 4 3/4" external diameter, with a taper to about 4 1/8" at the top, which would give a lip of about 1/4" if the nozzle internal diameter was indeed 3.5", as per my measurement of Kratville's photo.
Another UK correspondent says that the nozzles were down to 3" by 1947 as per the Locomotive cyclopedia, reduced from the original. Joe suggests this change might have been connected with the removal of over fire air at that time.
This all confirms the idea that a fractional reduction in engine efficiency was quite acceptable against other prioirties.
The specification in nozzle diameters to an 1/8" is interesting- variations of this magnitude shown for other types between coal and oil firing, winter and summer. We did the same in this country, but a random sample of locomotives brought in for testing in 'typical' condition revealed most of them had significant deposits in the blastpipe- some up to 1/2".(!). So much for drawings
Thanks- I was sent a copy of fig 2.112 on p171 of an unknown book- page titled Steam Locomotives- 4-8-8-4 Articulated types. I blew the image up, and measured the 33" shown across one of the chimney cowls as 63.5 mm. On this scale, the outside diameter of the nozzle is ca 9mm - about 4 5/8" as you say, but there is a taper at the top, leaving only 8mm", 4 1/6" though I quoted 4 1/8" ( margin of error).
I then looked at the pictures on p29 of Kratville; on the lower one the taper at the top is quite visible. I measured the outside width across the top of the taper as 50mm, the internal diameter about 41 mm, giving me about 3 3/8".
If the diameter is indeed only 3" as you say, that's equivalent to about an 8.5" blastpipe ( assuming 0.99 discharge coefficient)- not brilliant for something intended to steam at 100000+lbs/hr.
Very keen to get the nozzle diameters for the FEF also- can't get the relavant books that might have it on interlibrary loan here.
Dave,
Which drawing did he send you? UPdrawing 443 CA 33736 (Jan 16, 1951) is where I got the dimensions for the nozzle tips (3" ID & 4-5/8" OD press fit into nozzle plate). It also lists the nozzle sizes for most other UP steam locos. I'll send it and other pertinent front end dwg's via emal.
A correspondent has kindly provided me with a detailed drawing of a Big Boy front end, which shows the external diameter of the 8 blastpipe nozzles is about 4 1/8". Usung Kratville's picture of same, this puts the internal diameter of the nozzle at just less than 3 1/2". This gives an area equivalent to about a 9.5" single blastpipe- similar to the C&0 H-8. This means that even at a cylinder flow of of 100000lbs/hr ( say 5800 IHP at 20mph, 7000IHP at 40 mph) the back pressure will be less than 10psi- nothing too desperate, and under many operating conditions quite low. I imagine the FEF might be a bit narrower, but if like other 4-8-4s they were normally worked at no more 65000lbs/hr, the blastpipe pressures would also be low.
Thanks for this. The kinetics of combustion of CO to CO2 seem to be pretty fast at firebox temperatures, at the kind of excess airs present-that's why you can use GPCS, or fire a thicker bed with less primary air and reduce unburned losses from the firebed. As far as I know, firebox design isn't a constraint here. Some goes for combustion of volatile compnenets of coal I think, more secondary air allows combustion to complete, less black smoke. My suspicion remains that firebox design won't have too much effect on the combustion of dust from the stoker- the coal burning process much too low.
Just had information that the French used graded coal-1- 2" pieces to reduce coal losses with mechanical stoking, so this could be another variable in the 21st Century Sherman Hill trials. For completeness, a DMIR Yellowstone and SP cab forward need to be added, I think.
My understanding is that the path of the Big Boys et al in such locations is still marked by a layer of coal char on either side of the tracks, something for the 31st century archaeologist to ponder.
The firebox of the Big Boy locomotives, by loading gauge limitations, had to be long & shallow since it was above the rear 2 driving wheels. Kratville makes the statement that the firebox should have been about 2 feet deeper for proper proportion. For me the question is for the same firebox volume, what are the effects of a shallower firebox with long path length of combustion gases vs. deeper firebox with shorter path length of combustion gases? Intuitively, you would want a longer length for the combustion gases so that they have a greater time to complete the combustion process. As designed & built, the Big Boys & Challengers had over fire air openings on the sides of the firebox. These secondary air openings were later removed about 1948-1949, why I don't know considering the now well known advantages of introducing large amounts of air over the firebed. (This should probably be in a new thread about firebox design...)
Regarding the coal used on the Union Pacific, I've been trying to track down good information on this subject, ie the percentage break down of its elemental components, ash contents, typical moisture content, etc. The coal UP used for its steam locomotives came from Rock Springs or Hanna. It was not the Powder River Basin coal mostly mined now out of Wyoming (Btu value of ~8500 per lb) From various sources (like Babcock & Wilcox), Rock Springs coal has a heat value of about 11000-12500 Btu/lb. It's friability is a good question, since the small pieces would be sucked right out the stack in the stream. Also, what was the typical size of coal particles after passing through the stoker?
Regarding getting the Alleghany's and Big Boys fired up to compare...there's nothing 20 million dollars can't solve!
Dear Erik,
Many thanks-I'm in a parallel conversation with local experts who are making exactly the same points about the Big Boy firebox.
However, I remain sceptical about these arguments; the reason is that the lifetime of a particle of coal flying through the firebox is less than a second, whereas the burn time for particles of more than a millimetre is measured in minutes (Wardale provides the data). Backing this up, I have test results on the PRR Q2 with and without induction tubes, from which they conclude they have no meaningful impact on boiler efficiency.
Now there is UK data that clearly shows the benefits for overfire air with hand firing of lump coal- significantly reduced coal loss. When firing lump coal, there has to be a process which produces particles that are small enough to be swept away by the air flow. This is I think the production of fine coal char of a few mm in size from the coal surface as it burns rapidly- the stuff that lands in your eye when you look out of the coach window. With more overfire air, and less primary firebed air, there will be a cooler firebed which leads to less char production at a given rate of carbon transfer to the gas phase - the carbon is only partially oxidised to CO in the firebed. This is subsequently oxidised to CO2 by the overfire air. This is the basis of the GPCS system also.
However, with mechanical stoking, the majority of the unburned loss is from the stoker feed. I would bet these fines never reach the firebed, but are just swept away, perhaps fragmenting further in the heat of the firebox. A small increase in oxygen partial pressure from overfire is not going to influence this. Hence over fire air can influence loss from the bed, but not I suspect from the much larger stoker feed. Higher firebox oxygen can influence the rate of oxidation of volatiles, so reduce black smoke. This is consistent with Wardale.
So, my hunch is that the Big Boy's low boiler efficiency is indeed a consequence of the Wyoming coal used, and that its problem is that it either crushes more easily in the stoker screw or is more prone to fragementing in the heat of the firebox.
If you were to fire a Big Boy with Appalachian coal, (much more like UK coal) my bet would be it would behave pretty much like an H8. This very intersting experment would be a rather expensive to organise in 2011, but I'm sure quite a few people would turn up to Cheyenne to watch.
Somewhere in the memory banks is I think a quote by a UK CME, Sir Jasper Spock visiting Wyoming in the late 19th Century 'It's coal, Isambard, but not as we know it'.
The hand fired Niagara is simply the result of a number crunching exercise. There was research done into what the maximum quasi sustained shovelling rate for a reasonably fit human was, and the answer was ca 4000lbs/hr, possibly on terra firma e.g. a building site. In this country, with hand firing, crews are still reguarly trying to get maximum outputs from their locomotives on fan trips, to please the passengers. It is quite clear that the peak sustained performances achievable are about 2400 IHP; a handful have made 2600IHP. If you do the sums, this says that the fireman is able to sustain 4500-5000lbs/hr for 15 minutes or so. This is the bar to higher sustained outputs. In the short term, 3000IHP is the target, but this involves running down the boiler level. Here 's a loco that's still trying to join the elite 3000IHP club- best effort about 2950IHP over 5 minutes hauling 500 tons at 75mph up a 0.5% grade. In this clip, the exhaust sharpens around 0'15" because the crew have been running downhill more or less without steam for the last 35miles, and the crew are preparing to attack the mile of 1% beyond the station.
http://www.youtube.com/watch?v=v2rmtJ9EUBQ
(OK a bit gratuitous, perhaps best filed under 'Why this subject retains its interest'.)
So, two firemen could in principle fire a Niagara to its normal maximum output, were it not for the fact that a) they would have to perform some kind of ballet across the cab floor; the distance between the tender and the firehole door ideally needs to allow a simple one movement swing and b) they would have to have the energy to fling the coal to the back corners of the enormous firebox.
If the Niagara used most of its 42 tons of coal from Croton to Chicago in normal operation, hand firing would save about 6 tons of coal per trip, not enough to pay for an army of additional balletic musclemen, I think.
I'll comment on Wyoming coal in my response to Erik.
Re: Big Boy vs Allegheny
The biggest difference between the Big Boy boiler and the Allegheny boiler was the firebox. The Big Boy's firebox extended over the rear drivers and thus had to be made shallower than optimal. The Allegheny's firebox was entirely behind the drivers and was therefor deeper. The deeper firebox would lead to better flow of combustion gas plus unburned coal particles and present more direct heating surface. The Allegheny also had overfire jets, where the Big Boy just had ports in the walls of the firebox.
Hand fired Niagara? Is this a hypothetical or does this represent some test where they had guys working in relays?
With respect to the Big Boy, I believe they were using some kind of Western States low-BTU high-ash coal -- there is a lot of advantage with steam to use local supplies of fuel to save on shipping because you need so much of it. Perhaps the closest modern equivalent to the Big Boy fuel would be the famous Powder River Basin coal that many of the railroads today ship to power companies.
A short supplementary on Big Boy Boiler efficiency. I have checked the boiler efficiency data I have for the PRR T1, C&O H8 and NYC Niagara against a model of boiler efficiency I have developed. This is based on the correct first principles, but these principles are expressed as empirical equations derived from UK test data, i.e. are not fundamental scientific expressions, so might not apply outside the range of data on which the empirical equations are based. I was pleasantly surprised to find that the model gets the boiler efficiency on the above three types pretty much spot on. I had to use an expression for unburned coal loss for a mechanically stoked boiler based on the only UK locomotive ever tested this way, and I thought this might be a weakness. Evidently not. Unburned coal loss generally increases with increasing firing rate/sqft (better I think heat produced/sqft grate, which approximates to evaporation rate/sqft). The Table below, derived from the model, shows a smooth increase in unburned coal loss with evaporation rate/sqft for these types. I have included for comparison what a Niagara might achieve with two firemen manually stoking the grate.
Locomotive
Evaporation rate
Unburned coal loss as % fired
Lbs/hr
Lbs/sqft/hr
Hand fired NYC Niagara, maximum service rate
62150
622
11
Mech stoked Niagara, maximum service rate
21
C&O H8 High steam rate road tests
96050
712
26
PRR T1, Very high rate plant tests
1044
38
UP Big Boy on Wahsatch
84750
565
47
The use of mechanical stoking on the Niagara is predicted to lead to a 10% increase in unburned coal loss. At high in service steam rates, the NYC and C&O types are losing about 20-25% coal unburned. At the exceptionally high rates practised at Altoona, losses were approaching 40% on the T1.
The Big Boy tests show that at a lower specific evaporation rates than the other three types, unburned coal loss was approaching 50% of that fired. In other words, there is something fundamentally different about the nature of the coal fired, or the way it was fired on the UP which caused very high levels of unburned loss. It is possible that the Big Boy front end drew more air through the grate than other types, and this may have added to the effect, but I do not think this can possibly be a standalone explanation of these data.
Thanks. I’m not sure when Locomotive engineers began to come to terms with the fact that they were losing 4% of heat due steam from the combustion of the volatiles in the coal. It may have been when the Orsat apparatus for analysis of smokebox gases became widely available- with this you can begin to have a stab at the overall boiler heat balance. Not sure when this was, may have been when interest in the subject was waning. It may not even be mentioned in Lawford Fry’s book (1924)- haven’t got access to all of it..
In terms of overall boiler heat balance, for a simple boiler you lose about 12% of heat in the waste gases+4% in latent heat of steam (these two largely independent of firing rate)+ unburned coal losses which are about 15% at 100lbsq/sft/hr with careful handfiring, increasing rapidly with firing rate, say 70% overall efficiency at 100lbs/sqft/hr. However, if you recycle 10-15% of the steam through a feedwater heater, then in effect this is an improvement to boiler efficiency, giving you the 80% figure you quote. In other words, Johnson’s figure makes perfect sense to me, and against this the Big Boy figure does seem low. The Big Boys only had exhaust injectors, not feedwater heaters, but this would not explain the difference.
In the paper I sent you there is a graph of mine plotting unburned coal loss/sqft vs gas flow/sqft for handfired types, and the one mechanically stoked type tested in the UK. At the kind of gas flow rates we are talking about, mechanical stoking is about 10lbs/sqft/hr worse. However, I have always assumed (it’s not stated) that what the UK testers did was to use the stoker to feed a normal hand fired bed (say 6-12” deep) so the coal burns in the bed. I’m sure there will be references to stoker firing technique in the US literature, most interested to hear what you come up with.
"gravity in Utah is pretty much whait is everywhere else!" - Ha ha. For everybody's sake I hope so!
Thanks for the papers on Front Ends. I'll be looking them over quite a bit before I reply...you're ahead of me on this subject. I've really only been studying the steam locomotive from a design engineer's perspective for a little over a year now.
In a Railway Mechanical Engineer on the 3rd or 4th series of Challenger locomotives, it said they were designed to burn coal of 11,800 Btu / lb. I would assume the Big Boys and other UP locomotives would be designed for similar coal heating values, and 11,800 Btu / lb coal is consistent for coal from Rocks Springs, WY.
As a side note, I never understood why engineers don't use the lower heating value of coal. For example, the Hydrogen content of the coal could easily cause 500+ Btu's to be lost in the latent heat of the water vapor formed via combustion. It's only ~4% loss, but not insignificant. This excludes the moisture content of the coal before firing.
Other references have made comments about stokers causing much of the coal to be burned in suspension, never hitting the grates. I think "Last of the Giants" makes a claim similar. I'll try to look up the sources. The unburned fuel losses of locomotives was quite large, running upwards of 50% at high firing rates (175 + lbs coal per sq ft grate per hour). Simply not enough firebox volume & time for combustion to take place before the combustion gases enter the flues. Ralph Johnson states that modern locomotive boilers produced boiler efficiencies of 80% at firing rates of 100 lbs coal per sq foot of gate per hour due to large grates and combustion chambers ("The Steam Locomotive" pg 82-83). The Big Boys would fall into that modern category, but the low evaporation rate of ~4.5 lbs of water per pound of coal fired does seem low (rate of firing ~125 lbs coal per sq foot grate per hour, not excessive).
Thanks- the fact that the majority of the work is being done against gravity is why I was reasonably relaxed that even though the rolling resistances I was using were quite possibly wrong, I wouldn’t be too far out; gravity in Utah is pretty much what it is everywhere else!
I’ve now had a go at simulating the more detailed log of 4016 on p22 of Kratville as far as Echo. The UP gradient profile is wonderfully detailed, but when the train is often strung out over two or three changes of grade, the best you can hope to get from an analysis of train motion is to get into the right ballpark! In fact, even though I’m using what could be a completely inappropriate rolling resistance equation, I get incredibly good agreement with the reported ‘Drawbar horsepower Corrected’ (EDHP) figures, simulating to the reported speeds and overall time.
4016 was only being worked nearly flat out on the 1.14% from ca MP 981-987, easier on the other stretches (if you count 5000IHP as easy). The peak reported EDHPs of over 6000 are a consequence of the easing of the grade at the point in question, not deliberate advancing of cut off to achieve this. The easing of the grade allows the train to accelerate, and cylinder power naturally increases with engine speed at constant cut off, hence EDHP; at these speeds, the parallel increase in locomotive resistance is not as steep.
It’s at this point that my problems begin. I have the average IHP over the whole trip as 4350, about 7500 IHP-hrs. I could easily be 5% out, but given the agreement with the EDHPs, not much more than that. Now if you look at the reported cylinder consumption, it’s 144000lbs. This implies that the cylinder efficiency (SSC) is just less than 20lbs steam/ihp-hr. Now, making a conservative assumption about superheat (it mightn’t be top notch given the long tubes and huge combustion chamber, N&W A was only 650F), my computer programme, which is a pretty reliable friend says that SSC will indeed be about 20 for the 25 minutes flogging up the 1.14% in near full gear. However, on the easier sections, if the throttle was wide open, and cut off were eased right back to allow efficient expansion, SSC would fall to about 15lbs/ihp-hr, so overall I would not expect average SSC to be much worse than 16, say 120000lbs cylinder consumption. 24000lbs is a huge discrepancy. There will be some auxiliaries e.g. stoker feed of course, but nothing like this. The only reasonable explanation for this is that the power was eased not by shortening cut off, but by part closing the regulator, sorry, throttle, leaving the locomotive working in a very long, inefficient cut off. There are two bits of support for this idea. Firstly Kratville says of the run of 4014 (and with some pride!) ‘All the way up the throttle was open less than full’. Secondly, on the third run made, which was slightly faster, water consumption comes down by 17000lbs to 127000, pretty much where I would expect it. So maybe on the third run, because they were ‘hustling’, they did use full throttle, shorter cut offs. Even across this enormous distance in time, I have huge respect for the crews, so if inefficient part regulator was being used, there must have been a good reason. Question is, what?
Even worse, the report says that the specific evaporation, lbs steam/lb coal is only 4.3. Now the 144000lbs figure for water requires an hourly evaporation rate of about 85000lbs/hr, say 560lbs/sqft/hr. Under this condition, with 13800BthU/lb coal, you will normally find about 7.3 lbs/steam/ihp-hr everywhere in the world. Mechanical stoking causes higher unburned coal loss, because the screw grinds up the coal; absolute worst case scenario on the data I have, 10% worse, 6.4 lbs steam/lbs coal. Now Wyoming coal, from an authoritative source in another thread is only 11700 BthU/lb, 15% worse, hence 5.4 lbs steam/lb coal. But 4.3- another 20%- that’s terrible. The unburned coal loss must have been amazing- Kratville does talk about the ‘rain of the cinders on the roof’. I am at a loss to explain this- the Big Boys were not terrible- they used the same technology as everyone else. I have two hand waving, not thought through possibilities in mind. Firstly, could this be an altitude effect? Like an athlete in Mexico, the thinner mountain air means the Big Boy had to get a greater volume of air into its lungs to produce a given amount of heat. If there is any truth in this, then surely tonnage ratings for climbing steep banks near sea level (e.g. north from San Luis Obispo) would be greater than the top of Donner Pass. The second is the method of firing. I have a quote from a British observer riding an FEF west from Lincoln that I have never fully understood. ‘The coal used was small and dusty, and a feature of its use was that hardly any of it lay on the firebed. At the start of our trip, the fire was only about 1 ½” thick over the grate, and only intermittently bright, and it returned to the same condition at intermediate stops. The action of starting the stoker was akin to using oil fuel, in that the coal ignited and burned mainly in suspension in the firebox and combustion chamber……..as rate of combustion increased, unburned particles increased in volume, and their shooting forth by the powerful blast was the cause of those wonderful pictures of American trains in action where an almost unbelievable column of black smoke rockets into the stratosphere’. This way of working, if practiced would seem to me to be likely to lead to very high unburned losses, although there is no evidence from the Big Boy test photos, of ’unbelievable columns of black some’. Any thoughts on this welcome.
I have a very helpful paper on the fluid dynamics of front ends that I’ll send you. It’s not the finished article, in that at the end of it, you still can’t use the maths to design a blastpipe+ exhaust system that will give optimum entrainment for a given firebed and boiler resistance for reasons I’ll explain. However, I find the conceptual thinking it embodies very powerful.
Dreyfusshudson,
A lot of the work moving the train between Ogden and Wahsatch was fighting gravity. To move 4500 short tons (locomotive, tender & train) up the total climb of 2501 feet in 205 minutes requires about 3300 continuous horsepower. This neglects of course all the other resistances and horizontal distance covered.
Do you have any textbooks or papers that you found helpful for calculating the fluid momentum balance & turbulent jet theory? Fluid mechanics was a course that I didn't really pay attention much too at the university....oops.
Lately I've been studying steam flow through nozzles. The exhaust nozzles (or blast pipe) and smoke stack form the configuration of a convergent nozzle flowing into a convergent-divergent nozzle. The areas of each are known. The mass flow of steam and combustion gases can be calculated fairly well. Any one have references regarding the study of gas velocities through smoke stacks?
Thanks for these comments.
