Cossart Valves:
While those look a little like poppet valves, they don't close against seats like poppet valves, but just move up and down in cylindrical liners allowing steam to enter and leave through ports in the liners.
That's why I said they were piston valves "by definition"....
They aren't conventional piston valves but they are piston valves and not poppet valves...
M636C
Hi –
Here I found another – different in the word’s sense – drawing .
What this one shows is a valve that does slide in a sleeve in the block . Now – imho *by definition* this is *not* a piston valve because a piston valve can over-travel port edges , does not move rapidly then stop dead following cam profile and effects variation of cut off by variation of travel and thus degree of over-travel over port edges bot by variation of cam duration . The valve although sliding in a sleeve is thus *not* free to over-travel a port edge and steam control is *not* by timing of passage of port edges *at speed* while moving on what is defined as an elliptical track by the Zeuner diagram but has to stop dead and from there has to be rapidly re-accelerated as given by cam profile . I think this is enough difference to action of a true piston valve so these sliding valves should be termed piston valves . To whom this will not suffice may be added that mechanical forces and stress loads on valve differ from those of a true piston valve where hammering forces on the valve bodies such as caused by rapid acceleration / deceleration to follow cam profile and pre-loaded spring return movement do not exist due to said desmodromic action simple reciprocating travel without intermittent points of abrupt acceleration / stop .
I think that explain why imho it would be misleading to call these valves ‚piston valves‘ in spite of these substantial differences in character of travel , function , stress and maintenance demands .
Regards
Juniatha
Hi Juniatha! LOVE that French patent drawing, especially the title:
"Les piston-valves dans le fonds de cylindres"
Makes me wonder if I should order a Cabernet or a Beaujolais with that.
Wayne
I think the situation here is that the drawing quality is just a tad low in the critical area, the 'bottom' ring and seat area.
[ADDED COMMENT-- Take it from an ex-culprit (cue 'Amazing Grace' as having relevant historical semantics): civility is always an advantage in conducting technical discussions. As Riddles said in connection with Leader "if doctors may differ than so may engineers", but that does not extend to mocking the engineers rather than discussing the engineering. Riddles certainly didn't think much of Leader, but that did not cause him to judge Bulleid's experience or character.]
My thanks to all of you who have contributed to the actual topic of this thread - American steam that might have been in the 1950s and 60s .
Having meanwhile better informed myself about that valve gear , I have seen these valves *did* slide in sleeves - the drawings 10 a and 10 b were as incorrect or misleading in that aspect as was the other drawing , I have to say though .
Still , the valves moved *not* by a continuous to-and-fro path like a piston valve proper but by the abrupt open / closed sequence dictated by cam(s) . This marks a decided difference in how variation of cut-off was accomplished ! Also , that probably is why the French did *not* refer to these valves as 'tiroirs' which would be the French word for piston valves proper but by the foreign term of 'piston valves' which again they did *not * use for proper or conventional piston valves .
Now , the question is if we want to come back once more on the topical question :
What steam locomotive design and classes could have been expected to be built in the years following 1949 for about another decade - conventional reciprocating steam , that is ...?
Hi Firelock
I know that . You see , I could easily have cut that off - yet I didn't .
I'm not trying to pretend the French didn't call it piston valves -
In fact it is kind of revealing the French *did* refer to them by the foreign term of 'piston valves' within a text in French . The French word for proper ( or conventional ) piston valve is tiroir - yet the author did *not* use this term for these valves .
In my posting I explained why *I* preferred to call them poppet valves - that's all . If that is a problem to some , then it is their problem - not mine .
As I have meanwhile informed myself by help of a friend who has the 1938 Chapelon book 'La Locomotive à Vapeur' and has sent me a scan of the text and drawing on Caprotti and Cossart valve gears , the Cossart was a variation of the Caprotti and in fact *did* have valves sliding in a sleeve - so in the light of this information I would refer to them as 'sleeve valves' preferably ( possible alternative : 'sliding valves' )
And - as asked before : do we want to return once more to the original topical question
What designs and classes of conventional reciprocating steam could have been built after 1949 for about another decade ?
With Regards
photo : August 2012 by = J =
photo : October 2012 by = J =
# 307
Hi folks
After some cooling down I have decided to smoothen out the controversy - after all it wasn't but a rather trifling point basically about using terms if you boil down to it - and have re-opened the thread .
Everyone who still has some thought or idea to contribute is welcome to post .
Myself , I wanted to answer on that matter about DD1 electric motors driving coupled axles by jackshafts ;
Because in a jackshaft drive powered by electric motor *all* rods ( including what would be 'main rod' ) are fully *rotating* as but the coupling rods on a reciprocating steam locomotive , all masses can be perfectly balanced by counter weights in wheels , provided this includes proper cross balancing . The problem with steam locomotive main rods is that they are revolving at the hind end and reciprocating along a horizontal , longitudinal line at the front bearing - it is the latter part , approximately 2/5 of the rod*s mass , that causes problems with counter balancing in wheels . The drive action in an old electric with big frame mounted motor and jackshafts is thus fundamentally different from that of a steam locomotive because the motor offers rotating torque right away while the pistons offer but fore-and-aft thrusts that have yet to be converted into a turning moment by transmission via main rod . The consequences on rod and bearing pressures over a turn of wheel of this rotating moment by the electric motor surely will have taken some by surprise who had believed rod drive in such an electric could be dimensioned just as for a steam locomotive of equivalent power output . However : naw-naw-naw , that didn't quite do it and I think must have led to premature wear and slack in rod bearings - actually , in some vids showing such old side-rodder electrics in action rod clanking is clearly audible .
But that's another story ..
# 308
Hello all,
Juniatha, I am still interested in the valve gear you would use to create practical three cylinder North American steam.
Baldwin Locomotive Works:
What I think would have been built after WWII:
A 2-8-8-4
A 4-8-4
A 2-6-6-4
A 2-8-2
A 2-10-4
Anything I am missing?
I see Baldwin mostly sticking with tradition, after the turbine disasters...
NW
# 309
Hi NorthWest
Three cylinder valve gear :
I think the three cylinder engine unit has never been fully developed - most of these engines have been more or less sensitive with their midways drive gear . Often , my impression is that of a two cylinder design with a middle cylinder added ‘booster style’ or almost as an after-thought . The way it can be clearly read from the drawings , presets by customers had been like “ ok , make it a three cylinder engine , but keep extra parts to a minimum “ is appalling in cases .
Dual drive seemed to avoid some peak stress in crank axle as compared to common drive of all cylinders on one axle – yet it created other shortcomings not foreseen because consideration of chassis and engine unit under aspects of dynamic working was just fledging – if at all . In contrast to what it seems when standing besides one of these big , heavy , solid looking locomotives , framework did not provide a rock-like foundation for everything attached to it – the one piece cast steel engine bed came tolerably close to the ideal , yet bar frames with bolted connections between left and right sides and cylinders were something else – not to mention bolted plate frames . Running at speed and with throttle wide open the chassis was vibrating , flexing , and suffering from proper frequency around certain rotational speeds – as far as noted , it was wise to arrange for proper frequency to arise at speeds not run for prolonged periods in actual traffic , however in some cases the second or upper proper frequency speed happened to be unpropitiously close to constant travelling speed – such as in the DB 01-10 class three cylinder Pacifics which for much of their earlier carrier were limited by road speed limit to 110 km/h running (70 mph) while coincided with proper frequency in engines of this class when with frames sound mechanical condition while in worn engines proper frequency used to fall to some 105 km/h . In travelling speed of course exact maximum of proper frequency speed also depended on tire wear i.e. wheel running diameter . Friends who had been cab riding on these engines have told sensible drivers tried to avoid running right on proper frequency speed and allowed the engine to go a few km/h above it , then pulled by throttle to allow speed to fall below it . Worn engines worked loose frame work connections and even cylinder to frame fitting bolts – the latter probably helped by cylinders becoming very hot with these engines of very high superheating when worked hard on heavy trains over the undulating profiles with many 1 in 100 climbs of lines around Kassel , Bebra and Frankfurt . I have copied a sound recording from an old vinyl steam LP of an oil-fired 01-10 on an up 1 in 100 with 12 coaches and keeping 100 km/h flat out – exhaust just one fierce continuous roar like a jet engine at take-off power ; this must have involved a cylinder output substantially above 3000 ihp – not bad for a Pacific of 132,000 lb adhesion mass . When later on the engines were scheduled 135 km/h (84 mph ) on the mostly water level ‘Rennbahn’ Osnabruck –Bremen – Hamburg with up to 15 coaches trains ( ‘race track’ – so called by railroaders for its mostly straight-on routing ) another weak spot showed up in that the inside drive with its shortish main rod driving the leading coupled axle tended to develop hot cross heads which when not detected in time would in cases result in formidable destruction of inside drive when cross head was disintegrating and breaking loose from slide bar . While I had originally quite admired the 01-10 engines for their ‘fast’ looking roller bearing rods , it was to my disappointment I had to learn neither axle bearings nor inside rods had them – arguably where they would have been most needed ! As for the inner crank , putting a roller bearing to a forged one piece crank axle probably was just beyond technology of the 1950s , while distances between axle box guiding pedestals in existing frames just did not allow for larger roller axle bearings to be fitted . Thus the 01-10 came to be equipped with a rather peculiar mixture of plain and roller bearings and this had consequences not foreseen and made these engines sensitive to neglect and indifferent adjusting of axlebox guiding wedges since that affected exact keeping of design value of wheel spacing ( 2300 mm / 90.551 in )
What looked good to me was their independent triple valve gear , each cylinder its own , inside gear driven from a short crank on the center drive axle . Although rather so-so tuning of valve timings had started early – Kassel shed especially must have had a repair gang believing right angle was anything between some 88 to 92 degrees – the engines *could* be tuned to very regular exhaust beats – only it sure did involve a bit of craftsmanship at some progressive yoga positions with the forward mounted slightly slanted middle cylinder . The valve gear’s tendency to loose once established good valve timings clearly was a design flaw common with all DR standard three cylinder engines as was their dual drive and independent valve gear on all cylinders , it stemmed from early to mid 1920s when establishing design features – a time when passenger trains were mostly run at 100 km/h with a maximum of 110 km/h ( 70 mph ) and freight at 60 km/h ( 40 mph ) – that with the heavy Decapod design three cylinder still won over two cylinder type was purely due to comparison of early pre-series engines showing decidedly lower wear and frames trouble in the three cylinder type – thus the series production 44 class came into being and this class set the stage for principal arrangement of the three cylinder engine unit when later on the two cylinder 01 Pacific was to be redesigned into a three cylinder type .
Interestingly , when A Lipertz of ALCO visited Europe to learn about the three cylinder engine , he saw the Prussian P10 with common drive engine and inside drive derived from second crank on left side return crank – yet he adopted the same sort of dual drive the Reichsbahn was to adopt for their three cylinder engines , yet combined that with the Gresley conjugated gear . This type of design was one that clearly cried out “ I’m three cylinder – but only just !” It was characterized by all efforts made to have but a minimum of parts insider frames , disregarding what severe compromises this meant for function and durability of the middle cylinder drive . Lacking the meticulous care of important engines cultivated on the LNER , on the Union Pacific heavy loads and hard work flaws of the conjugated drive soon had to show and were never fully sorted out . Regretfully , ALCO had effectively joined the less-than-ideal German type dual drive with the quite sensitive Gresley conjugated gear – both of which for unquestionable and obvious reasons yet at the expense of endurance of the middle cylinder drive . In UP style heavy freight with lots of cinders raining down on the front of the locomotive when creeping up hill at utmost effort , the front end was necessarily a very dirty place for the Gresley levers to work and thus inevitably wear took its toll early and excessively . To run fast with such an erratically functioning valve gear must have been abortive for the locomotive – theoretical advantages of three cylinder type were turned into their negatives . No wonder UP embraced the Challenger SE Mallet and with its later improved variation appearing decision was made accurate .
I have always wondered what a reversed w/a twelve coupled three cylinder engine complete witch cast steel one piece engine bed could have been had ALCO not given up the three cylinder type but improved it instead . I think a 2-12-4 three cylinder type could have become the apex of the rigid frame freight steam locomotive in America .
All in all , I believe the three cylinder simple expansion engine unit had to have independent valve gear for each cylinder – very well Walschaerts with double intake / double exhaust Willoteaux piston valves , should by all means have uniform cylinder volume in all cylinders ( no uncompensated shortened inside stroke ) and preferably common drive , no drive on front coupled axle , and a very well cross braced framework , exceptionally stiffened against shearing forces .