I didn’t have the 18psi pressure for the Big Boy. If you look at the 1943 tests, they covered the 39 miles from Ogden to Echo in about 100 minutes, average speed 23mph, so probably got up to about 30 mph. Loads were about 3750 US tons, 3350 long tons. Average evaporation rate was about 82500lbs/hr, (just over 500 lbs/sqft/hr) blastpipe rate about 75000lbs/hr.
I have had an initial stab at building the line from Ogden to Echo. I covered the distance in about 100 minutes with 3350 long tons of UK passenger stock (!) by dint of a) accelerating slowly, keeping speed to less than 15mph over the first mile- I assume the train was leaving the yards b) accelerating up to 30 mph to the foot of the 1.14% grade 5 miles out of town, peak of 5000IHP whilst accelerating c) working at 90+% of nominal TE to maintain 18 mph up the 6 miles of 1.14%, about 5500IHP, 115000lbs TE and d) maintaining 26-30 mph on the gentler grades beyond there on around 3600-4400IHP. Average IHP was about 4400 over the middle 37 miles. Cut off would be 60+% on the 1.14%, pretty easy working at 30% or less on the easier grades at 30 mph- playing with the train really. Since much of the work being done is against gravity, the choice of the wrong type of stock won’t be too tragic for the power estimates. With the hardest working, there would be peak backpressures of about 18 psi on the 1.14% if the effective diameter of the blastpipes was between 8.5 and 9.0”. This is a bit less than I was expecting, similar to the PRR Q2, less than the 9.6” of the C&O H8, which would have given less than 10psi at this rate of working. However, these calculations are very much preliminary estimates, and hard data on the Jabelmann systems still needed. A lot can change when you know the facts!
If you raise the load to 4200 short tons, as was practiced, you do need full power working at maximum NTE to clear the summit of the 1.14% at about 15 mph.
I understand the logic of the greater perimeter argument, but think it not particularly helpful. You can analyse the front end using standard turbulent jet theory, which says that entrainment depends only on nozzle diameter and the speed of flow through it; with four interacting jets, the calculations can still be done. However, if you do the sums, this standard entrainment model does not give sensible answers. A more fruitful approach I found was using fluid mechanics/momentum balance. The purpose of the chimney set up is to minimise the outlet momentum. Unfortunately this is difficult to calculate from the front end dimensions. What I did was calculate how successful various chimney set ups tested in the UK were at minimising momentum. Quite significant differences, reasons not obvious, but it’s all a long time ago now.
One thing that is really bugging me about my analyses of US Passenger types is how no one seemed to expect to work them regularly at evaporation rates of more than about 600lbs/sqft/hr, 700lbs/sqft/hr an absolute limi, well inside what they could steam at, and well inside what the engine could consume if pressed. Now this is very sensible from the point of view of both coal economy, and I suspect boiler maintenance. So people were behaving very rationally. How strange! Surely this point of view is documented in contemporary literature?
DryfussHudson,
Just wanted to answer some questions about the use of multiple nozzles instead of a single blastpipe. The Challengers and Big Boys had two 4 nozzle plates that sat atop a stand that formed a receiver for the engine exhaust pipes. Also, the pipe for exhaust steam to flow to the exhaust steam injector was connected there. The downside to this configuration is that the back pressures on of one engines could interfere with another engine during its exhaust stroke. The April, 1943 tests recorded a back pressure of 18 psig.
Regarding circular nozzles, the flow of steam is simply proportional to its area. Other shapes of nozzles follow the ratio of area to perimeter, the larger the perimeter for an area reduces flow. This is known as "mean hydraulic depth". However, larger perimeters allow greater contact between the exhaust steam jets and the combustion gases which entrains more gases and thus greater vacuum. Comparing the 8 nozzles to a single large circular nozzle of equivalent area gives:
8 nozzles at 3" ID - total area 56.55 sq in - total inside perimeter 75.40 in
1 nozzle at 8.485" ID - total area 56.55 sq in - total inside perimeter 26.66 in
With regards to the equivalent coefficient on the program, I'll have to get back to you.
Cheers,
It’s maybe time to draw this all to a close, so let me try and summarise what I’ve concluded so far, thanks in great part to your inputs.
You shaped my question into ‘Could Chapelon have improved US Steam Passenger Locomotives?’
Chapelon is revered primarily for his efforts to improve efficiency. In essence his approach had five main elements; high superheat; compounding to maximise expansion ratio; generous steam circuit to ensure maximum steam flow to the cylinders; feedwater heating to recycle otherwise wasted steam heat; and low exhaust pressure through better front end design, all factors together allowing more power to be extracted from smaller, lighter designs. Now US designs had feedwater heaters. So, of particular interest are US front ends, the level of superheat achieved and steam circuits. In as much as there are differences in TE between US locomotives, those with higher TE will tend to work at better expansion ratios than those with lower ones at a given power output. Also critical is the Adhesion weight, for this is a key determinant of how fast you can start, and when you stall on a steep climb. For high speed working aerodynamic resistance (frontal cross sectional area and drag coefficient) is also important- speed was limited to 75mph in France so this was not so much of an issue there.
The Table compares the designs in my survey on these criteria. I have included the Challenger and H-8 for reference purposes. The first point to note is that I have not been able to locate the critical data for UP types, a most fundamental point to which I will return.
Class
J3a
F7
3460
S1b
T1
3765
FEF
Challenger
H8
NYC
MILW
ATSF
PRR
UP
C&O
TE, lbs
41680*
50295
49300
61570
64700
79700+
63800
97350
110200
Inlet steam max F
750?
??
750-800
800
680
???
730
Inlet steam Typical, F
700
670-690
700-750
630
Valve size “
14
13
15
Boiler- valve chest pressure drop, psi
10
Up to 30
Ca 30 ‘too much’
12-18
Blastpipe diameter,”
6.5
7
7.5
7.5?
8
9.6
Adhesion wt, lbs
210500
216000
213000
274000
273000
287000
267500
404000
471000
Frontal area, sqft
155 **
155**
167
155
188
Grate Area, sqft
82
96.5
99
100
108
132
125
*12400lb booster + in practice TE ca 70000lbs on starting, because maximum cut off 60%
** streamlined
To assess the benefits that might have been attained, two criteria are needed; firstly what was the maximum performance these types could deliver on test? My analyses of the actual running the passenger types achieved says that, with the exception of one circumstance, evaporation rates were normally in the 500-600lbs/sqft/hr range, i.e. even the largest types didn’t evaporate more than about 60000lbs/hr- with a reasonably good engine efficiency around 14lbs steam/ihp-hr this translates to typical maximum IHPs around 4300. This is my definition of ‘typical’ working level. (Note that types burning lower quality coal (F7, UP types) needed larger grates to generate a given amount of steam).
The one exception is what I would call the Cajon test. On this steep grade, and others like it, engineers had no choice but to work flat out, using the maximum TE the engine can deliver. This falls somewhat from its maximum nominal level as speed builds up, and steam rate hence backpressure increase, and the pressure drop between the boiler and valve chest increases. In this circumstance, then, the benefits of Chapelonisation would have immediate impact; there is a direct need to generate maximum possible TE as speed/steam rate rises. This logic also applies in spades to freight types such as the H8 hauling maximum loads over mountain districts. For all types, up to about 30mph maximum output is TE limited. At this point, steam consumption is roughly 700lbs/sqft/hr. Beyond this point it rises so sharply if you insist on trying to develop maximum TE that a boiler limit comes into play. Interestingly, on the long adverse grades elsewhere on the Santa Fe, where maximum power was desirable, but not essential- e.g. the long drag from Needles to Kingman, maximum power was not used, just steady steaming at ca 500-600lbs/sqft/hr.
Points to note from the Table are:
TE: the F7 and 3460 are higher than the J3a, and will develop a given power in shorter more efficient cut offs at speed. The ATSF 3765 will also benefit in this respect over the FEF, T1 and Niagara on account of its higher cylinder capacity, and is the ideal type for the Cajon test- the other three would fall by the wayside here.
Adhesion To keep adhesion factors around 4.2, given maximum axle loadings around 65000-70000lbs means that Maximum TE for three coupled axles is about 50000, for four axles 70000lbs. High TE is critical for certain tasks, so there’s no point in lightweighting the engine too much or you’ll reduce adhesion weight!
Superheat. There appears to be only limited data on this most critical of efficiency determinants. In this country, all tests (A type superheaters) showed that it increased smoothly with evaporation rate/sqft, the exact level depending primarily on the number of flues (more = better), tube plate distance (longer = worse) and firebox area (larger e.g. with thermic syphons = worse); it is however something that it was not possible to estimate from boiler dimensions- though I’m sure a good computational fluid dynamics programme could address this today. Superheater areas by themselves are meaningless. Generally typical working levels seem to have been around 700F- pretty good, but possible to improve, as the T1 test results showed, and Chapelon would have claimed. One notable exception is the ATSF 4-8-4, which is about 50F below their 4-6-4; this led to a measurable increase in water consumption vs the 3460 across the desert districts. This however passed without comment! This fits with the overwhelming impression formed form going through these test data – the need for operational simplicity/reliability/low maintenance was far more important than worrying about a few percentage points efficiency.
Steam circuits
There appear to be inexplicable differences here between types which achieved perfectly respectable boiler/ steam chest pressure drops of ca 10psi at normal working rates to the Santa Fe types that showed up to 30 psi loss, something they commented on their test reports. Clearly if this could be improved, the superiority of the Santa Fe types on the ‘Cajon test’ would become even greater.
Blastpipe pressure.
The figures for blastpipe diameter above are, for multiple blastpipes, equivalent single blastpipe diameters. It seems that US designers thought that blastpipe pressure up to 12-15 psi were ‘ok’, and they were happy with these levels for their number 1 priority was good draughting/steaming. At speed, at the maximum steam rates operated in practice say 60000lbs/hr, this criterion was normally beaten. At 60000lbs/hr, blastpipe pressure would be around 14 psi with the 6.5” J3 a nozzle, 10.5 with the F7 7” nozzle, 8 psi with the Niagara/3460/T1 7.5” nozzle and 6 psi with the 3765 8” nozzle. Going from 6.5” to 8”, brings a clear gain of about 500IHP, ca 12% at this steam rate. Going to 9.5” brings another 100HP, but diminishing returns are setting in, and there’s really no need to go above this for high speed working. Under the slow speed maximum output ‘Cajon test’ at ca 85000lbs/hr, 66000lbs TE at 25 mph, blastpipe pressure would be about 14 psi with the 3765 Layden set up. Here even a 10” blastpipe would still give about 5.5 psi, but there would be a very valuable gain of about 3000lbs in ITE.
Streamlining
The difference in aerodynamic resistance between the massive, blunt 178 sqft envelope of the 3765, and the streamlined 155sqft front end of the Milw F7 is of the order of 200HP, the benefit going straight through to the drawbar. I’m no aerodynamicist, but the streamlining on the T1 and J3a looks more cosmetic to me. Running at 100mph, there would be a big benefit for the F7; other types running more slowly less so; in fact its seems to me that the ATSF 4-8-4s spent much of the time either working hard but slowly uphill, or at high speed coasting without power and braking downhill, so for them streamlining would have brought virtually no economy benefit.
In general, I conclude that all but the T1 would have benefitted from higher superheat at normal operating speeds- the T1 boiler shows this can be done- perhaps 50 degrees more being a reasonable target. All would have gained significant benefit from blastpipe diameters about 1”-2 “ greater than they had. Together these two benefits together could have given improvements in efficiency of 10-15% at maximum daily working rates. Whether this would have been worth the maintenance trouble, I cannot say, but had steam had a future, any issues would surely have been solved.
The reason the absence of data for the UP types in general and the FEF in particular is so critical is that Jabelmann was said to be ‘obsessed with back pressure and superheat’. It is possible therefore, particularly in the absence of thermic syphons, and with a shortish (19’) boiler, that they already achieved superheat approaching 750F at normal steam rates. Kratville gives pictures of a Jabelmann exhaust, but nowhere can I find what the diameter of the eight nozzles were. 8*3” would give the equivalent of 8.5“ single diameter; 8*3.5” almost 10”. So it may be that the ‘Chapelonised’ 4-8-4 already exists, and is heading to St Louis as I type. It would not however be as good as an ATSF 4-8-4 on the ‘Cajon test’. The C&O H8 already ticks all Chapelon’s boxes.
I have a couple of leads I am still following up to try to establish what the Jabelmann nozzle diameters were. I shall wander the planet, a lonely soul until I find out. Perhaps the simplest thing to do would be to hop on AA 55 tomorrow with a tape measure, and head to St Louis on Friday night, and get into 844’s smokebox. I feel that these days such behaviour would lead to me being incarcerated for my troubles, but prisons can be quite comfortable now.
Dear CS,
Thanks for this.
Vuillet also claims that an F7+ 14 got from Milwaukee to Chicago in 65 minutes, which I thought might be possible, but looking more closely, some features e.g. the passing time to Sturtevant and cruising at 107mph are impossible and very unlikely respectively.
This is odd because Vuillet is a very respected and knowledgable source, so he must be quoting another source, perhaps word of mouth.
I don't wish to appear to be knocking the F7s, they would probably get my vote for the gold medal; accelerate from rest at the limit of TE, get up to 100mph and stay there. Take a few extra coaches if needed. No other steam locomotive I know of did that day in day out. But such gold medals are very subjective of course.
Vuillet's claim comes out to an average speed of 80.95 MPH based on 85 miles for Milwaukee to Chicago. Without even considering the locomotive's limitations, various speed restrictions in the terminal areas would preclude such a high-speed run.
Dear Firelock,
Many thanks- in a separate stream of correspondence I commented that 'it was a small world' even in the 1920s- the latest developments in US steam technology were avidly looked for in Europe, and I guess the same would be true in reverse. So, very interesting to have confirmation that US designers were aware of European efforts in exhaust design, but with hindsight not too surprising.
I'm busy trying to focus everything folk have sent me into an article along the lines of 'would Chapelon's initiatives have made any difference to latter day US steam', taking the J3a, Niagara, T1 F7, 3460 and 3765 classes as representative. (Sorry N&W and UP fans, data lacking). If anyone would like to critique this prior to submission for publicaiton, I'd be more than happy to have your comments- it'll be a couple more weeks in gestation.
If there's anyone still out there, I have come across something that's in the 'I don't believe this' category. In my analyses of 'Normal' F7 running on the Afternoon Hiawatha, I conclude that the locos were worked close to maximum NTE accelerating from rest up to about 30 mph, when it's the boiler rather than engine limit that comes into play, and worked at just less than 4000IHP to accelerate to 100mph, after which you could cruise at this speed at 2500- 3000IHP. There are several runs that fit this pattern, including a run from Milwaukee to Chicago in about 68 minutes net with 9 cars - the fastest the author knew of. ( See Hiawatha Story). This is comfortably within the computed performance envelope for an F7, but well ahead of what was expected from a J3a it would seem.
However, there is a claim in a book by Baron Vuillet of something of a different order. It is said that Milwaukee to Chicago was run in 63 minutes with sixteeen cars! This involved covering the first '12miles in 12 minutes' which is truly impossible with this load; I think even 13 minutes would be sensational, and would require e.g. crossing the Kinnikinnic drawbridge at 50+mph. The 'next 62 miles were covered in 37 minutes at an average of 100.7 mph'. Again, I think this is more or less impossible, unless there was a howling northerly gale blowing. That all gets you to Edgebrook in 49minutes, leaving 14 minutes for the last 11.5 miles; whether this is possible depends on what the speed limits at Mayfair, Pacific Junction and Westen Avenue were at that time (1943- Vuillet said some limits had been relaxed by then), but with a slow run in round the bend at Canal Street, this also sounds unlikely to me.
Does anyone have chapter and verse on this claim?
Just have to add one more thing to "The Tread That Will Not Die!" I was consulting my copy of " The Steam Locomotive In America" (1952) and lo and behold there was a comment on exhaust nozzles and backdraft. Seems the locomotive builders and railroads themselves were always aware of back pressure problems, but they never found a practical solution to the problem. They DID know about adjustable exhaust nozzles that were in use in other countries but decided that the constant attention they needed to get the best out of them, plus the increased maintanance costs made them just not worth the trouble. It was easier to live with the inefficiency, which really wasn't all that bad.
Many thanks for this, and firstly let me reinforce Firelock76’s comments about your work in Iraq. I travelled a lot in the 1980s with a colleague who had had the Middle East as his patch in the 1970s. I remember him saying that Baghdad (and Tehran for that matter) used to be a beautiful place and I am always haunted by this when I see images of there now.
I would be delighted to share the programme with you; I think I will need to mail it you with supporting documentation, since you can’t send .exe files by Outlook, so if you send your name and address privately, I will pull a package together and send it. One of its great virtues is that it’s incredibly easy to use, and you can do 10s of calculations in an hour to explore all sorts of hypotheses. Likewise I would be very interested in your work on the Big Boys. I now have Kratville’s book on loan, along with King on the As and Farrington on the Santa Fes Big Three. I’m giving full attention to the latter, since it is such a treasure trove of the kind of information I am after, not paid much attention to Kratville yet, though I do intend to ‘build’ the line Ogden to Evanston, to see if I can simulate the test runs of 4014/16 in terms of efficiency coal consumption and work done, a significant project some way off until I sort the Santa Fe. This will be a great test of how good the models I have of engine efficiency, boiler efficiency and locomotive and train resistance actually are, way outside the range of conditions I’ve tested them out in until now. Maybe you’ve gone a long way towards this?
Thanks for the dimensions of the UP front ends. The FEF2-3 have an area equivalent to a single blastpipe of just over 8¾ ”. However, looking at the pictures on pp 28 and 29 of Kratville, I am wondering what the discharge coefficient through these openings would be. As I understand it, each of these four orifices sit atop a single wide blastpipe? I ask because over here, single blastpipes were normally used; comparing the programme’s output with backpressures measured on test, shows that good fits are obtained, as expected, with a blastpipe discharge coefficient of 0.99. Late in the day, some designs adopted what I believed was an American idea, namely to sit an orifice plate across the nozzle, giving a sharp edged aperture; this not only reduced the diameter, but also decreased the discharge coefficient to about 0.85-0.90, thereby further increasing backpressure. It could be that this UP set up had a discharge coefficient of less than 0.99, hence in effect narrower diameters. (I am not clear what the rationale for the orifice plate was; if you say that the way to increase draught at a given exhaust choke is to increase the difference between the momentum of the exhaust jet and that at the top of the chimney, you can increase draught by increasing jet momentum, achieved by the orifice plate, but at the expense of even higher back pressure than normal; the purpose of the chimney is to get the minimum exit momentum for a given inlet momentum, so maybe the orifice plate (or perhaps also the cross wires found on certain nozzles) was to increase turbulence in the jet, which made the chimney more able to achieve a flatter exhaust profile hence lower momentum, hence better draught? It is the inability to easily calculate the momentum minimising ability of the chimney set up which seems to me to have hobbled exhaust design).
The dimensions of the Big Boy and Challenger exhausts, even if discharge coefficient is 0.99 are smaller than the figure I have for the C&O H8 (roughly 2 *7” blastpipes).
I am bit like a kid in candy store with the Farrington material, but among the things that strike me are a) that the Santa Fe was far more interested in the temperature of the motion bearings than that of the inlet steam! Clearly running hundreds of miles from A to B without running hot was a higher priority than saving a bit of oil. Strange people. In fact, though the data is not entirely clear, the superheat on the 4-8-4s is mediocre, certainly much worse than the 4-6-4s, and this seems to be related to very low gas temperatures at the end of the 21’ long tubes. b) I was amazed that the maximum cut of was fixed at only about 60%, reducing tractive effort from about 80000lbs, as given by the normal formula, to about 70000lbs. And yet they were sent up Cajon pass, with gross train weights of 1200 long tons i.e. about 60000lbs worth of gravity to deal with unassisted! Keeping the boiler pressure at 100% must have been absolutely essential. c) There is to British eyes a very surprising increase in the pressure drop between the regulator and steam chest as steam rate increases, much larger than found here, although of course here maximum steam flows were far less. d) At speeds above 60 mph, the cylinder pressure does not even reach steam chest pressure at the beginning of the stroke (the programme reproduces this effect on its predicted indicator cards for the 4-8-4 well) though this is not the disadvantage it might seem and finally, and sadly e) the speeds claimed for 2900s climbing from Dodge City to La Junta in excess of 90mph (96mph in one case) are absurd, though going eastwards on the slight downgrades would be easily possible. In fact, having now built the line all the way from San Bernadino to Barstow, if the regulator on the 3766 had been closed at Cajon summit when speed reached ca 40 mph, it could have remained closed all the way to Barstow, maximum about 123mph just before Victorville, still 60 mph at the West end of D yard(!) I hadn’t realised that the line goes back down nearly to sea level at Needles. The eastbound grades from there are awesome; next challenge is to see how 3776 coped with the 80 or so miles of 1/70 grade on the way to Wampai, and I will be ‘building’ this line next.
Zu Burgard540: Ach du lieber! Ein poster mit ein "Pour le Merite" avatar! Either you iss a real romantik or der last Fokker Dreidekker pilot livink! Seriously, welcome back from Iraq, glad you're home safe, and thank you and God bless you for your service!
DreyfussHudson,
I just saw this thread (haven't been on here for a while, in Iraq) and the program you're using sounds interesting. Can you share it? I've been doing research on the UP 4-8-8-4 Big Boys, and it would be interesting to see the results. The Big Boys, Challengers, and FEF's shared many design characteristics and by extension many of the blueprints are shared as well. Kratvilles books regarding each class are good, but often reference each other. I bit the bullet and bought all three. I'd be happy to send you the tables of data and analysis I've done based on the data from Kratville's Big Boy book that I have made in Excel.
Each class of the locomotives utilized a double stack design (except the first FEF class, which were later modified to double stacks), with two multiple jet exhaust plates that had 4 nozzles. It seams that UP made similar conclusions regarding nozzle design as Porta's Lempor drafting design. The inside bore diameter of the tips were (Reference UP Dwg 443 CA 33736, 16 JAN 1951):
FEF-1: 2-13/16" (Total Area 49.70 sq in); FEF-2-3: 3" (Total Area 61.36 sq in)
4664-3-5, Challengers: 3" For oil (Total Area 56.55 sq in), 3-1/8" For coal, (Total Area 56.55 sq in)
4884-1-2, Big Boys: 3" (Total Exhaust Area 56.55 sq in)
The April 1943 tests showed that from Ogden to Echo the per hour running time the average water evaporated by the boiler was 83,200 lbs/hr, condensate return (via exhaust injector) 6830 lbs/hr, coal fired 19,100 lbs/hr, developed 5300 drawbar horsepower, speed 29 mph, and drawbar pull of 68,800 lbs. Only 4.36 lbs of water were evaporated per lb of coal, and the coal was fired at a rate of 127 lbs per sq ft of grate per hour.
From Echo to Evanston the per hour running time the average water evaporated by the boiler was 80,600 lbs/hr, condensate return (via exhaust injector) 6450 lbs/hr, coal fired 17,900 lbs/hr, developed 4560 drawbar horsepower, speed of 16 mph, and drawbar pull of 103,500 lbs. Only 4.52 lbs of water were evaporated per lb of coal, and the coal was fired at a rate of 120 lbs per sq ft of grate per hour.
Average boiler pressure of 294 psig, 96% full throttle, & reverser 26 notches ahead of center (what that means in terms of cut-off I don't know). This was pulling a 71 car, 3883 ton train. Regarding total steam temperature, Kratville mentions on pages 13 & 18 of "The Mighty 800" that the Type E superheater would supply steam at 750 degrees or better to the cylinders igniting leaking lubricants!
I look forward to the discussion,
Thanks, Spain was great. The only thing that distresses me is that about 6 miles from where we live, we have a brand new TGV station, two island platforms long enough for double length TGVs, two arrow straight 200 mph centre road fast tracks separating them. At present, it is the terminus for two daily round trips to Paris with a single TGV, plus two connecting services to Barcelona. In addition there are four freights per week. The whole project to Barcelona is about 4-5 years behind schedule. I think I'm paying for this somehow, probably most other folk too.
Kratville's book on the UP had arrived. In the absence of solid data yet from Farrington, I checked out the run of the Super Chief on 31st April 1953, when a 2900 had to take over a consist swollen with business cars to 850 long tons due to diesel failure at Kansas City, working a 39 3/4 hr schedule. On the long uphill 2500' climb from Newton KS to La Junta CO, the 2900 covered the 355 miles in 4'50", just inside schedule, including a 10 minute servicing break at Dodge City, working around 4000IHP, and cruising pretty consistently at 80mph. Maximum was about 85mph in the slight dip before a 30 mph service slack through Hutchison.
Welcome back! How was Spain?
Always the optimist, I went to Englewood 10th May 1974 to check out the ghost of the Broadway /Century race. What I got was a picture of my conveyance there (the 17.40 to Valparaiso) with Tuscan red PRR coaches, hauled by two ex NYC E8s 4039 and 4042, which I guess might have hauled the Century? It all seemed very sad at the time, but on reflection, as good as you could hope for. I was the only person to alight at Englewood. From there, I think I must have walked under the tracks through this dangerous part of town to take the Dan Ryan El, for my next shot is at Pullman Junction, EL RS3 1042 hauling a short freight, on the third car of which was a small steam locomotive, possibly narrow gauge. Any ideas what this was?
On my vacation, I took the trouble to read the stuff I have already got more carefully. (Note to self: must do this more often). I have the executive summary of the NYC Niagara report, which has a big clue to one of my main queries. It talks in detail about the acceleration trials I mentioned earlier, and it was pretty much as I supposed- the locomotives were worked at full capacity up to 75mph, exceeding 6000IHP at about 60 mph, higher thereafter. However, there is another table which gives the game away on my basic query, for it gives the requirements to deliver 4000IHP required for ‘above average passenger train operation’. In other words, Niagaras could deliver 6000+IHP, but in practice were not required to deliver more than 4000 IHP, evaporating no more than about 650lbs/sqft/hr. This is remarkable concurrence with operating practice in this country- where it was of course possible, but rarely if ever required to steam above 650-750lbs/sqft/hr (this with 12500-13500Bthu/lb coal, as on the NYC)- economics of boiler maintenance, not to mention fuel economy probably has a role here.
To illustrate what 4000IHP can do, the Table below shows some simulations of running from Toledo to Elkhart (all weights long tons). The run of streamlined Hudson 5450 is a simulation of an actual run that got to Elkhart in 112.5 mins net: this simulation uses no more than 3000IHP at any point. The first run of Niagara 6023 has the same load, but working at no more than 4000IHP; it gets to Elkhart 10 minutes faster. Alternatively, the last simulation shows that 6023 could get to Elkhart in the same time as 5450 at 4000IHP with an additional 300 tons (say four coaches). (All these are ‘rough cuts’; there are some aspects of the resistance equations I’m using that I’m not happy with, to be fixed shortly).
Engine
5450
6023
Tare wt
900
1200
gross wt
950
1250
Engine wt
350
400
total wt
1350
1650
miles
est speed
est time
inst speed
0
Toledo
0.01
00:00:00
8.85
Holland
65.26
00:12:23
72.02
00:11:59
65.06
00:13:22
18.45
Swanton
70.58
00:20:48
79.55
00:19:31
71.76
00:21:43
24.35
Delta
70.93
00:25:48
80.55
00:23:56
72.48
00:26:38
31.7
Wauseon
70.57
00:32:02
80.61
00:29:24
72.32
00:32:43
36.35
Pettsville
76.19
00:35:50
85.92
00:32:45
78.25
00:36:25
40.45
Archbold
79.87
00:38:59
89.39
00:35:33
82.09
00:39:28
46.2
Stryker
81.42
00:43:15
90.94
00:39:22
83.80
00:43:38
53.45
Bryan
74.33
00:48:52
85.40
00:44:20
77.68
00:49:03
63.9
Edgerton
70.10
00:57:36
80.79
00:51:56
72.43
00:57:28
70.45
Butler
72.67
01:03:06
82.91
00:56:44
74.74
01:02:48
78.15
Waterloo
72.46
01:09:28
82.68
01:02:18
74.45
01:09:00
84.8
Corunna
67.34
01:15:12
78.14
01:07:17
69.38
01:14:34
91.05
Kendalville
74.85
01:20:25
83.20
01:11:55
75.31
01:19:44
102.3
Wawaka
85.17
01:28:45
92.74
01:19:31
01:28:00
107.8
Ligonier
84.92
01:32:38
92.53
01:23:05
85.75
01:31:51
114.75
Millersburg
80.44
01:37:42
88.46
01:27:42
81.36
01:36:51
123.1
Goshen
90.12
01:43:32
97.35
01:33:03
90.98
01:42:37
128.75
Dunlap
87.75
01:47:21
95.15
01:36:35
88.64
01:46:24
132.8
Elkhart
1.97
01:51:27
1.65
01:40:43
2.48
01:50:47
Looking at these outcomes, 70-80 mph start to stop averages with 14-17 cars, I would be surprised if more than this was achieved on a daily basis from a Niagara, or indeed any US 4-8-4, except perhaps when climbing steep grades. In this case, with exhaust rates not much above 50000lbs/hr (allowing for feedwater heater recycle), then with blastpipe diameters of 7.5+”, there really is no ground for Porta’s implied critique that more power/less fuel could have been achieved in the US with better designed front ends. Unfortunately the British interlibrary loans service is in chaos, so the books on the ATSF and N&W 4-8-4s that might provide counter evidence still haven’t arrived.
Thinking more generally about US designs, there was a practical limitation of about 30 long tons driver axle loading, say 210000lbs on 3 axles, 280000lbs on 4. Now there was also a desire to keep the adhesion factor (ratio of axle load: tractive effort) not less than 4, so TEs of around 50000lbs for 3 axles, 70000lbs for four axles. Now if you say that at speed, you don’t want cut off to exceed 30%, for efficiency's sake, then the question is, how much HP do engines with 50000 and 70000lbs TE develop at 30% cut off at ca 80mph? Now obviously this depends a bit on some design factors, but at reasonably good superheat, the answers are ca 3900IHP and 4200IHP. (Note that IHP does not increase in proportion to TE at speed). So at 4000IHP a 4-8-4 would be working in an economical regime. At the risk of getting caught in partisan cross fire, I would also observe that going to four cylinders, as per the T1 does allow slightly shorter cut off working to get the same steam flow and power, but that the difference in efficiency due to a few percent in cut off at 30% or less is pretty small.
We (the Broadway) got the highball a few seconds before the Century, and then it took about ten minutes for the Century to pass us. He started while our obs was about even with the sleeper ahead of the dining car, which if I remember correctly was the two-unit type. Our train was all red (E-units excepted, of course) and his was all grey. All equpment matched at the time. I think the year was 1959.
YOu could arrange for a porter to have your pants pressed on the Century at the time. I availed myself of this service and tipped appropriately. This was not noted on the Broadway. But the dining car food was excellent on both trains, and the interiors and general upkeep was very good at the time.
I contiued to use rail between the east and Chicago up to 1995, and I can also recall the worst Penn Central period. Also one trip on the Erie-Lackawanna Lake Cities --- with no complaint.
To daveklepper, that must have been a fun race to witness. I envy you!
Anyway, I am glad I witnessed this race. From Moutain View or Tower View on the rear of the Broadway (never dreaming that some day I would be a small shareholder in the owner of Mountain View, Pullman Classics Limited). As I recall we lost and the Century won. We had head end cars added on the Broadway, while the NYC (both still all-Pullman) began with the baggage dorm and had a shorter consist. Both trains were pulled by EMD E-units at the time. But I had traveled to Chicago on the Century, returning via the Brodway, change to Clocker at N. Phila, MU at Trenton, Princeton Jc. and back, and then mu to NYC.
You know, Dave, I have no idea! The original teller of the tale didn't specify what was pulling the trains or when the incident occurred. We just have to accept is as a fun bit of railroad lore and let it go at that.
Now I want to know: If the Century was headed by a Niagra, was the Broadway by a T-1, or two K-4's? If it was a J3A vs single K4 contest, well the winner would be quite obvious.
Concerning the New York Central's high speed runs to and from Chicago, there's a great story I'd like to pass on.
Both the 20th Century Limited and it's arch-rival, the Pennsy's Broad Way Limited, used to depart Chicagos Englewood stationat the same time on adjacent tracks. This always (well almost always) led to a race between the two, to the absolute delight of the passengers. Mind you, OFFICIALLY the railroads said they weren't racing, and OFFICIALLY the train crews were forbidden to race.
One day, a NYC official was aboard the Century, when he felt an abnormal acceleration. He looked out the window and saw the Broad Way running "neck and neck" with his own train. As the Broad Way sped up, the Century did as well. He saw a conductor passing down the aisle and called out "Conductor!" "Yes sir?" replied the conductor. "Step this way, sir." and the man came over. "Are we racing the Broadway?" "AHHHH, ERRRR, OHHHHH," stammered the conductor. "I see," said the official. "Get word to the engineer. If he looses, he's FIRED!"
Crandell,
I enjoyed reading Lars Loco's posts as well.
Just a FYI, and a bit startling, but lars loco has asked that his account be deleted. He reports that he can no longer devote the time to the forum that he was used to. I think it safe to say that he will be missed. I wish him well.
Thanks for more top rate inputs.
With respect to Firelock 76’s (20.4.11) comments, this is an almost perfect articulation of my belief, namely that, like any successful species, the North American steam locomotive was well adapted to its environment, (operating, social, political, economic and so forth). (It is by the way true that in this country as in the US, locomotive efficiency was relatively low in the operators priority list, getting the train over the road near the top; in the heyday of steam our railways were privately operated also, fighting to stay alive from the 1920s on as road transport took off, fighting for the country’s existence during the war years, so most of the other considerations about US railroads also applied). It has been suggested by Porta that efficiency improvements to US designs might have been possible by reducing backpressure. I am looking to establish if this was indeed the case; my view is that, given the relatively low priority of efficiency, any improvement would have to be very significant for it to have had a worthwhile impact. I do think that it is likely that the answer to this question will turn on how hard locomotives (let’s restrict it to passenger locomotives) were actually worked so I will come back to the point about the southbound mail later.
With respect to NM_Coot 20.4.11, we are I think in violent agreement on all points. I know the gentleman who translated Chapelon’s book into English, and he makes sure I don’t deviate from the straight and narrow, though of course there are factors that with 20:20 hindsight, which is what high speed computing can give you, aren’t quite as important as Chapelon thought. On point 4, you speak of ‘Nozzles’ How many did the 3460 class have? I will come back to NYC performance.
With respect to feltonhill 20.4.11, yes, I’m absolutely certain that the 750F for the NYC Hudson is inlet steam temperature, not superheat level. Given that I believe that superheat is the dominant factor in determining efficiency, it would be good if one could estimate what would be delivered from published dimensions. I have a book with a most unpromising title ‘The Great Book of Trains’ written by two top rank British writers, Hollingsworth and Cook, which is remarkable in that it makes a great general interest coffee table book, (super line drawings), and, as far as I can tell also provides reliable technical details on hundreds of steam locomotives from around the world, and exceptionally knowledgeable write ups. At $10 second hand on Amazon, it’s an absolute snip! I’m telling you this because last week, I spent 2 hours extracting all the technical data on the US designs featured, to try to address this point. I was surprised how different some of the key dimensions were, particularly in quoted superheater areas. (Sadly, my computer ate the spreadsheet, so need to start over!). However, I believe that superheater area itself is not the sole determinant of superheat achieved. Test results over here showed that superheat always increased with the number of flues; however they also showed that if the distance between the tube plates was more than about 18’, superheat begins to go down, possibly because the flue gases are cooler than the steam in the last few feet of the flues, so the steam gets cooled. Obviously, the longer the flues, the higher the superheater area at a given number of flues, but this does not guarantee high superheat. We found best practice was to have not much more than 17’ between the tubeplates. The class with the largest superheater area over here did not generate the highest superheat, in part I believe because it was 19+’ between the tubeplates. From the very limited US data I have, this also seems to apply, in that the T1 (18’ between plates, 1430 sqft superheater) delivered higher inlet steam temperatures (750+F) than the Niagara, (20’ between plates, 2060 sqft superheater, which gave ca 680F at 52500lbs/hr). This is a point I would like to explore further, hoping the texts I am getting will provide details on this important dimension. But 750F for a Hudson does not seem unreasonable to me.
With respect to NM_Coot, 21.04.11, thanks, this is just the kind of data I’m looking for. With respect to the 3460 class, the 44 sqins nozzle are must mean they had four 3.75” nozzles? Depending on the steam rates normally used, going down to 40 sqins wouldn’t be too tragic (See Below).
Now for some more analyses, which I’ll try to keep simple. Hollingsworth and Cook give a quote for a Hudson hauling a 940 tonne 20th Century, which made it from Toledo to Elkhart (133 miles) in 112.5minutes net, and on to Englewood (93.9 miles) in 79.5 mins, both about 71mph start to stop. I am assuming this is an example of very good running. Since it’s basically downhill form Elkhart to Chicago, and I’m not sure if speed restrictions applied beyond Gary, I’ve ‘built’ the line from Toledo to Elkhart. What I found was that, using standard resistance equations, and assuming a relatively modest start, I could get to Elkhart in 111.5 minutes without exceeding 3000IHP. On the gentle upgrades to Wauseon, speed rose to about 71mph (32mins). In the gentle dip down to Stryker, speed rose to 81 mph (43 ¼ mins), before falling to 67 mph up to Corunna. Finally, on the descent to Elkhart speed rose to 90mph. covering the 44 miles on to Dunlap in 32 minutes, and averaging 84mph over the 38 miles from Kendalville. If a more vigorous start had been made, power required later would have been less. Now my computer programme says that the Hudson would need about 42000lbs/hr to the cylinders to deliver 3000 IHP, and even if the Hudson blastpipe diameter were only 6.5”, this would not lead to excessive back pressure. So, adding this to my analysis of the MILW F7, and extrapolating wildly from just two data points, this says that the actual working rates of these Hudsons in normal service would not lead to efficiency losses due to restricted blastpipe dimensions. Now I’m sure an NYC Hudson could sustain much more than 3000IHP. The question is, whether there is a run on the proverbial 8 hour late southbound mail that demonstrates this.
The second set of analyses was to look at the acceleration tests with a Niagara east from Utica with 1005-1875 short tons. The line is pretty flat, perhaps a very slight downgrade to the canal bridge. Now these represent the most astonishing running I have ever come across. I only have the bare details of loads, times, and distance to reaching 75mph, but the times reported can only have been achieved by working the Niagara in full gear right from the off, with power reaching about 6000IHP at about 45mph. From then on, cut off was gradually eased to maintain just over 6000IHP until 75mph was reached. I’m almost speechless! The sight must have been unbelievable, blastpipe pressures phenomenal. Now I recall David P Morgan reflecting on how every steam fan in the early 1950s knew that beyond about 20-30 mph, a Niagara could outperform a contemporary 6000 HP ABA E unit, and this is what these tests surely demonstrate. The point Morgan was really making however was that, even though this was true, it was irrelevant. I believe also, that even though this performance was surely possible, it was rarely needed in practice, except on the 8 hour late mail. If you put a Niagara on 1200 tonnes east from Toledo, accelerate more gently than the above tests, and work at up to 6000IHP on to Elkhart, then, having self imposed a 90mph limit, I get to Elkhart in 96mins, running at 90 mph up hill and down dale from Swanton on. Maximum power is about 5700IHP, less on the downgrades. Did a Niagara ever do this in normal service? I doubt it, but would love to be proved wrong. So, I suspect even the Niagaras were actually worked at steam rates which did not involve excessive blastpipe pressures. All data to the contrary very welcome.
Heading off to Catalan Spain for 10 days from tomorrow, to a computer free zone, so will be off the case for a while.
NM_Coot - Thanks for the figures! I have two of the three sources: NYC 4-6-4's (Reed) and Santa Fe's Big 3 (Farrington). I assume that Profiles 26 is the Milw F7 and/or A? Don't have that
Farrington's book is the real gem here. He apparently had the clout or $$ to put what he wanted into his books. As you already know, and others should be aware, he frequently puts in copies of actual test reports. He doesn't paraphrase or editorialize. He lets the railroad's test department do the talking. It's great stuff.
On the other hand, Reed tends to include a lot of numbers so that his analyses look technically rigorous, but he doesn't always use consistent data in his comparisons. Some of his other articles where I got into the details, he got a bit fast-and-loose at times.
This will keep me off the streets for a little while. Thanks again!
Here are some numbers with sources.
Loco Profile 2, Table II, text. Loco Profile 26, text. The Santa Fe's Big Three, text, tables, figures. I've compared three contemporary 4-6-4's. I put in some numbers that may be of use to others. That's all! This is not a mine's bigger, better, whatever, effort.