# 310
Since there are one or two users who have exceptional difficulties in understanding my way of writing and find it impossible to let go about one certain dead end side track , I feel obliged to repeat just *one more* time :
to quote my own words on re-opening >> Everyone who still has some thought or idea to contribute is welcome to post .<<
'To contribute' of course meant the topic of this thread - I will not repeat it here , the obviously uninformed is free to look it up by the headline of this thread and description in my introductory post .
I hope this will now be clear enough to be understood .
Thank you , and sorry , my excuse to all the other readers for having to clean up the same thing twice .
# 311
Thank you for your detailed reply.
So you would use a Walschaerts for each cylinder. Probably the most trouble free arrangement. I also wonder if Holcroft valve gear would have found an application in the United States.
Back to Baldwin, I see them working on a 2-8-4 as well.
# 312
OK, now will whoever is bending over backwards, it seems, to annoy Juniatha just knock it off, for cryin' out loud? I don't know what was said and by whom and quite frankly I don't care, but will some of you out there in railfan land just grow up and get a grip? Please? I'm askin' you nice now, aren't I?
There's a general impression among the non-railfan public that railfans are a bunch of social misfits. Let's not re-enforce it OK? You never know who's looking in, and we don't want to scare off any new posters, DO we? Hmmmm?
As I understand it, we're supposed to have fun here. Aren't we?
# 313
Thanks , Firelock , I guess it will now be all nice and easy again ..
NorthWest
>> So you would use a Walschaerts for each cylinder. <<
I forgot to mention Baker valve gear is always an option - depending on what you prefer - an arrangement of levers that *can* be equipped with needle bearings or a simple expansion link and die combination . The Willoteaux piston valves are better at upper rpm speeds than they are at lower ones compared with regular piston valves and because of their in creased heat exchange from live to exhaust steam have been used preferably on LP cylinders in compound engines . As with a very large and heavy American Super Power design of a locomotive their double admission / double exhaust feature in combination with three cylinder simple propulsion should have made for an exceptionally free running engine if equipped with a well-designed progressive B or W valve gear . With 67 in drive wheel diameter a 2-12-4 with one-piece cast steel bed the engine could have been regularly run at 70 mph in competitive fast freight service . ( I'm not so much interested in what one time top speeds an engine could be tweaked to attain on test runs than I am in what the locomotive can do in actual daily traffic )
# 314
Right, getting back to the subject at hand. Three cylinder locomotives on American 'roads? Gee, I don't know. Traditionally American railroaders were pretty cool to the idea of multi-cylinder locomotives, that is. more than two cylinders. Certainly the UP had the 9000 series, and a few other 'roads who's names escape me tried them, but the added maintanance was a real turn-off, both for management and shop crews alike.
Maybe if a three-cylinder design that wasn't a maintanance headache was developed it might have had a chance, but I'm not sure. Here in the US they liked 'em simple as possible, even if it meant sacrificing some efficiency. That's just how it was.
# 315
Yes , that was the general attitude – they preferred the two cylinder simple . Yet , that was a consequence of the steam locomotive’s perennial need for servicing and maintenance . Everywhere in engine design and construction , the trend was from maintenance à easier accessibility to reduced maintenance à maintenance freeness . The steam locomotive building and running trade was the *only one* which had remained stuck on the second stage ‘easier accessibility’ with open ( unprotected ) plain bearings that had simple one-way or run-through total loss lubrication that had to be hand-refilled for about each trip .
Had this level of bearing technology been applied to the diesel engine it would have remained a very marginal power ! Since quite clearly it was unthinkable to go around and oil a twelve cylinder diesel engine every couple of x00 miles and since again it made its way on the rails only after full circle mechanical lubrication had long since become a vital integral component of the engine , comparing diesel locomotives with traditional steam say of the 1920s was *by default* unfair since the steamers had to do on those simple bearings needing frequent attention if only to refill oil cups or grease gunning around .
Inability to develop maintenance-free bearings of comparable level of technology – or maybe just unwillingness to leave the beaten path – was one of a couple of major factors in post WW-II years causing insupportably high daily maintenance work hour costs with steam locomotives otherwise almost overly robust and amply dimensioned . Diesel comparable mechanical maintenance freeness was only reached with roller bearings and some adjacent developments supporting their application . However , difficulty with 1940s – 50s technology was to apply roller bearings on an inside drive with crank axle(s) That is where imho mechanical lubrication for plain bearings should have needed to be developed . It would have demanded to securely fix supply lines to and fro bearings along rod sides or drill oil lines into axles and journals , drilling lines in wheels to supply pin and rod bearing .