Railroad NYC CMStPP ATSF
Class J-3a F-7 3460
builder ALCO ALCO Baldwin
year 1937 1938 1937
boiler pressure (psi) 275* 300 300
grate area (ft^2) 82 96.5 99
total evap surface (ft^2) 4187 4166 4760
superheat surface (ft^2) 1745 1695 2080
superheat temp (deg F) 750 *** 775
superheat/total evap (areas,%) 41.8 40.6 43.7
total evap/grate area (ratio) 51.1 43.3 48.1
wheel dia (in) 79 84 84
valve diameter (in) 14 12 13**
piston diameter (in) 22.5 23.5 23.5
max travel (in) 8.5 7.5 7
exhaust nozzle dia (equiv in) 7.25 7 7.5
choke/stack dia (in) 21.5 18.5 24
max hourly evap rate (lb) 93,000## ? 89370#
max inp (hp@mph) 4900@75 ? 4350@65
max dbhp (hp@mph) 3935@65 ? 3600@50
* Later 265 psi. Designed for 300psi.
** Open throttle indicator cards show drop of about 30 psi, boiler to valve chest.
# 1st District, Arizona Division, heavy grade at 45 mph, 3997 ihp, steam consumption was 68700 lb/hr.
## Test, Selkirk.
*** Class A superheat was 680-700 deg F. F-7 no reference value.
? No reference value in my library.
I think the numbers support a conclusion that the F-7 superheat was low typical. The author of Loco Profile 26, Brian Reed, says, with regard to the A class: "This was not a very high degree of superheat." I think the numbers above would tend to indicate a similar condition for the F-7. The users of this data are completely free to draw a different conclusion!
Regards and happy contemplations
NM_Coot
Add flue sanding, keeping any necessary and installed smokebox netting clear, petitcoat pipes mounted correctly....they were complex machines that needed skilled and experienced, not to mention attentive, operators.
MEGO also!! Help me out here. I'm as good a number cruncher as the next person but I need some assistance.
The NYC J3a is credited with a superheat of 750 degrees. I'll bet this is total steam temperature, not superheat. The J3a's operated at 265 psi, which would be a saturated temperature of about 410 deg F. For a total steam temp of 750 deg F, the superheat would have to be about 340 deg, which is a very high figure. Where did it come from? I'm not doubting it, just want to know the source.
I've re-read most of the thread and would also like to know the source of the MILW F7's "low" superheat? Anyone have the figure? I've never read anything that indicated the F7 was deficient in that area. But stuff happens in any design. Maybe I haven't read enough.
NMCoot - I believe you're in good company with the water treatment comment. Wasn't Porta a proponent of this idea?
Some brief comments if I may.
1. Water treatment is far more important to good boiler operation (and maintenance) than feedwater heating.
2. Chaplelon was far more concerned with the entire steam circuit than he was about the exhaust configuration per se. The exhaust point is but one of many critical points in the steam circuit. He was as concerned about the drafting interaction with combustion as he was about back-pressure.
3. I don't quite understand the pokes at Chapelon. He was a modest engineer who achieved remarkable results. In fact, I will go so far as to say that if you do not know and understand Chapelon's work you do not really understand all the important elements and inter-relationships of modern Stephenson locomotive design. His book which summarizes his fundamental work was published (in French) in 1937 and revised and expanded in 1952. A English translation was published in 2000. The contents are as valid today as they were in 1937.
4. Stephenson steam locomotive design is a constant trade-off of factors. Once you get the exhaust about right, continued tweaking isn't going to gain much in day to day service. For example, soft exhaust often led to problems with smoke and water vapor obscuring forward vision. Farrington, in The Santa Fe's Big Three, reprints many Santa Fe test reports. The 3460 class developed 4350ihp at 65 mph and 3600dhp at 50 mph. Because of drifting smoke, the ATSF test engineers gradually reduced the nozzle openings from 3 3/4" to 3 1/2" and the stack diameter from 26" to 24" without any appreciable change in performance. The smoke problem was particularly problematic when working lightly. There are also some interesting relationships between firebox draft, smoke box draft and oil atomizer pressure. How about that folks! Anyone interested in seeing how a railroad evaluated locomotives in service should read this book.
5. As best I can research, the Milwaukee F-7 class is a development of the A class. These two classes handled low tare weight trains at high speed over a relatively flat line. ALCO essentially designed both. The 84" drivers were used to keep piston speeds within reasonable limits. ALCO did pay attention to the steam circuit. The relatively low superheat is unusual. the A class could do about 3000-3300ihp and I would venture that the F-7 probably could do about 3500-3700 ihp (primarily because of rather low superheat). In the example given, a total weight of 855 tons with an engine weight of 395 tons really means that the train was there to provide braking effort. Both classes did what they were designed to do but of the two the A class is really the big step forward in performance.
6. NYC J-3a's regularly handled trains of 800 to 1000 tons tare and regularly achieved schedule service speeds of 80-90 mph. In my opinion a J-3a could handle the F-7 service and I doubt the opposite is true. The only question might be the 79" drivers on the J-3a but I think the proven lively performance of the J-3a would compensate. The ATSF 3460 class Hudsons had 84" drivers and could handle trains of 800 to 1000 tare at 90 mph when schedule compliance was required.
7. Ther are numerous examples of excellent US design practice. So the point isn't making "the case against US front ends." The point was simply that when small locomotives significantly out-perform very much larger ones, the designers first step in making the larger ones perform better is to look at what makes the small ones perform so well.
Regards
Ok folks, I think we're getting to the point here where we can't see the forest for the trees. All this technical data is starting to make my eyes glaze over and my head swim! I'm not an engineer, but I am an amateur historian, so let me give you all a little history lesson about American railroads, especially to our brothers and sisters "across the pond".
As opposed to railroads on the European continent which were state run, American railroads were strictly a private enterprise. No national treasury to back those boys up in the 19th Century when things got rollin'. American railroad builders had to raise the capital themselves. AND those investors were looking for a return on their investment as soon as possible. Ever hear the old saying "the best is the enemy of the good"? In the United States that meant a good roadbed, a good locomotive, good rolling stock, was a lot better than someting perfect in the future. The roads had to start making money NOW. Improvements could always be made in the future when the money started coming in. Certainly mistakes were made, often with tragic consequences, but it's too easy to critisise from the comfort of an armchair when you're not the one on the spot with stockholders looking over your shoulder.
So, American locomotives had to be built tough and durable. If some efficiency was lost, well, that's something they were willing to live with. Remember, if you build a perfect locomotive you need a perfect road bed to run it on, and in the old days that was usually lacking here in the U.S. By the way, those tough American engines were also a pretty attractive puchase for those building railroads in what we'd call "Third World" countrys, but I'm digressing.
Now if the locomotive builders got it wrong at times, well who's to blame them. No computer assist back in the old days. All the boys at Baldwin, ALCO, or Lima had to go on was their own knowledge of physics, geometry, metallurgy, and prior experience, and if they had a slide rule to help them they were lucky. Sure they got some things wrong, but boy when they got it right! Niagaras, Mohawks, Big Boys, Berkshires, the Norfolk and Westerns mighty Class A's, J's, and Y's! Even the 4-4-0 "American" type was a masterpiece in it's own right. By the way, even the present day diesel builders get it wrong from time to time.
So, remember the American philosphy was "keep the tonnage moving!" Get 'em in, get 'em serviced, get 'em out again. If the engine can't take it you get rid of it and get another. Keep them trains rollin'!
Mind you, I'm not attempting to denigrate any other nations locomotives. British, German, French, all were good machines for the evironments they operated in. All were gems in their own right. Hey, some were so beautiful they LOOKED like pieces of jewelry! Would they have been sucessful here in the U.S? Well, I don't know. I DO know that American designers were aware of what was going on with European steam, but there wasn't much they were willing to borrow. Superheating, the Belpaire boiler, the Walschaerts valve gear, as far as I know that was as far as they went.
So how hard was American steam run? Well, as the song says "I'll run her 'till she leaves the rail, 'cause I'm eight hours late with the southbound mail!"
Gotta go, my eyes are glazing over again! I guess it beats writers cramp!
Thanks for these comments; with respect to the video of UP3985 posted by Hamltnblue, then if you want to know why I find this topic compelling that explains everything perfectly. It has to be high up in my top ten of steam videos on You Tube. Unfortunately those of us with over active left hand sides of the brain try to explore this magic, and try and reduce everything to numbers!
Thanks to Lars Loco for the link to the previous thread, which contains some helpful stuff on feedwater heaters, a topic I’m going to need to get my brain round (they never caught on in the UK, we settled for exhaust steam injectors, which recycled roughly 6% of the heat in the exhaust steam), since this determines how much of the cylinder feed goes up the blastpipe, a very important correction especially at high steam rates. Also, the data on what the Big Boys put out at 14 mph (ca 4200IHP) is the kind of stuff I’m after. Lots of other interesting stuff in the thread too, don’t want to divert onto that!
With respect to NM_Coot’s data, that’s helpful, but not quite enough. Let me tell you what I’d ideally like, what I’d happily settle for, and why I think the point is important.
Between you, you have helped focus my original query into something like ‘How much would Chapelon style exhaust technology have improved the efficiency of latter day US steam locomotives?’ My original Porta quote implies he thought the answer was ‘a lot’. I think the answer is more likely ‘a bit, nothing game changing’. This depends on the rates at which locomotives were habitually worked, not what they could do on the occasions when they were worked flat out for short periods, for it is overall rate of working that determines overall efficiency.
Let me illustrate with some calculations I did some years back on the Milwaukee Road F7s. (Don’t take numbers too literally, my thinking has moved on quite a bit since then, but they are good enough to make the general point). In Jim Scribbins book ‘The Hiawatha Story’ are what I’d really like- detailed logs of the working of the heaviest/fastest US expresses. (My strong suspicion always was, as NM_Coot writes that freight is not the place to look for sustained high power; there are of course mountain grade where locos will have had to be worked flat out at the maximum end of their tonnage range, but a) because the speeds will be low, the cut off will be long, the cylinder efficiency down, and hence the maximum power will be some way below what the locomotive is capable of, as the Big Boy example shows) b) this is probably not typical of what they had to do over the whole trip: the short stretch from Cheyenne to Laramie is different to Laramie to Ogden or North Platte-Cheyenne. I don’t know what US freight speeds on reasonably flat terrain were, but am assuming it was probably at best 50 mph(?)), which will not require too much power - about 4000IHP with 4500 tons).
I have analysed the runs in the book, from Milwaukee to La Crosse. There is a northbound run with 100 plus 9 cars than got to Milwaukee in 73 minutes, 2 minutes inside schedule, having lost 3 minutes as far as Pacific Junction, possibly due to adverse operating conditions. Accelerating from Mayfair, #100 produced about 4000hp for a couple of minutes to get up to 90mph by Glenview. Thereafter, 2500-3500 ihp sufficed to cover the 56.4 miles on to Oakwood in 36 ½ minutes, maximum about 100mph. Coming south there is another run with 100, from Milwaukee to Chicago with 9 cars in about 68 minutes, net of a slowing at West Lake Forest, maximum claimed speed 110mph, well within the schedule and I suspect at the top end of the range of daily working. A full analysis is shown below; times were clearly noted at every milepost. Note that IHP rarely exceeded 3500 and was often less. The very detailed time course provided suggests there was a brief maximum of 107 mph just before Gurnee. The two claimed miles at 110mph are quite impossible- accelerating from 103-110mph over a mile is quite beyond an F7- an artefact of the timing process used!
Engine 100
Tare/gross wt
430/460 tons
Loco wt 395 tons
Total wt 855 tons
Miles
Schedule
Distance/cut off
Est speed
Log speed
Log time
Av speed
Av ihp
Milwaukee
0/25
01:38:03
1.1
Washington St
3
1.1/35
33.45
01:41:29
19.22
824
1.95
Kinnickinnic Draw
2.0/47
44.56
64/64/67/68
01:42:48
38.53
1923
7.55
Lake
9
6.6/65
72.59
75/78/88/92/92
01:48:24
60.13
3369
12.55
Oakwood
7.55/40
94.65
97/97/97
01:51:55
85.02
3505
16.05
Caledonia
9.2/44
95.49
97
01:54:05
97.33
3560
17.25
Tower A68
17
15.9/0
93.44
90/97/100
01:54:51
93.68
2671
19.3
Franksville
16.55/50
98.45
97/97/100/103
01:56:09
94.03
3919
23.5
Sturtevant
19.1/35
100.05
97/103/100/103
01:58:40
100.74
2833
27.65
Somers
22.6/0
102.97
103/103/103/103/103
02:01:09
100.16
2867
32.65
Truesdell
23/34
101.58
02:04:02
103.79
2587
33.65
Ranney
28
31.9/0
101.20
103/100/100/100
02:04:38
101.39
2770
38.1
Russell
32.65/32
99.87
100/100/100/103/100
02:07:19
99.37
2697
42.4
Wadsworth
34
38.6/45
102.45
103/110/106/103
02:09:51
101.86
3350
46.55
Gurnee
39.85/30
103.21
103/100
02:12:14
104.52
3366
48.35
Wilson
41.65/48
100.22
100/103/103/103
02:13:18
101.38
1962
53.1
Rondout
41
45.25/20
103.71
103/103/110/64
02:16:04
102.59
3642
57.05
West Lake Forest
48.2/44
71.96
73/78/88
02:19:14
74.87
-1700
61.1
Deerfield
51.5/40
92.87
88/90/97
02:22:07
84.39
4242
64.15
Northbrook
54.25 brake
95.14
103
02:24:00
97.05
2250
64.8
Tower A20
49
55.15/0
97.57
02:24:25
96.38
4844
64.95
Techny
55.75/60
98.10
92/100/103
02:24:30
97.84
4915
67.95
Glenview
59.05/46
100.89
02:26:17
3608
69.15
Golf
62.6/0
99.97
100/100
02:27:02
96.50
2253
70.9
Morton Grove
64.65/58
98.28
97/97/103/103
02:28:05
100.51
73.65
Edgebrook
66.25/30
101.67
02:29:43
100.80
3720
75.1
Forest glen
69.1/0
68.81
72
02:30:41
89.17
-8437
76.45
Mayfair
70.5/48
51.58
46
02:32:20
49.32
-1908
77.25
Grayland
73.45/0
64.09
59/65
02:33:13
54.59
3848
79.05
Healy
Mostly
50.92
56
02:34:52
65.39
-843
79.9
Tower A5
Brake
48.02
02:35:54
49.04
135
79.95
Pacific Jn
62
and drift
48.65
02:35:58
48.33
2500
82.45
Western Avenue
28.80
02:39:11
46.56
-317
85.45
Chicago
75
0.50
02:47:39
21.26
-118
These and the other logs I have analysed suggest that a continuous output of 3500IHP from an F7 was more than enough to keep time with the standard load. Now there is also a report from the French Baron Vuillet that in 1943, an F7 took 16 cars from Milwaukee to Chicago in 63 minutes, averaging almost exactly 100mph from Oakwood to Edgebrook. This would require about 4200IHP sustained over the whole distance- perhaps a truer measure of what an F7 could do, though, I suspect, rarely required to. It is claimed that an F7 averaged about 120mph with 9 cars on test, and this would also require this level of power.
Now from some source or other, trustworthiness unknown, I have a blastpipe diameter of 7” for the F7. Using this, and making an intelligent guess on F7 superheat, I reckon that 3500IHP would require about 50000lbs/hr to the cylinders, and if all this went up the blastpipe (which it didn’t), this would give a backpressure of about 10psi, probably in truth nearer 8psi.
However, suppose what was actually required on a daily basis was more like 4500IHP. The rapidly rising backpressure means that efficiency would decrease and steam to the cylinders required would balloon to ca 72000lbs/hr, possible from a 96.5 sqft grate with Indiana coal, but seriously high blastpipe pressure. M. Chapelon could then claim he could deliver 10+% in coal savings, which would surely have got the MILW’s attention.
My underlying suspicion is this: everyone likes to quote what a steam class could do; it is reasonable looking at the data already to hand that if they operated at ‘could do’ level then improved front end design would lead to significantly more efficient locomotives; however if you look at ‘daily needed to do’ as the basis, most of the case against US front ends will disappear.
To understand this for other designs needs similar information on what they were actually required to do on a daily basis.
Now, whilst we had tens if not hundreds of folk timing trains all over our country with great precision from about 1900 on, so any train that moved was logged on some occasion, I’m not sure that detailed logs of the kind for 100 given above are common in the US. It would be great if they existed for the heavy high speed workings on the NYC, PRR, and Santa Fe, say, or any other railroad for that matter. If they don’t, just knowing typical loads, and schedules between the stops on say, the steam hauled 20th Century, Broadway and whatever the fastest steam hauled Santa Fe service was would suffice. The data would need to be for fast, straight stretches where it may be reasonably assumed there were no speed restrictions for curvature nor frequent speed restrictions for other causes, and what the speed limit, if any was. Routes that fit the bill would I think be for example Buffalo and Crestline to Chicago, Garden City to La Junta.
In all this, I am making the assumption that the second benefit of Chapelon style exhausts, namely providing more air to the fire, hence more reliable steaming was not really an issue in US practice, the need to operate reliably in harsh environments having already been given priority over other considerations. (It was in this country when people tried to draught increasingly large boilers with single blastpipes, in an environment where the distance between the top of the blastpipe and the top of the chimney was severely restricted by the loading gauge).
Thanks again for all your inputs.
Whenever I see a thread like this I usually do a quick google for the challenger freight video.
It puts it all into perspective. The fact that you have so much black smoke shows that there is a lot of unburned fuel/inefficiencies, but she is a beaut.
http://www.youtube.com/watch?v=XhgHrDbN4EU
Springfield PA
Volumes have been written on this! North American freight service, with a few exceptions, is NOT the place to look for continuous performance capabilities. Here are a few passenger examples. ATSF 4-8-4 classes worked 1791 miles, Kansas City to Los Angeles, 12 crews, Raton, Glorieta, continental, Cajon passes. Average engine worked 18 to 20 thousand miles per month. (2900 class 4-8-4 on freight service via Belen cutoff, only averaged 8000 to 9000). NYC 4-8-4 S-1b, in field tests, consistently produced 6600ihp at speeds of 75 to 85 mph and were capable of accelerating a 1000 ton train to 75 mph in 19,400 feet. They averaged over 20,000 miles per month. Kiefer was never able to optimze them because of dieselization taking place and bumping them from passenger service. SP GS3 regularly handled 925 ton trains up compensated 1.0% grades at sustained speeds in excess of 55 mph, which I believe is equivalent to about 5500ihp. NYC J-3a Hudsons worked the 20th Century, usually about 1000 tons, from Harmon to Chicago, 925 miles, in 16 hours with 7 intermediate stops. In practice the J-3a boiler, in line with most modern boilers, could produce more steam than the engine could consume. What made the J-3a so much more powerful and efficient was the care taken with the dimensions of live and exhaust steam passages and the high degree of superheat (about 750 degree F). J-3a's on average worked about 12,000 miles per month.
Not sure if this what you wanted. In any case it is just a sample.
"I’d like to understand; exactly how hard and for how long were US designs worked?"
Have a look here...
http://cs.trains.com/TRCCS/forums/t/161948.aspx?PageIndex=1
Great discussion, turning into a full time job to keep up!
I am in sympathy which most of what is written.
I do agree with CJ Hegewisch (12/04/11) that the perspective of operators was different to that of the gurus running testing stations. Our test stations were interested in cost savings from efficiency, because that is what they were designed to measure. Their work led to a few, trivial improvements to efficiency, all of which were ignored by the operators and not implemented. The only things they did that were accepted were to narrow the blastpipes of some designs, which made efficiency a bit worse, but improved the reliability of steaming. The basic philosophy (not well executed) of latter day UK designs was low maintenance. Efficiency was pretty good, but draughting stone age, because of the predilections of those in charge of design. Clearly if any feature which improves efficiency adds to maintenance costs, that needs to be factored in.
So, as Juniatha so eloquently puts it (13/04/11) efficiency is one thing, operating a railway another, and operational needs win out over efficiency if they are in conflict.
(Great memories of Hegewisch by the way; I remember well trying to walk from the station to State Line Crossing in a foot of snow, 13 degrees, early 1974, Train Watcher’s Guide to Chicago in hand. Crazy. Have a slide of an old CSSB unit there the following summer).
With respect to NM Coot 09/04/11, 10/4/11 and Juniatha 10/4/11 and 13/04/11 on the subject of Chapelon, he surely was one of the all time greats, and his designs demonstrated that. Juniatha’s stories of what he achieved are well known and oft repeated over here.
Secondly, it is in my experience nearly always the case that we often selectively cherry pick data to support our cases; A number of our 100ton (110US ton) Pacifics have achieved 3000IHP, one approached 3500IHP. Did they get anywhere near this in daily service? No. Could this have been expected? No, certainly not with handfiring, only at great inefficiency with Mechanical stoking. So, you will find me less interested in extreme claims, more interested in what operational practice and needs actually were, for this will determine how much benefit could be achieved e.g. by better draughting.
Whilst it is true that going faster and heavier, which is what efficiency allows you to do ought to be a universal good, experience over here says that speed and train length were not benefits always avidly sought. Those operators again.
Thirdly, with respect to his designs, the improvements obtained were simply by applying known best practice- three simple tenets, high superheat, low back pressure, high expansion ratio- nothing magic. (And, in the firing department, he was helped by Chauffeur Marty, who it seems was able to shovel approaching 10000lbs/hr of coal, for short periods at least. Over here, our guys seem to peak out at about 4500lbs/hr. Wimps. (I think I’m good for about 200lbs/hr)). High superheat was found in best US designs. Expansion ratio was sometimes poorer than it might have been because of the use of two cylinders only in many designs, which means that to generate high powers, particularly at low speeds you need to use longer, less efficient cut offs. However, I think it would be wrong to criticise this; I prefer to believe that the engineers had done their sums and in net terms, the simplicity of two cylinder designs far outweighed the benefits of going to more complex three and four cylinder arrangements, which could have operated more efficiently, except when they had no choice. So that leaves a draughting system with lower back pressure as the potential easy win.
As I said right at the beginning, it does seem to me that with 20:20 hindsight, it could well be that some US designs were suboptimal in draughting; this seems to me where Chapelon might have helped. I am interested in finding out if this is true, and whether under the operating conditions which were actually required, improvements would have made much difference. I suspect not, certainly not to a degree that would have saved the species for any length of time. There is no doubt, as Juniatha writes that with later European practice, e.g Chapelon, you can break the mould of the single blastpipe, where you do generally trade efficiency for steaming, and get the highly desirable combination of better draught and reduced back pressure. This would appear to be a low maintenance efficiency improvement that could have brought about some economy, potentially an easy win. How much is what I would like to establish.
Over here the high ups in British Railways were so anti fancy foreign practices that even the Giesl was killed off. In fact, this particular decision was not too unreasonable, since it turned out that the class on which it was tested were not often steamed at rates where its benefits would have kicked in. So the real issue was that the class was not being used to their potential. The Giesl could have been applied to good effect on other designs.
One of our railways, whose CME was a good friend of Chapelon, introduced Kylchap to this country; after his death, and following nationalisation and the expiry of the patents, it was widely adopted on their passenger designs, but only in the face of strong opposition from the top. Even as late as 1960 there is a vitriolic letter from a high up slating the use of such equipment, even though it had patently transformed the classes concerned. So I think an additional point is that, over here at least, design was determined as much by personal opinions and rivalries as rational considerations.
I agree with Juniatha’s critique that what I am interested in doing is looking back with 20:20 hindsight; I am not trying to resurrect the species, which is dead.
On Juniatha’s original input, point by point
1. My original query was exactly to ask what the backpressures experienced actually were. I am making no assumptions as to what they are. I agree that measurement is a problem. The model I use uses the standard equation for flow through a plain orifice, with the possibility of varying the discharge coefficient. This fits the measured values of blastpipe pressure with a discharge coefficient of 0.99 in nearly all UK tests. The exceptions are a) some higher values, when orifice plates, which I believe were a US invention were placed over the blastpipe nozzle. These reduce the discharge coefficient to 0.85-0.90. i.e. pressure is higher than it would be at the same orifice area without these devices. (Incidentally, although our blastpipe diameters were specified to the nearest 1/8”, and adjustments of that magnitude made, in service there could be deposits of up to ¾” of carbon below the lip- so clearly these locomotives would be very well draughted! b) lower values. These occurred in three cases. In two cases this was due to the pressures being measured at the base of the blastpipe stand, and there being very slight taper in the blastpipe. Allowing for this, the predictions are spot on. The other exception was the Lemaitre set up, where pressure measured was a bit below the theory. I have not investigated the reasons for this.
2. I was specifically vague about the Niagara blastpipe pressure at high outputs, because there are lots of unknown factors. Precisely what it is, is part of data that I am looking for, and precisely what I think it should be will fall out when I have other design details. My programme gets the pressure for the Q2 more or less spot on. It says the T1 had a free nozzle area equivalent to ca 8” diameter- this must be a matter of record, so this will say how good the programme is for this type. The T1 was 29 psi at 100000lbs/hr. The Niagara had a diameter of 7.5”, so by my estimation it would be higher, and the relationship is not linear. And the higher the backpressure, the more steam you need to generate a given power, so at a given target power you lose twice. So I don’t think my approaching 40 psi is too far wrong and I’ll wager it’s not too far out. We’ll see. The programme says at 52500lbs/hr the Niagara nozzle was behaving like 7.2”. This could be a measurement problem, a design feature, a build up issue, or a small inadequacy in the programme.
3. The H8 quote is from a very respected British Engineering Journal, quoting in great detail what I suppose are primary sources; some of the original graphs are included. Blastpipe pressures are given for both front and back engines, the former being somewhat lower than the latter. The 13psi is a fair average of back and front on the highest rate test, where the evaporation is quoted as 100800lbs/hr, the tank feed 91300lbs/hr. I understand the question of where the measurement was made is important (See above), and this may lead to misleading values.
4. Firstly I should say that in UK testing, the difference between exhaust port pressure and blastpipe pressure was negligible, especially considering the difficulties in measuring both temperatures and pressures of fast moving steam. On your comments about the Gresley set up, you make assertions, but provide no facts. The facts I know of are these a) There are plenty of UK tests on engines with similar lap to the original A1 that show no deficiency in thermodynamic efficiency. b) tests on the original short lap A1 showed the steam chest pressure was typically about 110psi. So, to develop any kind of power, engines of this size had to be worked in 50+% cut off at speed whereas the later designs with longer travel valves were worked at full pressure in ca 30% cut off. This makes a big difference to economy. c) one of the high preists responsible for perpetrating the original efficiency story recanted once he’d figured out what the computer programme was doing and saying d) At around the time the valve gear design was changed the superheat was also increased, then boiler pressure, which is why the A3 and A4 were more efficient still. I agree that the performance of the original Castle it was compared to with was far from inspirational. The claimed 2.83 lbs coal/dbhp for that design is quite unrealistic, as later tests showed. I think I’ve figured out how the 2.83lbs claim was made; a little not unreasonable leger de main was involved to generate a great publicity claim. Like all claims, once you’ve made it, it’s very hard to deny it. (The reason the Castle got to Plymouth faster was basically that the own line driver ran through speed restrictions over the limit. The A1 was actually developing more power going uphill, but because it weighed 30 tons more it didn’t go much faster). I’m not claiming that high MEPS generated in long cut off are not a bad thing- of course they are; but as noted above they are a consequence of sticking with 2 cylinder designs. My point’s a bit more subtle, but I’ll spare everyone that.
5. I agree I’m not taking pumping efficiency into account; I’m rather assuming that everyone understands that with more sophisticated exhausts you can, as noted above get a highly desirable reduction in back pressure whilst actually improving pumping, so, reducing back pressure looks like a bit of a no brainer now.
6. More than happy to look at the differences between the German types if you have the relevant dimensions.
7. Well, I suppose that’s another thing I’d like to understand; exactly how hard and for how long were US designs worked? I could ‘build’ the line from say North Platte- Cheyenne- Laramie, and if people have passenger and freight schedules, and tonnage ranges, I could work out how much power was actually needed on this long uphill stretch. Experience in this country says that whilst the draughting usually allowed 800-900lbs/sqft/hr evaporation, 1050lbs/sqft/hr on occasion, much over 600-700lbs/sqft/hr was rare. Maybe it was different in the US. That’s what I’d like to find out. But even the 20th Century only averaged 60 mph, and that on a Water Level Route.
Nor is it topical.
daveklepper Just for the record, N&W diesilization started with the Harrisberg Lline and its 4-8-0's....
Just for the record, N&W diesilization started with the Harrisberg Lline and its 4-8-0's....
No it did not!
.
Last operating steam on NKP was an 0-8-0 at Calumet Yard.
Just for the record, N&W diesilization started with the Harrisberg Lline and its 4-8-0's. Plus, of course, replacement of J's by run-through Southern Ry E-Units Monroe - Bristol.
blue streak 1 J: I believe that N&W retired their big steram last Switchers were first to go in my area.. I think that a "Y" was the last to drop its fires under regular service. Anyone more info?
J: I believe that N&W retired their big steram last Switchers were first to go in my area.. I think that a "Y" was the last to drop its fires under regular service. Anyone more info?
An S1 or S1a was the last to drop its fire. Sorry, don't have the exact engine # at my desk, however, it was written up in "Classic Trains" mag a year or two ago.
Well , I see what you mean . I agree with you as regards virtues of American Super Power steam . Yet , may I kindly invite you to read my paragraph on why the 141.R out-lasted the 141.P on the SNCF ? In a nutshell :as far as decisions were taken upon technical reasoning – there were numerous other considerations playing a part in which steam on which lines of which sheds etc had to go first – it was not regarding matters thermal efficiency in any which way , it was because of lower maintenance of the newer and more numerous , standard engine against the older and more traditional mechanical design in view of lesser assignments of steam where highest specific outputs were no longer needed . The Achilles’ heel of the 141.P was their out-of-date mechanical design inherited from PLM predecessor 141.D . However , composite plate frames and plain bearings of traditional design for instance were neither stringently correlated with nor sequel , effect or outcome of high (or low) thermodynamic efficiency . As a matter of fact , the mentioned Chapelon engines all were rebuilds of old engines inevitably inheriting old fashioned concepts of construction , mind it ! In contrast , the engines he had planned were to be of the same standards of mechanical design as was then practiced in the States – so where , then , was the difference upon which to state those engines would have been more costly to maintain ? Or in simple words : higher thermodynamic efficiency in cylinder work did not as such necessarily demand any extra expenses or complications in mechanical construction and therefore didn't have to change maintenance costs either . A steam loco of higher thermodynamic efficiency wasn't costlier to maintain just because of its high efficiency . Or in other words : you can't save on maintenance costs by sending consumption soaring . And further : costs of running coal traffic for locomotive coal were one of the issues that stood against steam ! The higher fuel consumption for a given work , the more fuel needed , the more traffic for loco fuel , the more costs for loco fuel and for bigger tenders (which added up to locomotive total mass and reduced edbhp for same ihp [2] ) and so on . Efficiency did matter in the steam / diesel competition years . [1] Coming back on the 242.A.1 – this also was a rebuilt engine . Still the 241.P based on the PLM 241.C.1 prototype was chosen for building a final series . This design concept was inferior both thermodynamically and mechanically at the same time , suffering for instance from overheating crank axle bearings and crank axle crackings and thus clearly was second choice in comparison to the 242.A.1 having a much sturdier single crank axle and no problems with it . [2] So , where was the advantage gained for maintenance by sacrifice of excellent efficiency and performance for second choice ? In fact , in determining decisions as to which designs were chosen there were other than technical reasons as I had already hinted with the pathway of the Berkshire / Mikado project . The American Super Power types certainly were magnificent engines for their time – no doubt about that . However , to propose they could not have been improved and improved upon would mean to deny technical progress . Let's not forget more than half a century has passed since these things were an issue . So , we don't need to get too deeply involved in all that , I don't think there is too much in discussing matters all over again . We will never find an undisputed lasting truth that would apply in general – too widely differing were the circumstances and the influences that intervened and played a role . Who was to decide what was right and what was wrong in individual decisions ?
Well , I see what you mean . I agree with you as regards virtues of American Super Power steam . Yet , may I kindly invite you to read my paragraph on why the 141.R out-lasted the 141.P on the SNCF ? In a nutshell :as far as decisions were taken upon technical reasoning – there were numerous other considerations playing a part in which steam on which lines of which sheds etc had to go first – it was not regarding matters thermal efficiency in any which way , it was because of lower maintenance of the newer and more numerous , standard engine against the older and more traditional mechanical design in view of lesser assignments of steam where highest specific outputs were no longer needed .
The Achilles’ heel of the 141.P was their out-of-date mechanical design inherited from PLM predecessor 141.D . However , composite plate frames and plain bearings of traditional design for instance were neither stringently correlated with nor sequel , effect or outcome of high (or low) thermodynamic efficiency . As a matter of fact , the mentioned Chapelon engines all were rebuilds of old engines inevitably inheriting old fashioned concepts of construction , mind it ! In contrast , the engines he had planned were to be of the same standards of mechanical design as was then practiced in the States – so where , then , was the difference upon which to state those engines would have been more costly to maintain ? Or in simple words : higher thermodynamic efficiency in cylinder work did not as such necessarily demand any extra expenses or complications in mechanical construction and therefore didn't have to change maintenance costs either . A steam loco of higher thermodynamic efficiency wasn't costlier to maintain just because of its high efficiency . Or in other words : you can't save on maintenance costs by sending consumption soaring . And further : costs of running coal traffic for locomotive coal were one of the issues that stood against steam ! The higher fuel consumption for a given work , the more fuel needed , the more traffic for loco fuel , the more costs for loco fuel and for bigger tenders (which added up to locomotive total mass and reduced edbhp for same ihp [2] ) and so on . Efficiency did matter in the steam / diesel competition years . [1]
Coming back on the 242.A.1 – this also was a rebuilt engine . Still the 241.P based on the PLM 241.C.1 prototype was chosen for building a final series . This design concept was inferior both thermodynamically and mechanically at the same time , suffering for instance from overheating crank axle bearings and crank axle crackings and thus clearly was second choice in comparison to the 242.A.1 having a much sturdier single crank axle and no problems with it . [2] So , where was the advantage gained for maintenance by sacrifice of excellent efficiency and performance for second choice ? In fact , in determining decisions as to which designs were chosen there were other than technical reasons as I had already hinted with the pathway of the Berkshire / Mikado project .
The American Super Power types certainly were magnificent engines for their time – no doubt about that . However , to propose they could not have been improved and improved upon would mean to deny technical progress . Let's not forget more than half a century has passed since these things were an issue . So , we don't need to get too deeply involved in all that , I don't think there is too much in discussing matters all over again . We will never find an undisputed lasting truth that would apply in general – too widely differing were the circumstances and the influences that intervened and played a role . Who was to decide what was right and what was wrong in individual decisions ?
There is romance in steam's lost case and we like to ponder it and contemplate what could have been – that's all . We should never forget about our hindsight perspective . I've offered my view – no insult intended . That said , I step out of this thread . Don't get at loggerheads about this matter , folks – the locomotives never come back anyways . [1] Regards Juniatha
There is romance in steam's lost case and we like to ponder it and contemplate what could have been – that's all . We should never forget about our hindsight perspective .
I've offered my view – no insult intended . That said , I step out of this thread . Don't get at loggerheads about this matter , folks – the locomotives never come back anyways . [1]
Addenda :
In view of the latest post commenting on comparison of SNCF Mikado types 141.P and 141.R I feel obliged to disclaim what seems to be a miss-interpretation of my comment written earlier on this matter . Inviting to re-read my text , I pointed out reasons why SNCF concentrated dwindled remnants of steam loco park on mainly one loco type , the 141.R , were largely disconnected from engine-technical considerations . Quite certainly , the fact a small number of 141.R engines outlasted practically all other types of steam on the SNCF was caused by vastly eased demands of engine performance with steam traction at that stage during the transition having dramatically lost mainline assignments – all first class train running handed over to electric traction and much of the more important secondary train running done by diesel traction , 141.P high performance standards were no longer needed . To secure remaining few secondary and lower ranging traction services , SNCF therefore preferred to keep the 141.R Mikado because of universal presence on the system , younger age , more sturdy construction and simplicity in design , parts being readily available .
No conclusion can be drawn on that in trying to find an answer to the age-old steam design issue ‘two cylinder simple expansion against four cylinder compound’ . There never was anything like the colloquially alleged ‘mass engine competition test’ carried out on the SNCF and thus , logically , there could be no winner. The question which of these two series of Mikados , 141.P or 141.R , was ‘more efficient’ could not be answered in such a widely undefined , general way – simply because engine characteristics vastly differed . Looking at heavy express run assignments formerly steam handled , the ‘P’ could clearly outperform the ‘R’ ; the ‘P’ could handle demanding runs the ‘R’ lacked both speed and power for – that left the ‘P’ without alternative in this class of train running . At the other end , in low ranking secondary train traction where performance was of little or no importance , clearly the sophistication of a 141.P was not needed . Yet , steam depots in the western part of France continued with their 141.P engines in such services as long as they kept steam running , never asking any 141.R for replacement . Both ‘P’ and ‘R’ stoker fired , the ‘R’s coal consumption was vastly higher at both ends of performance range – at the upper end because of degraded combustion efficiency and lesser thermo-dynamic efficiency and at the lower end because of grate too large for this sort of service . Yet , the ‘R’ was better digesting indifferent coal qualities because of larger grate and sharp draught . In average freight traffic , there was an advantage with the ‘R’ that grew with increasing load and decreasing speed (until getting at a disadvantage against the 150.X , German 44 class , in what was a good working range of that three cylinder Decapod) – yet this was not so much a question of two against four cylinders as it was a profit gained by sturdier construction with cast steel locomotive bed, Franklin automatic wedges , roller bearing on axles , boxpok drive wheels and a very good boiler design for high steaming rate . Beyond the obvious difference in cylinders , ‘P’ and ‘R’ differed drastically in mechanical design and this blurred any practical comparisons of maintenance costs as to seeking an answer to the old ‘two cylinder SE against four cylinder C’ issue . So , in actual service at different regions and in various qualifications of service there could be all sorts of answers to this issue – even on the same railway system ! I think this explains why there never was and never could be a definite answer found or agreed upon in the perennial ‘two cylinder SE against four cylinder C’ issue .
Edit (lighter shades of color) :
sentences added [1]
sentences clarified plus two missing words added [2]
Addenda inserted on issue two cylinder simple against four cylinder compound [3]
It's time to throw a bomb into this discussion. Chapelon's 242.A.1 may have pointed the way to improved efficiencies and performance, but that won't mean too much if such locomotives spend an inordinate amount of time being maintained. French steam was alloted a lot more shop time than would be tolerated by a North American CMO, so the fuel/maintenance trade-off needs to be considered in any discussion of efficiency. Vernon L. Smith, longtime Superintendent of Motive Power of the Belt Railway of Chicago, opined that North American steam locomotives were the best at what a locomotive should do by producing more ton-miles and passenger-miles with less down time than anywhere else.
Hello All
It wasn't my intention to divert the thread. It was my intention to indicate that you can't consider either predicted or actual performance without taking Chapelon's work into account. North American practice has always differed significantly from the rest of the world. The rest adopted some of the best NA practices, but it seems to me the converse is not true. I did try to indicate that what Chapelon achieved with 242.A.1 was simply a starting point for what could be achieved. Other engineers offered changes with considerable potential improvements. For example, Bulleid's concept for oil-bath valve gear was a massive conceptual leap but was let down by very poor detail work. Having said all this, the classic Stephenson steam locomotive could never compete with either electric or diesel-electric performance or operational benefits. But it is interesting to speculate on what could have been...
Three men, two 4-8-4 and one valiant Mikado
Ok, NM Coot, you put attention on André Chapelon – and rightly so: no discussion on steam locomotive efficiency could be complete without due consideration of his work. So far, I for one had kept my comments strictly limited to the initial topic of this thread, more precisely: the question in mind if locomotive performance and efficiency can be ‘predicted’ (questionable wording in view of ol..erh, classic! engines) respectively, can it be assessed closely enough and with tolerably correct correlations between various types of engines? So far, I didn’t want to extend discussion onto ‘how to optimize classic steam loco efficiency?’
In my remarks on draughting I was already hinting it would actually take more than just widening blast nozzle area if you wanted to lower back pressure at cylinders while to ensure gas pumping capacity always remains slightly above demand of complete combustion (i e λ ≥ 1.0 up to nominal boiler steaming rate, wherein ‘nominal’ is somewhat relaxed from maximum (like nom ≈ 80 to 85 % of max) if ‘maximum’ is correctly defined by grate limit, not by premature front end limit or by secondary considerations).