At the same time bearings would have had to be provided with protection against dirt entrance and lubrication loss . This was a technology ready to be adapted from other machine construction , it should not have been overly tasking a development job to tailor bearing seals for steam locomotives – yet it was never accomplished although a few designs were put up and test applications *were* made – among these even a quite ingenious lubrication system that had oil circulating within each the bearings – however this with an imminent risk of running dry very quickly if a bearing seal should be damaged .
In my ventilating ideas concerning three cylinder engines I assume a system of mechanical lubrication has been sufficiently developed to allow for its safe application . Three cylinder engines should thus have had roller bearings on axles , protected and mechanically lubricated plain bearings on all power transmitting rods and mostly needle bearings on valve gear inside and outsides .
I guess , typical steam railroad view on economy could have been to take a dim look at extra first costs of such a system – and apply it to inside drive exclusively , thus perpetuating the engine’s need for ‘per trip’ hand lubrication just the same . That would of course have prevented making a full step towards maintenance freeness as in fully roller bearing equipped two cylinder engines such as the J class or NYC Niagaras and some J-3a Hudsons . Let’s assume good rational common sense would have won over beans counting together with a spirit of progress , steam locomotive so equipped would have been able to rise to 9000 ihp and even touch the 10000 ihp mark – i.e. six contemporary EMD diesels . To compete economically each of these diesels would then have been left with a margin of but 1/6 the running costs of such a big steamer – this would have been a degree of saving indeed challenging to meet , to put it mildly .
#316
Referring in general to #315 without a bunch of cutting and pasting quotes:
One principal difference between diesel (and other IC motor) lubrication and steam-locomotive running gear is that there is no great reason to 'seal' the hydrodynamically-lubricated surfaces: the great innovation being pressure oiling, and the enclosed crankcase and sump being used for more than just convenience in oil-gallery routing. A substantial amount of lubricant can pass out of the main bearings, for example, with no evil consequence so long as adequate pressurization in the oil flow can be maintained. This is not likely on either the axle bearings or rod bearings of a typical reciprocating steam locomotive.
Having said that, I do think, and have thought for many years, that pressure lubrication would be valuable on modern, high-thrust reciprocating locomotives. My original interest came from a sentence or two in the Encyclopedia of World Railway Locomotives describing cross-drilling of cranks and bearings in Chinese locomotives (I was never able to locate the actual locomotives this referenced). It is relatively simple to provide hardlines along the rods and then appropriate passages to the pins and/or journals (cf. an Isothermos axlebox for one technique of communicating passages without physically weakening pins). The oil supply for a whole pressure-lubricated rod assembly can be taken from an appropriately-armored flexline from crosshead to mechanical lubricator, with the return through, say, the bore of a hollow axle to a take-up and recirculating dry sump. Note that the plain-bronze-sleeve "UP-style" rod bearings, as mentioned in Bruce's book, would excel over roller bearings if provided with this type and level of lubrication, and would probably do so over a more than sufficient service lifetime without other attention (compare the situation with the Timken extended-life bearings for freight-car and locomotive wheelsets, which are able to go the equivalent of 500,000 miles without servicing, but the wheelsets they're installed on have a useful tread life far less than that...)
Part of the argument against such a system is that, as Juniatha mentioned, any leak in the system can result in fairly rapid drainage of large amounts of lube oil and then lack of flow to the bearings, which could then fail remarkably quickly and with very short effective 'notice'. If you are using something like Alemite, and one of the cellars pops off or loses its 'spring', the same result might occur, but only in the affected bearing, and probably with more warning time. I am not sure how well a cost-effective application of pressure lubing would hold up in typical practice, but it should be worth trying.
I have to wonder whether the long experience in developing 'sealed' railcar roller bearings with M-942 compliant lubricant could be applied to some areas of locomotive practice. While there are some technical disadvantages to relatively soft grease (instead of oil) lubrication of bearings from an efficiency standpoint, the maintainability in a wide range of practical operating conditions may make it preferable. (I might also mention that in my opinion most of the seal issues regarding rod bearings have been addressed and solved with respect to freight-car bearings...)