The Kylchap, often used in double arrangement on larger European engines, for a long time featured the highest pumping efficiency with neither Bulleid, nor Ister, nor Kiesel, nor Jabelman multi nozzles types coming close in performance and extent of efficient working range. In post WW-II years, an at least equally high if not superior efficiency was claimed for the Giesl ejector with Adolf Giesl-Gieslingen himself pointing out simplicity of his device, its ease of installation and (claimed) self-adjusting and the advantage gained by slimline shape of improved access to smoke box tube plate for cleaning and maintenance. A direct comparison of performance between Giesl ejector and double Kylchap on two engines of one suitable class of locomotive should have been interesting but I have no knowledge of it ever having been done; testing a double Kylchap 556 against a Giesl 52-80 light Decapod would have come closest – likely however, no comparative tests have been made, or I haven’t heard of it, since the classes were of different, if neighbouring, state railways: DR (Deutsche Reichsbahn) and ČSD (Československé Státní Dráhy). Drawing conclusions from what I know of relative performances of these locomotives it would appear it’s a close cut with the Kylchap 556 almost as much heavier as superior in maximum power output at speeds above some 40 mph – yet in absence of established data it is left to conjecture how much of that must be attributed to somewhat larger if closely similar boiler (equipped with stoker firing) and more efficient cylinders of slightly smaller relative volume (556 class b p of 256 psi against 228 psi was only 95 % of pressure for same t e with 21.7” x 26” as compared to 23.6” x 26” cylinders and identical 55” wheel diameter, disregarding difference in adhesion mass).
Later, Livio Dante Porta evolved the Kylchap into the Kylpor and from that derived the Lempor type, involving a Kordina and vortex or swirl flow of steam from blast nozzles. As by diagrams put up by David Wardale with these types of draughting devices even higher pumping efficiency can be reached. However, implications of these techniques must be considered with some caution. These highly effective draughting devices may lead to heftier than ample air aspiration through grate and firebed with abortive rather than supportive effects to continuous up-keep of combustion. Basically, these devices were developed with regard to running a thick firebed as used with gas producer combustion, involving a high degree of secondary air induced above the firebed.
Wardale has described intricate teething problems with his rebuilt 19-D 2644 to keep the firebed from being partially lifted and tumbling towards the front of the grate. I could imagine the typical railroad steam testing department having given up after a number of runs to just dump that tricky matter including rebuilt engine(s). While a comparatively steep incline of grate in the 19-D class certainly was not supportive of thick firebed plus high draught operation, Wardale identified several further adversely influencing factors and by scrutinizing sequences of testing / analyzing as well as committing co-workers and crew he was able to sort things out in the end. By reading his account on the testing period I got the idea these late hour advancements while holding a promising potential to increase both performance and efficiency of coal fired steam locomotives have much remained in a prototype stage of development (in spite of a time of regular running on the quite special Rio Turbio) and are far from simple to realize, keeping in mind Wardale’s noting of continuous and close support by Porta. This is not the sort of technique to put into effect hands down by your average preserved engine group. In the end, goals very much were accomplished with 2644 ‘Spooky’ and 3450 ‘Red Devil’ – in view of time then running out fast for steam on the SAR during the very years of rebuilding and test running these engines, they should be considered about as much of a success as could be realized under ruling conditions. While his superiors choose to remain blind for reasons obvious, I can read between the lines how reassuring it must have been to Wardale when the driver of 3450, ascending a bank at exceptional speed yet on shorter than usual cut off, in plain words summed it up by stating “there is a difference!” The engine certainly performed inspiring.
Much could be commented on the admirable 242.A.1 – a rebuild prototype to a proposed new family of types, an engine that stood all road tests, shattering established marks of steam traction top performance in France on any mainline tried and even beating electric traction on two runs Paris – Le Mans, arranged at a time when plans for Chapelon’s proposed family of new steam had been shelved and construction of the 2-10-4 had been cancelled in pretense of an alleged shortage of coal which was more likely imagined by politics than real, transient at any rate and concerning grades of coal used by steel industries rather than the railways. The proposed line of three cylinder compound 2-8-4, 2-10-4, 4-6-4 and 4-8-4 of 6000 ihp nominal output were to have many components in common, design incorporated many modern features of American practice which Chapelon had seen and appreciated during his visit to the United States.
As things turned out, this was not to be. However in the years of high hopes during tune up at Vitry and trial running the 242.A.1 attaining 5500 ihp, the series of test runs on the Ligne Imperiale, the classic mainline of the old PLM (Chemin de Fer Paris Lyon Mediterranée) passed with flying flags with a formidable sustained speed on the long 1 in 100 ascent to Blaisy Bas by the first and only 4-8-4 in France – Europe’s most powerful steam locomotive – marked André Chapelon’s triumphant return to the railway of his beginnings, the railway he had initially started to work for because of his early admiration of trains passing by on the PLM line near his parents house in a town in the vicinity of Lyon. However, leading heads of the engine department had turned down his ideas of technical optimization of existing locomotives and so he had left the PLM and ‘went west’ so to speak, to join the PO (C d F Paris Orleans), later PO-Midi, in the western part of France and covering hilly lines over the Massif Central. There, he found his proposals well received and in 1929 first materializing in Pacific 3566 massively rebuilt to his design. The story of success written by this engine and the following, renumbered 3700 series (later 231.700) is well known. Astonishing as these engines performed, they were topped when Chapelon even more radically transformed a number of smaller 73” wheeled Pacifics into 4-8-0s (4700, later 240.700) followed up by more rebuilts in the early years of SNCF, séries 240.P, shortly before WW-II. It was one of these engines that turned out 4700 ihp at continuous rating (real time steaming at constant water level) marking the highest specific power output per unit of engine mass to that date. It was with this series Chapelon engines started intruding the Ligne Imperiale as the smaller 4-8-0s vastly outperformed the large boilered yet cumbersome PLM Mountain types of various configurations of cylinders and drive axles of 79” and 71” wheels (PLM séries 241.A, .B, .D with prototype 241.C.1 and rebuilt 241.E).
With SNCF André Chapelon having been appointed head of the DEL (Division d’Études de Locomotives – locomotive development division) a new Berkshire type (pardon my expression) four cylinder compound of 5000 ihp was being considered. However with the onset of war, the PLM league had fought off what seemed hard to accept with established thinking and in view of austere times managed to bend the project into a Mikado basically intended to be an SNCF follow-up series of the PLM 141.D. Chapelon was then trusted with the task of turning a tolerably mediocre design into a good one – hopefully without changing all too much. In 1942 this gave birth to the 141.P series, the first lot allocated to high ranking Lyon-Venissieux, mainly for passenger service on the Ligne Imperiale. These were soon followed by more engines until later on early Lemaitre engines were exchanged for Double Kylchap engines and a number of the former got shelved serviceable in nearby yards (while lesser ranking sheds continued with old 141.D). The petite slender yet athletic 285 psi, 65” drivered four cylinder compound Mikado featured a steam flow arrangement reversed from the old 228 psi PLM design with l p cylinders now between frames while below the deflectors compact h p cylinders with large steam passages for live and receiver steam flow were set slightly angled to better comply with the angled position of the inside drives. Mechanically, the well balanced engine unit was designed for 150 km/h (93 mph), in actual service speed was limited to 120 km/h (75 mph). Power output characteristic differed from the original PLM Mikado as a superheated steam engine differed from its saturated predecessor: while output curve of the old 141.D leveled off at some 2000 ihp around speeds below 60 mph starting to fall off towards higher speeds, the 141.P curve kept climbing throughout the tested speed range reaching some 4000 ihp at 75 mph without attaining a maximum level – i e what had been a rather dull mule had been re-designed to become an exceptionally free running sparkling universal performer.
When the PLM mainline was electrified in the early 1950s, many 141.P were sent on a tramp never really ending to the last of their miles. With 318 engines built between 1942 and 1952 by several manufacturers – or ateliers, as railway workshops are sometimes called in France – during the main period of transition from steam to electric and diesel traction on the SNCF, from mid fifties to mid sixties, 141.Ps could be found at the head end of about any sort of train on the SNCF from miscellaneous freight to heavy express as well as any odd jobs on rural country lines while with electrification expanding and first class assignments getting lost to steam, increasing numbers became shelved with a cloudy future on doubtful tracks in the quietness of weeds detached from the busy centers of the roundhouse yards.
American built two cylindered 141.R proved better withstanding reduced maintenance and neglect as in the twilight of steam traction the formerly prevailing system of fixed assignation of engines of major classes to permanent crews had been replaced by common user system. The 141.P’s Achilles’ heel was their PLM design ancestry, mechanically out of date in certain ways from composite plate frame structure to plain bearings throughout or boiler with small firebox. ‘Les R’ as they were tersely called were less demanding in maintenance, yet reliable and thus ‘destinée d’écrire la dernière page de la vapeur sur l’SNCF’ – destined to write steam’s last page on the SNCF.
In his 1952 proposal of future high performance types of steam, André Chapelon was to conjoin the best of American practice in mechanical design with advanced thermodynamic design features – or in presenting his paper at the Arts et Métiers in Paris he offered exactly what was perceived as worst case by those who were already planning for steam’s abdication …
Juniatha
front elevation of planned Chapelon / DEL 152.Q – drawing: SNCF / bearings added plus coloring: Juniatha
In all the discussion on this topic I'm really surprised to so rarely see the name Chapelon mentioned. His designs and performance standards really exceeded everything being discussed in this thread. The ultimate steam locomotive, 242.A.1, was a rebuild of a poor performer and comfirmed Chapelon design principles and showed what could be possible in the future. Both his original designs and his rebuilds of poorly performing engines gave dramatic increases in power with significantly reduced fuel and water consumption. In North America, compounding had a horrific history and would never be accepted. That left 1, free exhaust clearance, 2, soft but effective exhausting, and 3, relatively high super heat to increase steam fluidity. I previously mentioned Chapelon's own book but there have been others that give considerable detail for both test bank and field trials of his locomotives.
Imagine a steam locomotive capable of continous output of 4000 hp at the drawbar at 50 mph on the test bank and 4000 to 4200 hp on service trains. Total weight was 148 tons and adhesive weight was 84 tons. Boiler pressure was 292 psi and a triple klychap exhaust was fitted. The grate was 54 sq ft. Because of the three cylinder configuration the locomotive showed no tendency to slip even at maximum outputs.
if you want dream steam imagine a Chapelon-designed 4-8-4 to the North American loading gauge. (Or better yet, a Chapelon-designed Garratt 4-8-4+4-8-4...) No North American locomotive would even come close either performance or economy.
@ Dreyfusshudson
A few more notes
quotes in >> French Croissants <<
– 1 –
>> Now these backpressures look horrendous to European eyes <<
Umpfhh - erh, do they, really? well, maybe to eyes in 'Old Europe' as Rumbling Rummy would have said ;-) From your previous postings I thought you knew these back pressures since originally you stated that the formulae you work with would allow to predict performance (basically) from just three parameters and back pressure was one of them – although not precisely defined where measured?
– 2 –
>> .. Niagara already a back pressure of 10 psi at 52500lbs/hr, 44 sqins blastpipe, which means that at the quoted rating of 6000IHP the blastpipe pressure would be approaching 40psi! <<
That compares ‘back pressure’ against evaporation on one hand with ‘back pressure’ against cylinder horse power on the other. Back pressure essentially depends on mass rate of steam flow and steam temperature for a given design of blast nozzle. It is not the same for different shapes and designs of draughting blast nozzle(s) of given numerical free cross section tip area at given steam flow! As I remarked on the Porta statement: draughting could have been a lot worse in that poor dimensioning would have produced much higher back pressures with same numerical cross section nozzle tip area or produced much less draught at same given back pressure. To make it plain and simple: Porta’s statement, as I had modestly let on before, was characterized by his eagerness and love of exaggerations – thumps down for anything FGS / thumps up for his SGS. In this context, mind that given parameters were severely constricting design of draughting arrangement with large engines and inevitably it got worse with larger diameter of boiler within a given loading gauge!
Further, an extrapolation of back pressure from a given lower steaming rate to back pressure at an upper steaming rate about double the former, first of all needs to take into account the specific flow coefficient and the pumping efficiency curve of that draughting arrangement – only to remain a vague enough approach with an exponentiated rate of inaccuracy as the quotient higher : lower steaming rate increases. Again, there is a decided difference between round nozzles of plain ordinary type and De Laval type as in the Niagara! Not to mention effects of a – properly designed – Kordina! Without knowledge of the design specs and behavior of the draughting this numerical extrapolation will lead astray. That is probably why your estimate back pressure (40 psi) for the upper steaming rate is way too high. The draughting arrangements of the NYC S-1b vs PRR T-1 (non-regarding the Q-2) were not directly comparable for several reasons and showed characteristic differences in performance. The UP Jabelmann double arrangement of the 4-8-4 (again the Challenger and BB should not be considered in this comparison) was as inferior in some aspects as it was superior in other ways.
– 3 –
>> .. C&O H8 which showed around 13psi at the same steam rate <<
Ooops ?? where did you get that figure from? Without considering very special interfering conditions this would appear an exceptionally low figure, actually unwarranted. In SE Mallet type locomotives there is a long exhaust line from the second drive unit cylinders to the draughting, therefore it would be all the more important to know exactly where this ‘back pressure’ has been measured. Principally, for a number of reasons inherent with basic arrangement of live and exhaust steam conduct in these engines as they were, back pressures on pistons were rather higher – not lower – than in ‘straight’ engines i e of but one drive unit with cylinders under the smoke box. In SE Mallets, what percentage of back pressure on piston would then have been left as pressure at blast nozzle(s) highly depended on the degree of free or not so free flow of exhaust steam to that nozzle(s) and could indeed vary widely.
– 4 –
>> Firstly the % loss of efficiency you get from high back pressure decreases as MEP rises << and
>> US MEPs it seems were higher than elsewhere <<
Why, actually it doesn’t. Again, it’s important to distinguish between back pressure on piston and ‘back pressure’ somewhere in the exhaust line or pressure at blast nozzle!
For an increment of exhaust pressure base line in the cylinder, for example from 10 psi to 20 psi, including consequential higher compression – provided the higher exhaust pressure base line does not lead to over-compression! – the absolute amount of lost energy conversion is directly proportional to the diagram area defined by the two exhaust pressure and compression lines from exhaust lead opening to admission lead opening. Never mix intake and expansion processes with exhaust and compression processes. In the Walschaert’s / Heusinger valve gear characteristics on both sides of the piston are mechanically interconnected but that does not say losses on the exhaust side could be compensated or downsized by beefing up on the intake side.
In fact, evaluating progression of steam pressure on the intake side during admission and expansion has to distinguish between depression by throttling or drawdown by expansion. As I understand, in paragraph three of your reply to me, you postulate valve gear characteristics while not irrelevant are of but small influence. Next, you offer the known example of the Gresley A1 valve gear redesigned for the A3, where, to quote >> well told story .. about how changes to these settings led to a massive improvement ..<< were proven by your program to have been >> wrong << and instead >> the benefit was all about the redesigned gear being able to operate in shorter cut off compared to .. the original design. << Well, now that exactly was the difference the redesign (basically from short lap to long lap) did make and that was what yielded higher thermo-dynamic cylinder efficiency in the A3 and in the A4 further improved along the same line of development, allowing higher outputs in the upper speed range, and higher rpm speeds the A1 did not reach. So, it did make quite a difference. After all, it is known the original A1 proved a slouch, frankly, when in the exchange trials it was compared with the Swindon Castle class four cylinder simple – which by itself certainly was not too inspiring in thermo-dynamic cylinder performance with its valve gear mounted between frames and derived piston valve actuation on outside cylinders (quite the opposite of what would have been mechanically preferable) and with Churchward’s cherished low degree superheating.
However, as you acknowledge, to quote from paragraph four in your reply to me, >> beyond 40 % ‘very big’ differences << we would agree there is a decided loss in cylinder efficiency at c/o ~ 50 % or over as often found in large American loco types at flat out performance. High >> US MEPs << were thus no protection against losses by elevated exhaust back pressure. On the contrary long c/o caused losses by truncating expansion rate. To claim with these high mean cylinder pressures, increased piston back pressure didn’t matter much would mean to suggest when driving a Big Block Chevy of the early years of unleaded gas and catalyser, a badly tuned 4 barrel carb doesn’t make much of a difference in mpg.
– 5 –
>> If one were to set a design goal of a back pressure of no more than 5 psi at some maximum design evaporation rate- let’s say 800lbs/sqft/hr, i.e. before boiler efficiency really plummets, then .. a blastpipe area sqins of 0.887* grate area in sqft- 2.14 would be required, so the target for a Niagara would be 86.5 sqins, for the Q2 about 106sq ins, about double what they had <<
You never seem to account for pumping efficiency of draughting and for boiler characteristics, incl grate and firebed resistances. Principally, with draughting arrangements as they were, already compromised in proportions by inevitable restrictions of loading gauge, just to increase the cross section of blast nozzle(s) would have lead nowhere. This sort of ‘re-draughting’ would simply have repeated faults that had been found in the German Reichsbahn standard types intendedly equipped with Wagner ‘wide’ blast nozzles in an effort to minimize piston back pressure. While with these comparatively light designs in contrast to the massive American engines here considered the desired design aim was reached tolerably well, effect on draughting left something to desire. A suggested ‘Super-Wagnerization’ of American engines would have taken out draughting to a much more serious degree and reliable steaming could no more have been sustained even at moderate steaming rates!
Although steam locomotives weren’t too complex as compared to more modern engines, they represented the type par excellence of inherently interweaving, interrelating design characteristics in most all of their components and functions. That is why no item could be modified without effects on several related elements and according changes needed to be made to the whole chain of functions if any real advantage was to be realized. That was one of the underlying basics of steam locomotive design obviously not fully appreciated, to put it mildly, by many designers, engineers and craftsmen. It may well be so your formula would predict an increased performance by super-enlarging blast nozzle(s) – yet in reality the contrary would have happened.
– 6 –
>> Exhaust design is historically a trade-off between steam raising ability and back pressure; in the UK, and I guess everywhere else, steam raising won out over backpressure i.e. efficiency every single time. <<
If the first part aims at balancing draughting arrangement between lowered piston back pressure and increased draughting over a wide range of steaming rates, then there is something in it, if considerations are strictly limited to single, round plain nozzle, single round chimney type of design – and then it still lacks to take into account boiler design in view of gas flow resistance.
As for the second part: on British Railways steam locomotive design in the 1950s had been decidedly advanced by Riddles in general – famed for his 9F Decapod were ‘F’ actually proved to stand for ‘fast’ as much as it was meant for ‘freight’ – and by S O Ell developing his theory on draughting, enabling a couple of engine classes formerly notorious for bad steaming to shape up with new draughting arrangements fitted. In France, SNCF equipped a large number of engines with more or less sophisticated draughting arrangements and attained – more or less – substantial improvements in reaching both aims at the same time: good steaming and low back pressure. In Germany, with introduction of standardized design of Reichsbahn series, ‘soft’ draughting intentionally preferred low back pressure at the expense of compromising steaming rate ceiling; the rather extreme original proportions were set right on DB by the 1955 re-draughting campaign while on DR improvements to grate, air conduct to grate and ashpan design were introduced with or without Reko type combustion chamber boilers which proved steaming well with Wagner type of draughting because their characteristics were especially designed to suit. This did not mean the same boilers wouldn’t have steamed even better with more advanced draughting arrangements but this never materialized as the Giesl engines were used to burn low grades or indifferent mixtures of coal.
What should have been expected of a reboilered 5035 or 5280 class light Decapod with high efficiency draughting can be concluded from CSD 556 series Decapod of sensibly similar leading dimensions if at 99 t compared to some 90 t of the DR classes. With a very similar if slightly larger boiler these engines equipped with double Kylchap attained 3000 ihp. The Czechoslovakian railways post war two and three cylinder S E engine series by Mares, were of both carefully evolved external lines and highly competent internal design, combining – if given suiting quality of coal, which was rare during most of their time in regular service – high specific power output with comparatively very good cylinder efficiency.
It would be interesting to try your formulae on the 556 versus 5035 or 5280 classes to see how improved features in the former's design numerically nearly identical in leading dimensions will show up.
– 7 –
>> Secondly, these figures, .. are probably at steam rates, which whilst achievable on test may not have been called for in daily service <<
Oh my, on exacting never-relaxing Union Pacific, in fast freight services these engines were usually run flat out at anything test runs had proven them capable of – and then some if conditions were favorable. After all, there were many small adjustments made over the years and incessantly things were being optimized, all in view of improving reliability and capacity of every day revenue service – so, maximum output was very much the daily challenge of performance asked!
I bet-cha!
Edit: my usual odd words and formatting .. plus seven sentences clarified, one intro added - April 9th
Thanks for all your replies. I’m now getting some data, and have reports on the PRR T1, Q2, NYC S1a S1b and S2a, and C&O H8 to hand, and have ATSF and UP references on order. It will take me some time to digest and analyse everything, several weeks, if not months I would have thought. I will report back then for those who are interested.
An initial look see at the data says that on Porta’s critique ‘US exhaust design could not have been worse’ there is a case to answer; at top steam rates of 125000lbs/hr the PRR Q2 had an exhaust pressure of 31 psi, blastpipe area about 58 sqins; the Niagara already a back pressure of 10 psi at 52500lbs/hr, 44 sqins blastpipe, which means that at the quoted rating of 6000IHP the blastpipe pressure would be approaching 40psi! If it were possible to reduce that figure to 5 psi at the same steam rate (and I’m not claiming it is), you would get about another 1200IHP for ‘free’. The T1 had a back pressure of 25-29 psi at 100000lbs/hr which implies a blastpipe area of less than 50 sqins; star of the show so far is the C&O H8 which showed around 13psi at the same steam rate.
Now these backpressures look horrendous to European eyes, but there are mitigating factors. Firstly the % loss of efficiency you get from high back pressure decreases as MEP rises, and US MEPs it seems were higher than elsewhere. Secondly, these figures, apart from the H8, are probably at steam rates, which whilst achievable on test may not have been called for in daily service; remember to maintain 100mph with 1000 US short tons on the level needs only around 5500 IHP from locomotives of this size, exhaust rates less than 80000lbs/hr because of feedwater heater recycling. So an important piece of contextual evidence is what the steam rates required by the most demanding schedules were. If logs of actual running are available one can work back to required evaporation rates. (I’ve just piloted this approach by driving a brand new T1 + 1000tons from Crestline to Lima in 46 minutes, as one of its forebears did; all works well- the locomotive was not flat out. The residents Upper Sandusky were taken by fright as it passed at 111mph at the foot of the downgrade; most didn’t know that something approaching this used to happen several times as day. Lawsuits were threatened.)
If one were to set a design goal of a back pressure of no more than 5 psi at some maximum design evaporation rate- let’s say 800lbs/sqft/hr, i.e. before boiler efficiency really plummets, then a few calculations suggest that a blastpipe area sqins of 0.887* grate area in sqft- 2.14 would be required, so the target for a Niagara would be 86.5 sqins, for the Q2 about 106sq ins, about double what they had; relaxing the criterion to 10 psi gives about 60 and 75 sqins respectively. The Jabelmann exhausts on the UP 4-8-4 and Challenger, should offer a significant improvement on the Niagara and Q2 respectively, so getting these areas is a high priority to understand what US best practice was.
Exhaust design is historically a trade-off between steam raising ability and back pressure; in the UK, and I guess everywhere else, steam raising won out over backpressure i.e. efficiency every single time. This is the real constraint when aiming for lower back pressure.
Dreyfusshudson Dear Lars, Thanks for these comments. Where should I look out for the private message?- this is new to me. I would be very intertested to see the Big Boy data when we have established direct contact
Dear Lars,
Thanks for these comments. Where should I look out for the private message?- this is new to me. I would be very intertested to see the Big Boy data when we have established direct contact
On the index page, where all the trains.com forums are listed, nearer the top you'll see a clickable post from anyone who wishes to start a conversation with you. Similarly, if you click on the bold username in any text message posted here, a page will open and one of the options will be to start a conversation.
Dear NM Coot,
Thanks- understand and largely agree with what you write. Will check out the reference if I can get it. One of the first books I bought on US railroading was Duke and Kistler's photographic treatise on the Santa Fe in California, and I remember being very struck by the fact the 4-8-4s ran the 1765 miles from Kansas City to Los Angeles, over as many obstacles as you would care to mention. Inspired by the photograph of 3777 climbing Cajon Pass with 11 vehicles unassisted, I had a go at producing the power map for a 29xx class, again using guesses for many key parameters (Seems I can’t post this Excel graph)
I then got a crew with a miraculous ability to work at a constant 65000lbs/hr to the cylinders to drive 2926 all the way from Seventh Street, San Bernadino to the Summit, using this map. This is what I got:
2926
650
429
1129
instant ihp
speed mph
total time
av speed
av ihp
av edhp
av dhp
height ft
av grad
San Bernadino
1060
0.8
Seventh Street
1946
22.38
00:02:48
17.19
1422
1339
875
1086
160
Verdemont
4252
28.37
00:15:38
31.32
4260
4035
2604
1706
57
10.4
Devore
4269
29.05
00:21:39
28.92
4266
4008
2576
2001
52
Keenebrook
4205
26.12
00:25:00
28.56
4251
3995
2567
2171
14.1
Old Keenebrook
4159
24.48
00:30:06
24.76
4165
3942
2517
2423
44
18.6
Cajon
4304
30.62
00:40:07
26.94
4219
3980
2550
2903
Alray
4188
26.42
00:51:51
26.08
4196
3963
2536
3487
Silverwood
4383
34.34
00:57:34
32.53
4349
4057
2624
3751
27.45
Summit
4447
39.03
00:58:41
34.90
4412
4092
2658
3794
81
It’s so long ago I can’t even remember where I got the heights and distances from- probably off Google Earth, so it’s all very approximate, but this is another illustration of where I’d like to get to.
In addition to Farrington, The Big Three, you might want to check Brasher, Santa Fe Locomotive Development. There is quite a bit in it but not in easy to use format. Another series that might be useful is the "Loco Profile" series published by Profile Publications. There are issues on NYC Hudsons, Nord pacifics, LNER A4, the American 4-8-4, and many others. The ultimate of course is La Locomotive a Vapeur, by Andre Chapelon. There is an Engilsh edition translated by G W Carpenter. If anyone knows high efficiency steam, it is Chapelon.
I think it was Churchward that said he could improve the efficiency of any engine by 10% simply by painting the funnel blue. His point was that day to day operation depended much more on correct operating procedues than pedantic design detail. If a crew believe they are being closely watched, they'll do things right. And of course what's best depends on whether you want speed, drawbar effort, water economy, simple maintenance, etc.
The objective of the boiler is to produce steam faster than any combination of engine parameters can consume it. The objective of the engine is to produce an effective output using all the steam the boiler can produce. To pass great quantities requires large cross-sections through-out the exhaust path. Higher superheat gives more fluid steam. Numerous designers got the passage part right. Fiddling with just the exhaust outlet is only a part of the answer. High superheat in daily practice really is more a matter of cost and reliability than it is pure engine efficiency.
Having said this, few steam locomotives operated near their maximum for any length of time. The NYC Niagaras were great but for day after day long, high speed runs, you have to give the blue ribbon to the ATSF 4-8-4s.
First of all, let me express my appreciation for all the comments you have all made- a very considerable efforts in some cases. Thank you very much indeed. I will try to produce a single response to all these points, starting with some generalities, then getting to specific individual comments and queries later.
Let me start by referring to Feltonhill’s comment ‘I’m not sure where this discussion is going’ and reiterate my hope as to where it might go. What I have done, as Paul Milenkovic observes it to link together a series of engineering models of the key components of drawbar efficiency; engine efficiency, boiler efficiency, and Locomotive resistance (LR); the first is a fully worked through first principles model, the latter based on detailed estimates of the likely magnitude of the components of LR; the boiler efficiency model is simply an empirical model based on test results.
Now such theories are wonderful things, and the question is rightly posed, but what of the experimental data? As I reported, I have explored the value of these models by comparing their predictions to a wide range of UK data collected at the test plant, about 1500 2 hour long test plant runs, about a dozen different locomotives from different builders, with different design philosophies, plus road testing. The models predict the outcome of these tests remarkably well. In this one ought to bear in mind that there are differences of 5% or more in measured efficiencies of different members of the same class, both ostensibly in ‘as new’ condition (a fact not reported), for reasons I think I understand, and tests also showed that between visits to the shops, efficiency could deteriorate by 10% or more. This needs to be borne I mind when trying to validate the models against data. Also, it was not unknown in our country for enginemen to waste steam through safety valves, draincocks whistles etc., use it for auxiliaries such as braking, worst of all actually using precious steam to keep passengers warm.
I’m interested in finding out how well these models apply to US types. The stimulus was the Porta comment I quoted, which implies that there might have been major scope for improvement; I’m a bit sceptical about this. My hunch is that latter day US designs would all have efficiencies within 10% of each other, and at most the best could have been improved by 20% by the kinds of technologies advocated by Wardale and Porta, half of this from the GPCS firing system, significant, but not game changing I think.
What I therefore need is US data of two kinds.
Firstly, input data for the models. The prime need is for information on blastpipe dimensions and superheat levels, for the work I have done shows these are the things which will most impact engine efficiency. The engine efficiency models also require detailed knowledge of 19 separate dimensions of the motion, to calculate accurately the openings of the ports during the engine cycle; I also need to know lap, lead, exhaust lap, port width, the perimeter of the ports (i.e. % of valve circumference with open ports), clearance volume, operating pressure, cylinder diameter and length. Hoping to keep things simple, I asked just for superheat and blastpipe dimensions. I have access to some of the other data, and can work my way round what I don’t have. This would allow first pass estimates of engine efficiency to be made.
The second need is then for US test data to see how well, or otherwise, these predictions fit with experimental reality; in this it should be noted that as well as taking into account the variability in different class members, in the UK there were in initial plant testing some very basic problems in measurement- fortunately there’s enough raw data available to be able to pinpoint what these problems were. So measurements are not always as reliable as they seem. There is evidence of this in the Altoona test results too.
Even if such test data were not available, I think these first pass calculations would allow one to say how much scope for upward improvement existed, and give an idea of how much the engine efficiency of latter day US designs in fact varied. I would be happy to publish the results of these calculations in this forum on a ‘for what it’s worth’ basis. The quality of the output will depend on the quality and amount of input data I can get- garbage in, garbage out. To improve estimates of boiler efficiency and LR would require more data still, though for first pass estimates, I think what I have is good enough. I am acutely aware that there are aspects of US practice that my models may not incorporate.
One final generality. My interest is in how much efficiency can be improved, as Wardale, Porta and the 5AT group, with whom I am in regular correspondence. However, locomotive efficiency is but one element in the overall financial mix. My understanding is that even in diesel days, US philosophy has been to send out freight trains with just enough power to get over the road without stalling (I exaggerate). Working heavy trains at low speeds leads to low drawbar efficiency, as my illustrative calculations show. Drawbar efficiency could be improved simply by running shorter, faster trains; even then this is not the full story, because if you run faster you need more dhp-hr to cover a given distance, so putting coal consumption up. How this all pans out is a question for accountants; all engineers can do is provide an accurate assessment of efficiency over a range of operating conditions, to allow you to estimate operating costs. Overall it appears to have made sense to run longer, slower trains, thus depressing drawbar efficiency.
Now to specifics.
Firstly, Juniatha, March 26th 09:31. Thanks again for such a detailed set of comments. I’d obviously be very interested in knowing the dimensions of the of the Jabelmann arrangement. It was watching the soft exhaust of 844 and 3985 on You Tube that led to me wonder about Porta’s assertion. With respect to your question about draughting, estimating this is not part of the process. I have done some work, based on ideas of the author of the engine efficiency programme, to attempt to work out what the entrainment ratio (ratio of flue gas to exhaust steam) of a front end is, hence how much coal can be burned. This foundered on the inability to estimate how much the chimney is able to flatten the ‘peaked’ velocity profile of the exhaust leaving the blastpipe. Others are working in detailed computational fluid dynamic modelling to address this. The engine efficiency programme only needs to know the blastpipe pressure, which is determined by the flow and temperature of the steam. Since the programme does not know this at the outset, it does a series of iterations until the flow and temperature match the flow into the cylinders, and efficiency of expansion.
On the steam rates, I’m in no position to challenge your assertion that the UP types can and did exceed 100000lbs/hr, but it would have been very uneconomic (though see comments above about economy). However, I think a distinction needs to be drawn between what could be done, and what was regularly done. If I may illustrate; in this country, we had a class of 10 wheelers, the Scots, that had small grates (31 ¼ sqft) relative to the amount of work they were expected to do. Consequently, the specific evaporations they achieved were higher than any other type. On their best runs they sustained about 1650IHP and produced shorter term bursts of 1750 IHP. This translates to about 700-750lbs/sqft evaporation. Crews claimed they ‘wouldn’t steam’ (I think they meant they burned too much coal!), and one was sent to the test plant where they got 1020lbs/sqft/hr on second grade coal, the highest recorded there. One of the class is still working one of the many steam specials we have. Without wishing to suggest anything other than top class professionalism, crews love to thrash their steeds. In 2009, it delivered a short term effort of 2350IHP needing nearly 1150lbs/sqft/hr, 50% more than they habitually delivered, higher than anything else I can find in UK history, but hardly representative!
Understand your comments about Porta.
Comments on US draughting systems; understood and agreed; it is just how much more might have been possible with better draughting that interests me.
According to my programme, if you had a blastpipe area equivalent to a 10” diameter single pipe you could get 8000ihp from a Niagara, 45% cut off at 100mph, cylinder rate about 117000lbs/hr if superheat were 750oF, quite possible from a 100 sqft grate + feedwater heater, but not I think a regular practicable proposition. Because of the need to have sufficient cylinder volume to develop high power at low speeds, all engines can deliver far more power than their boiler could hope to produce at high speed: the issue is boiler capacity, not engine power.
I’m not saying that lap, lead etc. make no difference, just that the effects are not that large, this on the basis of analysing UK style dimensions; it might pan out differently with US dimensions, though I doubt it. There is a well told story in this country about how changes to these settings led to a massive improvement to the efficiency of the original Doncaster Pacifics. One of the ‘high priests’ of valve gear design in this country, and a Doncaster man too, saw the results of the programme I am referring to just before he died, and, understanding the basis of the calculations, had the grace to admit that the explanation put about for the previous 75 years had to be wrong; the benefit was all about the redesigned gear being able to operate in shorter cut off, compared to the long cut off/ low steam chest pressures allowed by the original design. I have corresponded with Wardale, but not on this subject.
There is indeed no magic in 30% cut off; it is just that if you plot efficiency vs cut off, you find that the decrease is relatively small increasing from 15 and 25%, and then progressively begins to get bigger and bigger for each additional % cut off- 30% is an arbitrary breakpoint between ‘not much difference’ and ‘quite a lot’ of difference, and beyond 40%, ‘very big’ differences. As in Germany, drivers in this country often, for the practical reasons you describe, drove on part throttle, and longer cut off than they needed to, offending the purists who correctly said that the most efficient way was full throttle, short cut off; however, providing cut off was not much above 25%, any efficiency losses were marginal; test plant results confirmed this.
Poppet valves are another big discussion point; maybe this should be shelved.
On the point about it being cylinder back pressure, not blastpipe pressure being important for efficiency, I agree. However, what the programme I am describing demonstrates is that to all intents and purposes the two are the same thing, or at very least, highly correlated.
With respect to your comments on the SSCs of the UP types at 5000 IHP being too optimistic, this is entirely possible! As I indicated in my comments, the input values I chose for superheat and blastpipe area were completely made up; I am looking forward to repeating the calculations with the correct values. The first column of my Table shows that develop 5000IHP at 100 mph is about 30%; the exact figure will depend not only my guesses about superheat and blastpipe dimensions, but also other guesses that I have made with respect to engine and valve dimensions. The allowance for feedwater heater on boiler efficiency is a bit of a botch; I am in essence assuming that the amount of heat from coal needed is about 10% less than that would be needed to produce the flow of cylinder steam in no feedwater heater were present; this is just another assumption that needs tightening up if one were to do more accurate estimates. No account is made of the steam required by auxiliaries. This may sound like a damning series of admissions, but remember my purpose was to illustrate what could be done, if the relevant data were available, not to produce accurate estimates, these will come later if the necessary data is available. Having said that, I do believe the drawbar efficiency estimates are in the right ball park, and the changes I alluded to with respect to speed realistic.
The same goes for you questions about resistance; the underlying formula for resistance I use does separate mechanical and rolling resistance as you suggest. I think the approach used to define mechanical resistance, developed by a friend, is sound, and with data about the mass of parts of the motion, pistons, valves, journal sizes, spring forces etc, could be applied to US designs. I am not confident that my extrapolation from UK locomotives is sound; this is another big project to put right, but I think we are only talking of 10-20% deviations from the figures I quote, so I am I the right ball park at least.
I am wondering, having written this, if it would not be better rather than to try to do a broad brush survey of many types with partial data as inputs, to take one example for which all the details are known, and for which there is test plant data- e.g. the Niagara as mentioned, just to see how applicable the models are to US practice. It all depends on the data which is available.
Feltonhill, March 26th 2.01pm Good to know the T1 and Niagara reports are still around- how can I get hold of them? This would go a long way to answering my query. I might be able to source the Santa Fe book over here. I got to know Phil Atkins well when he was librarian at the NRM, and if he still has the same e:mail address, I’ll see what he can add. (He was the one who hired a pick up truck and went to the Rugby test plant just before it was bulldozed, and saved all the raw data and correspondence associated with UK testing- probably the only surviving raw data we have). I’m not sure what gives the impression that I have used total evaporative heating surface- the surface area referred to is that of the grate.
With respect to combustion chambers, there were some attempts to incorporate these into UK fireboxes, but not I imagine of the size possible in the US; I’m not familiar with the theory but suppose the idea was that the more space you have the greater the time for combustion to complete. Now, Orsat analyses of smokebox gases showed that it was rare on UK designs for combustion to be incomplete, except with respect to unburned particulate coal, so if they did have an effect on combustion it would be to burn some of these small particles; however, given that the extra lifetime of these particles in the firebox would be small, I am not sure there would be much benefit- that’s where the US data would be most helpful.
The Professor who produced the first principles model of the engine began by trying to do a first principles model of the boiler. He abandoned the attempt because he felt he couldn’t do a proper treatment of radiation vs convection- if it wasn’t robust, he didn’t want to know. I think there is an even more fundamental problem that he didn’t address, namely the loss of unburned coal from a lump coal firebed. You can measure it, but as far as I’m aware, no research was ever done to try establish the principles involved (there is some work quoted by Wardale, but I think this is the wrong concept).
Your description of the changes to the Y6 fireboxes is of great interest to me, because there is something about US boiler design I don’t understand. In this country, 19th century engineers believed that a long distance between tubeplates was a good idea to maximise heat capture. When superheaters were added, this meant that whilst the steam was hot near the firebox tubeplate, as the flue gases cooled along the length of the flues, they actually began to cool the steam in the superheater! Bad news. Boilers tubes reached a maximum of 22’ here; it took about 30 years to figure out that if you wanted top superheat, it was inadvisable to go above 17’. And, in fact if you work out how much residual heat there is in the flue gases, the extra amount you extract between 17 and 22’ is pretty small anyway. So, looking at the massive size of some US boilers, how, I ask, did they manage to get top superheat levels?
Your SSC figures for the UP types give about 15.6lbs steam/hp-hr; with the original assumptions I used I get 6000IHP at 70 mph in ca 33% cut off, 89600lbs/hr to the cylinders: at this stage this is remarkable agreement! So we’re not too far apart, though I’m a bit better at 14.9 lbs steam/ihp-hr; from what I read, my blastpipe pressure of 19.7 is way too high, which seems to suggest the superheat is not as good as I suppose.
Lars Loco 26th March 5.34pm
Interested in how the discussion on 100000lbs/hr pans out- see my comments above on ‘normal’ vs ‘maximum possible steam rates’
Lars Loco 26th March 6.26 pm
I’m more than happy to share the engine efficiency programme with you, if you can suggest how I can get it to you; as far as I know Microsoft won’t even let you send zipped .exe files now. It’s only 500KB!
I can also send two published papers from its author, describing its fundamentals; there is also one from me describing how I went about applying it, though I have to say this is rather long! I should also say that I’ll send you a 1 sider on what’s wrong with it- its author would have wished for nothing less than complete honesty, but this is not to undermine its value. It is in my view a complete game changer in our understanding of steam engine efficiency, provided it is handled with care.
Paul Milenkovic March 27th 12.53 am
Thanks for your supportive comments; as outlined above, you are correct about the gist of what I am trying to do. I have Wardale’s book and it is indeed a treasure trove. The point about Wardale however is that he is very proud of the fact that he designed his 5AT using only techniques which were available in the steam age! He has used the programme referred to above to check out his calculations on the 5AT, and found good agreement, though I have the suspicion he prefers the old ways!
Thanks again for all your responses; 9 in 24hours; that’ll teach me to go off enjoying myself doing other things!