Now, in the 'bad old days' of American railroading, where trains had to stop more frequently than they do now, it was not so bad to 'oil around' at those times, or at least check reservoir or cellar lubricant levels and top off where needed. I suspect this was a factor in why good pressure-lubrication systems were not developed for North American steam. Given current and prospective EPA regulations, I'd expect most modern steam would have to get away from total-loss lubrication anyway.
With respect to posts earlier than #315: All points about three-cylinder power are appreciated.
# 317
Hi Overmod
>> #316
One principal difference between diesel (and other IC motor) lubrication and steam-locomotive running gear is that there is no great reason to 'seal' the hydrodynamically-lubricated surfaces: the great innovation being pressure oiling, and the enclosed crankcase and sump being used for more than just convenience in oil-gallery routing. A substantial amount of lubricant can pass out of the main bearings, for example, with no evil consequence so long as adequate pressurization in the oil flow can be maintained. This is not likely on either the axle bearings or rod bearings of a typical reciprocating steam locomotive. <<
Overmod , I’m not sure at what level of technical understanding we are conferring here – in my regard it’s trivial even to remark upon IC enclosed lubrication circuit engines do not have individually sealed bearings , on the contrary oil spill helps to lubricate cylinder walls . I mean – J – everybody who has owned a car knows *that* don’t they? I had earlier pointed out there are some fundamental differences between such engines and the layout of a classic steam locomotive , one of them being there is no enclosed motor block , nor crank case nor oil sump . Because of what is comprised in a motor block in IC engines is spread out over the length of an open design framework and wheel sets , a mechanical lubrication if not to spill oil to the open with intolerable oil consumption and unbearable oily staining of the locomotive has to have some kind of –
a) return of oil flow
b) sealing of bearings supplied by such a mechanical lubrication .
I would very much appreciate not everything which I have left off mentioning ( because I consider it common sense , otherwise generally known or side track ) should *not* be spotted as a detail that I must have blinked , overlooked , turned a blind eye to .. etc .
If I should be expected to mention each and every detail naturally to be co-considered in an actual application in ways self-understood as properly poising for a steady hand when using eye liner to draw eye lines then I may as well give up posting here since for every form of design or construction I would have to deliver a whole hand manual of how-to-do-it .
We all want to keep postings as concise to the point as possible , everything peripheral may be considered as self-understood until explicitly asked to explain .
>> Having said that, I do think, and have thought for many years, that pressure lubrication would be valuable on modern, high-thrust reciprocating locomotives. <<
Ah , there you are ! I remember having read your mentioning it before and I agree with that – although I would prefer arrangement the other way around supplying oil via drilled lines in journals , wheels and pins and using external oil lines for returning to avoid pressurized oil in these lines and in case of damaged line to make sure supply of concerned bearings can be up-held at the ‚expense’ in the word’s sense of lubricant and staining by leakage may be detected in just time . Reliable fixation and best impact-protected routing of oil lines along rods – probably within groove of I-profile at inner side – will be a thing to be carefully contrived and I would have oil lines exit both valve gear and power rodding via combination link and articulated links of oil lines although flexible lines with well guided , smooth non-vibrating U-turn might be an alternative – depending on what period we are talking of ( here : the 1950s where endurance of available materials likely may not have been supportive of such a solution ; today’s it would be no problem )
I would still prefer roller bearings on axles – they are just ideal for that application as they provide low rolling resistance from the start and *no* break loose resistance ; btw grease lubricated . It is true their rolling resistance rises lightly with rpm speed while that of mechanically supplied hydro-dynamically lubricated plain bearings rather falls with rising rpm speed – provided however conditions of running remain the same throughout the entire rpm range , which was not the case with steam locomotives since with rising speed the whole chassis was increasingly submitted to vibration and flexing which inevitably affected precision of alignment of bearing and journal surfaces and upset even spreading of loads over gliding surfaces .
There is a decided difference between axle bearings of freight cars and bearings of powered wheel sets on reciprocating steam locomotives ( btw same as with their carrying axles ) While the former are basically loaded in vertical sense by mass load on wheel sets plus dynamic loads by rail joints and imperfections of track , bearings on powered wheel sets of steam locos are submitted to variable horizontal forces potentially very much larger than mass load on them , especially at high speeds . Further , any play between axle box and pedestal liners is bound to submit axle bearings to pounding which may become destructive to roller bearing if allowed to grow enough – which seems to have happened in DB locomotives during the last years of steam when engines were run to do or die .