To Lars Loco: Oh yeah, I've seen "The General"! A very funny movie, but seeing that old 4-4-0 get wrecked on the burning bridge makes my skin crawl! Oh well, at the time it was OK. and the locomotive was due to be scrapped anyway, but I can't look at it now without thinking of it as the loss of a fine railroad artifact. And doing it to a Class "J" doesn't even bear thinking about! By the way, I misspoke on the "J"s tractive effort. Post- war modifications brought them up to 80,000 pounds of tractive force, making them the most powerful "Northerns" ever built. Oh, and the Norfolk and Western always called them "J"s, there was no way a Virginia road was going to call them "Northerns"! "Old times there are not forgotten...."
Firelock76 To Lars Loco: How much power could an all roller bearing axle locmotive have at the drawbar? Well, have you ever ridden on a jetliner? I rode behind Norfolk and Western's Class "J" 611 several times before the excursion program was stopped, and got that same set-back in the seat when the engineer opened the throttle! Not as much, mind you, but it was certainly there! A "J" could produce 73,300 pounds of tractive effort, can't tell you the horsepower, none of my books have the answer, but my dear God, what a magnificent machine! If General Lee had one available to him in the 1860's, along with "Seven-Elevens" and pick-up trucks, the South would have won The War!
To Lars Loco: How much power could an all roller bearing axle locmotive have at the drawbar? Well, have you ever ridden on a jetliner? I rode behind Norfolk and Western's Class "J" 611 several times before the excursion program was stopped, and got that same set-back in the seat when the engineer opened the throttle! Not as much, mind you, but it was certainly there! A "J" could produce 73,300 pounds of tractive effort, can't tell you the horsepower, none of my books have the answer, but my dear God, what a magnificent machine! If General Lee had one available to him in the 1860's, along with "Seven-Elevens" and pick-up trucks, the South would have won The War!
while reading your last lines...
do you know Charlie Chaplin´s "The General" ? It shows the most dramatic stunts ever made in movie´s history, now imagine he would have had a 611-class engine available ... phew... too though even for Charlie...
Well, all right! Juniatha's back! You go, girl!
I commend you on your work on this topic, and I offer you encouragement in the face of some of the naysayers.
What you offer is an engineering model, that is, a somewhat simplified version of a collection of much more complicated formulas and equations. The purpose of your model is 1) to predict the performance of known steam locomotives, with the view of understanding the rough contribution of various system components -- the grate, the evaporative surface area, aerodynamic resistance in the steam circuit, exhaust back pressure, and so on, and 2) armed with the ability of the model to predict known steam locomotives, to make educated guesses regarding what enhanced or improved steam locomotive designs could have done.
The fact that you get into "the ballpark" of the 5-6 percent thermal efficiency often quoted for 20th century U.S. steam, and that you get into that ballpark based on thermodynamic, fluid dynamic, and empirical relationships tells me that leaving aside the naysayers, that the quirks of individual steam engines are unknowable apart from much more complicated formulas, that you indeed have a good model of steam locomotive thermal efficiency and are to be commended for your efforts.
I am sorry I don't have data or measurements on steam locomotives to contribute. My only suggestions is that the David Wardale 5AT steam locomotive project has a Web site along with links to technical information you could check out (does his prediction of efficiency on the low to mid teens for this proposed design square with your methods?). David Wardale also has this out-of-print book The Red Devil and Other Tales from the Age of Steam -- if one could get a copy somehow, I am told it too has considerable technical detail on steam locomotive efficiency.
I commend you on your efforts and look forward to more results from your model.
@Dreyfusshudson
Thank you all for this interesting post + data, can you tell more about this program? Is it shareable?
Kind Greetings
-lars
Juniatha Er-hm .. if you may pardon my joining your discussion, may I just add a few words? The UP Jabelmann double chimney arrangement included quadruple blast nozzles each. With that draughting both Challenger and FEF-II could and likely did exceed 100,000 lbs/h steaming rates. Regards Juniatha
Do you have the source of "100,000 lbs/" ? Sometimes, I have those "weirdo" thinking, that UP-Engines were pretty underestimated, concerning that...
Looking at the BB-tests in 1943, the engines produced all less than that rate (but have a bigger boiler than a Chally or Northern). All evap.-rates were beyond 100.000 lbs/h. Though, they were not running fast, what is is the missing link here?
feltonhill While we're tossing numbers around, my estimate for a UP FEF-3 (844) would be about 5,500 IHP at 80 mph and 86,000 lbs/hr total evaporation. For the 4-6-6-4 (3985), it would be about 6,000 IHP at 70 mph and 94,000 lbs/hr total evaporation. These are relatively moderate figures, not full-flog readings.
While we're tossing numbers around, my estimate for a UP FEF-3 (844) would be about 5,500 IHP at 80 mph and 86,000 lbs/hr total evaporation. For the 4-6-6-4 (3985), it would be about 6,000 IHP at 70 mph and 94,000 lbs/hr total evaporation. These are relatively moderate figures, not full-flog readings.
talking about efficiency, how much power could modern "all axle roller-bearings based" steam-engines
produce at the tender's draw-bar, including mechanicary and air-resistance, actually?
I have never seen a "conversion table", but starting with your data, maybe ~20%?
So, 4500 DBHP for a FEF-2/3, 4800-5000 DBHP for the Chally? Just as a rule of thumb...
-edit-
Those figures in Mr. Dreyfusshudson's table talking ~10% ? Is not it awesome, from that point of view that modern diesel-electrics do not do better, regarding shaft power to drawbar.
Regarding modern US locomotive data:
Test reports for both the PRR T1 and NYC Niagara survive, but they are not published for sale. I know they exist because I've used them extensively over the years. I believe the Q2 test report also survives, but I've not seen it.A book called Santa Fe's Big 3, by S. Kip Farrington, has considerable operating data on the last classes of ATSF 4-6-4, 4-8-4 and 2-10-4. I have it and it's proved to be a good source of information. This book is usually available on line from several used book vendors.There's a very good book available in the UK called Dropping the Fire by Phil Atkins. Although this is not technically rigorous, it has exceptionally good write-ups on several US locomotive classes (Niagara, PRR T1, N&W Y6's...).Back to the original post. I notice that total evaporative surface is used. I believe this is shortsighted, particularly with US locomotives. After 1930, more locomotives had large combustion chambers, and this altered the proportion of evaporation produced between the direct heating surface and indirect heating surface.Two examples -(1) The PRR M1 4-8-2 of the mid 1920's had PRR's "standard" grate area of 70 sq. ft. ( used for the K4 4-6-2, L1 2-8-2, I1 2-10-0 and later the K5 4-6-2). However, it had a very large combustion chamber which allowed higher firing and evaporative rates. This increase in furnace volume enabled this improvement(2) A much later example can be found in N&W's modification of the interior proportions of its famous Y6 2-8-8-2 compounds in the early 1950s. Initially they had 24 ft flues and a short combustion chamber. This was later modified to 20 ft flues and a 48" extension to the combustion chamber all within the same boiler shell. This produced in increase in the direct HS and a reduction of the indirect HS. The total evap heating surface was reduced, but the overall steaming capacity was increased because most of the evaporation takes place over the direct HS rather than the indirect HS. N&W stated that this modification improved the combustion characteristics of the firebox and the evaporative capacity of the boiler substantially.Juniatha has also brought up valid points regarding the basic construction and configuration of UK locomotives vs those in the US, not to mention differences in operating philosophy. The frame design differences alone have an effect on locomotive resistance.
Based on appearances, the UP 4-8-4's, 4-6-6-4's and 4-8-8-4's all had similar exhaust designs: double stacks with double blast pipes separated into four exhaust nozzles each (a total of eight). It bears some resemblance to later European designs. This was done primarily because UP used low-quality on-line coal, which had a fair quantity of fines. This required an easier exhaust blast, so that the fire wouldn’t be lifted off the grates.
I’m not sure where this discussion will go, but I’m still reading and trying to understand. Exhaust design is not my best subject by a long shot!
Er-hm .. if you may pardon my joining your discussion, may I just add a few words?
The UP Jabelmann double chimney arrangement included quadruple blast nozzles each. With that draughting both Challenger and FEF-II could and likely did exceed 100,000 lbs/h steaming rates. However may I ask how backpressure, draught vacuum and efficiency of this or any other draughting arrangement should be calculated indiscriminatingly of design specs?
In paragraph one of your original posting you correctly quote Porta’s saying – yet this must be seen in context. First, as from what I’ve had an opportunity to lend and read of his papers it appears Porta had a liking for strong expressions and exaggerations to make his point clear (he enlightened in using the word ‘exaggeration’ for some of his own improvements, sometimes meaning to provide a design margin against malign influences, at other points meaning an emphasise on going for sth full stride – very best efficiency, mostly). From what I have read of his, as much as he tended to elevate importance and effect of his own developments, he also tended to degrade earlier technical standards.
There is no denying, draughting arrangements were simple enough in most of US steam loco designs right to the end of the development line. However, they sure could have been a lot worse. In fact, within their very plain and simple layout, more often than not they came tolerably near to best proportions and performed reliably, if at rather high rates of back pressure. In those times this was not considered too harmful in view of relatively long cut offs and consequently high exhaust release pressures with engines run flat out. Of course this consideration was short sighted as high back pressure did in fact reduce mean cylinder pressure – and that did of course adversely affect gross cylinder output and efficiency. Interestingly, the relatively frequent exchange of worn blast nozzles necessary as a consequence of highly abrasive effects of extreme flow rates of steam at rather high pressures was considered forbidding to more sophisticated designs that would have been more costly to build and to install. What was lost however: such designs should have worked on much lower steam pressures, thus at much less aggressive flow rates that again would have made these designs last much longer. Clearly, lack of introduction of more advanced draughting arrangements was due to the early and precipitous end of development of the American steam locomotive.
Some years ago I had written an abstract on thoughts of my own as how indicated power output of the Niagaras could have surpassed 8000 ihp at speeds around 100 mph just by more sophisticated draughting and valve gear, including improved cylinder tribology and accordingly raised live steam temperature, all that on the same amount of net steam heat, i.e. even at slightly eased steaming rate, or in other words exclusively by improved thermo-dynamic cylinder efficiency. Since these improvements would not be reflected in data you consider for calculations by your general formulae, how then could it get correct results for different steam loco types of differing qualities of design? For sure, it should be interesting to read your explaining about >> lap, lead, clearance volume, port width etc << and >> many factors thought to be important in the steam era << .. >> tens of thousands of calculations all say that [in] view of gross efficiency .. these things are of second order << to quote from your first reply paragraph three, lower, and paragraph seven, upper. So far I had believed on the contrary these factors do make a difference – and have further believed Porta, Wardale have believed so, too, when they remarked upon how these things had been unwarrantably slovenly developed?
In paragraph three of your original posting you mention 30 % cut off c/o as something of a dividing point above which length of c/o has a decisive influence on specific steam consumption s.s.c. while below it doesn’t. I can see no theoretical basis for that assumption. The effect of improving s.s.c. by ever shorter cut off levelling out around a certain point depends entirely on imperfections of actual engine characteristics against the ideal steam cycle at the same live steam values. The more these imperfections bear on the actual working characteristics, the sooner a minimum point is reached beyond which further shortening of cut off does no more improve s.s.c.
This happened to be around 30 – 25 % c/o in German standard type two cylinder steam locomotives such as the 01 and 03 class Pacifics and their Decapod counterparts 50 and 52 classes and to an extend also the 42 class. 30 % cut off was generally used for constant speed running, often with laminated throttle and I think drivers were even told not to go below 30 % in view of mechanically hard running – which to me seems arguable at least and was proven wrong by Wardale for his improved SAR 19-D and 26 engines. Clearly it largely depends on mechanical design specs and actual engine condition as much as on valve gear characteristics. Certain poppet valve equipped simple expansion engines could advantageously run on c/o as low as some 5 %!
So, with engines of advanced cylinder tribology and valve gear short c/o does pay while with the more indifferent kind of designs it didn’t. Clearly these contrasting behaviours cannot be banged together in one common formula. Power output graph over speed is not just higher in a more advanced engine per same steaming rate but its line also evolves into a different shape: while the more indifferent type of engine shows an apex of indicated output at a certain speed above which the line starts to fall off like the flight path of a ballistic missile, the advanced engine shows incessantly increasing power output over the entire practical range of speeds, i.e. as a Niagara improved along the lines mentioned above would have shown power output increasing over the entire then legal speed range of 0 – 120 mph.
Back pressure on piston is the one that counts as for cylinder output and cylinder efficiency – not blastpipe pressure (measured at which point?); pressure in blastpipe, or actually: at blast nozzle does neither depend on speed nor cut off but on steaming rate only – logical if you come to think of fluid dynamics and the variable mass / time rate of steam passing nozzle(s) of fixed cross section and flow coefficient. Sorry, I believe your relative data on hourly steam to cylinders for your given 5000 ihp looks somewhat inconsistent, it suggests too low values for s.s.c. Also, the Challenger having a slightly lower s.s.c. at the same speed and ihp relative to the FEF-II would surprise me indeed, taking into account that it is a simple expansion Mallet type while the FEF-II is a straight two cylinder engine.
What c/o did you calculate for the FEF-II at 100 mph / 5000 ihp? In data in column four you allow >> 10 % benefit << for feed water heater – 10 % of exactly what? why don’t you use the same net steam rate to cylinders as in column three? what evaporation data does your boiler efficiency relate to? if all your data so far should be based on net steam to cylinders, then overall drawbar efficiency lacks to account for steam to auxiliaries / if not all of your previous data is based on net steam to cylinders then what data is based on what steaming?
In calculation of locomotive resistance, mechanical resistance (drive), mechanical resistance (rolling) and aerodynamical resistances should be kept separate or you get a blurred picture; I’d like to see the formula that allows for correct data of the total of all of them to be established over speed and then based on data extrapolated from British steam locomotives of, as it was, rather special mechanical characteristics with their composite plate frames, very light rodding and extremely small diameter pins, design prone to flexure under high stress, and their smooth external lines of small loading gauge. Maybe this has lead to rather high locomotive resistance hp losses in the upper speed range and somewhat low values at low speed.
To quote paragraph 5, last sentence of your first reply: >> I was only trying to illustrate the kind of calculations that can be done; what are the most relevant calculations is entirely discussable. << Well, sure there are many ways to put up calculations. Ok, likewise I’m only throwing in a few words of mine to an interesting discussion – no insult intended.
So, to quote once again from your last paragraph in your reply: >> You will need to forgive my enthusiasm! << Well, I guess there will be no problem here, we are all enthusiasts of railroading power one way or another – steam, diesel or eclectic electric.
In videos I have seen of UP 844, there are actually two stacks. Or, so the twin plumes emitting from the 'stack' suggest.
Didn't the late UP locomotives have Le Maitre exhausts?
Some photos of these show multiple jet nozzles arranged in a circle and the large diameter stack suggests that arrangement. That is at least part of the way towards Porta's ideas.
M636C
Thanks for the tip- unfortunately the Altoona reports on Google are the ones I have access to! Reports 1-32 were presented to the British Institute of Mechanical Engineers in two bound volumes ca 1925, personally signed by (I think) the PRR CME. These are now in the Science Museum Library in London, disintegrating rapidly. They are wonderful documents, comprehensive data, crystal clear write ups, confident without being overbearing- a model. In terms of testing the PRR was way ahead of the game at the time. I think the 1914 Bulletin 24 on the effects of superheat is one of the most influential ever written; I'm sure Chapelon digested it and in this country, Gresley immediately copied the variations in superheater set up on his own Pacifics in 1926.
Still hoping someone has access to details of the final flowering of US steam from ca 1926-1949
Google books has tests of locomotives from the Pennsylvania RR done on the test plant. These are very interesting reads. The PRR testing department tested everything. Even the coal is broken down to its BTU and ash content.
Here is a test on an E2s Atlantic loco from 1910.
http://books.google.com/
[Edited to shorten URL]
I pray every day I break even, Cause I can really use the money!
I started with nothing and still have most of it left!
Thanks very much for your response. You are correct to pick me up on the loose use of the word 'Superheat'; I do of course mean inlet steam temperature not the level of superheat in the steam.
The use of constant IHP in my calculations was just for illustrative purposes. If you use constant steam rate as the basis, then you do indeed get IHP values that asymptote as speed increases. There is a possibility that IHP will decrease if you go above a certain speed at constant steam rate, i.e. efficiency decreases despite the fact the cut off is shortening and expansion ratio improving. UK test plant work was normally not above 80mph, so with your diameter +10% rule, you would not expect to see this, and you don't except for some hints thereof on older GW railway designs with low clearance volume, where there is serious overcompression at high speeds.
You can use the computer programme to establish when this would occur; it depends on the operating condition. With late UK designs, e.g. the streamlined A4, (80" drivers) it says that at constant steam rate in the normal working range, efficiency begins to plateau around 100mph and shows positive decreases about 120mph. The point is that where this turnover point is can be now be calculated, if you have the detail of the engine dimensions (Lap, lead, clearance volume, port width etc). This is the next layer of detail that I would need to understand about US types, but tens of thousands of calculations all say that from the point of view of gross efficiency, the point in question, these things are second order.
Now in practice locomotives were worked neither at constant steam rate nor constant power of course, and my argument justifying the 5-6% overall thermal efficiency is indeed a bit simplistic. However, I have done lots of work simulating logs of reported steam performance on various main lines (the only US one on which I have sufficent data is the Milwaukee from Chicago to La Crosse), and you can use equations derived from the outputs above to work out fuel consumption on a trip, and average DHP. It is really this experience of translating predicted performance into practical estimates of drawbar efficiency that leads me to believe that my conclusion about drawbar efficencies is sound- but that's another huge discussion! Incidentally, the MILW calculations surprised me in showing how little power relative to the 96.5 sqft grate was needed to do what the F7s did when cruising at speed; the secrets of fast schedules were barnstorming starts and super slick braking. O to have witnessed them.
I would be very surprised, like you, if a 4-8-4 with a 100sqft grate was ever steamed at 100000lbs/hr, even a Challenger, so I do understand this point; in the UK outputs above about 750lbs/sqft/hr were rare, even though they are surely possible. Bear in mind again I was only trying to illustrate the kind of calculations that can be done; what are the most relevant calculations is entirely discussable.
The calculations are very easy to do- as long as it takes you to input 3-4 numbers and press. In the 1950s, UK test plant results were reduced to a 'power map' which had IHP on the ordinate, speed on the abscissa, and lines of IHP at constant cut off, and constant steam rate, together with the coaling rate at each steam rate- a one stop shop that tells you everything you need to know about power and efficiency of both boiler and engine. If equivalents for US designs exist, I haven't seen them, but it is these power maps that I would like to create.
The calculation of engine efficiency is by no means simplistic, as you seem to fear; it is standard fluid mechanics, with cylinder heat transfer built in, plus steam tables, and requires simultaneous integration of 5 separate partial differentials. The big surprise is the simplicity of the headline conclusions that it leads you to, perhaps the most important of which is that many factors thought to be important in the steam era aren't. And once you start to think about it, it all makes sense. In the UK at least, the data available in steam days was simply not good enough to work out all the subtleties; whilst people knew what might be important, they coudn't be precise about how important; this is what the computational model helps you establish, and it is quite revealing. So I believe there are indeed simple conclusions, but this does not mean the process used to achieve them is simplistic, far from it. You will need to forgive my enthusiasm! Having said this, having compared the computer predictions to about 1500 fully instrumented test plant runs, it is clear there are things that are not so clear, and you do need to be aware of this, but these uncertainties are factored into my enthusiasm.
If only we'd had high speed computing in 1910, things might have been different. Unfortunately, we needed the steam locomotive to do its job, and pass on, to put things in place that allowed us to invent computers!
I believe we need to clarify some terms first. The steam temperatures you mention are not superheat but total steam temperature. Big difference. 250 degrees superheat was considered conservative for modern steam locomotives. Tests of the NYC Niagara and PRR T1 showed superheat temps of well over 300 degrees at their respective boiler pressures (275 and 300 psi).
Another item is the constant IHP as speed increases. Test data indicates that this is not how things work. Most estimating methods compute theoretical IHP as being asymptotic as speed increases. However, actual IHP tends to increase up to about driver diameter plus 10% then starts to decrease.
I'm also suspicious of the very high steaming rates at low speeds for both examples above. Whereas both boilers could probably produce over 100,000 lbs of steam an hour, neither would be particularly efficient at such high rates.
Be very cautious about simplistic ratios when trying to estimate locomotive performance. I've seen a bunch of them and almost always they produce only a meaningless average. Worse, they tend to overlook various components that are important to consider in the overall scheme of things. Steam locos are not easy to predict. I've used generally accepted industry methods and tweaked them a bit for modern construction and some other items. There are abut 20 equations involved, none of which I would call simplistic. Careful with this.... I'm interested but skeptical.
Our community is FREE to join. To participate you must either login or register for an account.