>> Given current and prospective EPA regulations, I'd expect most modern steam would have to get away from total-loss lubrication anyway. <<
That’s what I would expect , too .
#318
Brief snips from #317 for context; may be snipped/edited only for purposes of brevity or concision.
Juniatha in my regard it’s trivial even to remark upon IC enclosed lubrication circuit engines do not have individually sealed bearings , on the contrary oil spill helps to lubricate cylinder walls . I mean – J – everybody who has owned a car knows *that* don’t they?
in my regard it’s trivial even to remark upon IC enclosed lubrication circuit engines do not have individually sealed bearings , on the contrary oil spill helps to lubricate cylinder walls . I mean – J – everybody who has owned a car knows *that* don’t they?
I only mentioned this because I thought there would be many readers who did not recognize that. (Did you think I would presume that someone familiar with 300Gs would not know it? I may be a fool but I'm no idiot...)
... mechanical lubrication if not to spill oil to the open with intolerable oil consumption and unbearable oily staining of the locomotive has to have some kind of – a) return of oil flow b) sealing of bearings supplied by such a mechanical lubrication. I would very much appreciate not everything which I have left off mentioning ( because I consider it common sense , otherwise generally known or side track ) should *not* be spotted as a detail that I must have blinked , overlooked , turned a blind eye to .. etc.
b) sealing of bearings supplied by such a mechanical lubrication.
I would very much appreciate not everything which I have left off mentioning ( because I consider it common sense , otherwise generally known or side track ) should *not* be spotted as a detail that I must have blinked , overlooked , turned a blind eye to .. etc.
There is a partial exception, in that it is at least possible ... with modern materials, not '50s materials ... to seal bearings with the right kind of lubricant without formal pressure lubrication. There are some differences in load between steam engines and the usual sort of modern sealed roller bearings, but in New Zealand at least it has proven practical to run roller siderod bearings with M-942-compatible grease and good seals, and I see little reason why needle bearings in Walschaerts or Baker would be any different. That is just for the record; in our period you would want or need oil lubrication if you wanted long-term unattended supply, and it should be pressure lubrication, and the pressure lubrication REQUIRES a good return line, preferably imho one that drains by gravity (although that is scarcely a necessary design condition).
We all want to keep postings as concise to the point as possible , everything peripheral may be considered as self-understood until explicitly asked to explain.
I agree with this, but based on a fairly good number of previous questions, many of the people following these threads do NOT recognize why many of the mechanical details are made as they are. Tribology is likely to be one of those areas -- perhaps I am wrong, but I'd rather err on the side of 'too much explanation' in some of these cases. It is not intended as teaching people to suck eggs.
I would prefer arrangement the other way around supplying oil via drilled lines in journals , wheels and pins and using external oil lines for returning to avoid pressurized oil in these lines and in case of damaged line to make sure supply of concerned bearings can be up-held at the ‚expense’ in the word’s sense of lubricant and staining by leakage may be detected in just time.
Perfectly practical, and well-reasoned of course. I was proceeding from a different assumption: that maintaining pressure seal integrity between the wheels might be more difficult to inspect and adjust (perhaps especially inside a cannon box) than doing the high pressure with easily-sealed hoses and lines from the lubricator down to the bearings. (You will note that I don't extend the pressure lubrication to the driver-axle roller bearings, and that would likely continue to be true even if they were oil-lubricated as in the '40s -- the oil return goes up the hollow axles ).
... the 1950s where endurance of available materials likely may not have been supportive of such a solution...
I think it was; armored pressure hoses for aircraft applications were mature by the end of the War, and could have easily handled the peak pressure required for this service; I think materials aging would have been a far more serious concern than flexing-induced failures or oil degradation of the hose lining. I agree that with more modern materials there would be no insurmountable problem in any respect.
I would still prefer roller bearings on axles – they are just ideal for that application as they provide low rolling resistance from the start and *no* break loose resistance ; btw grease lubricated.
Full agreement. I think it was well-established even in the days of the Timken Four Aces demonstrations that roller bearings belong on driver axles, and are advisable on lead-truck axles; I believe they should be used on trailing trucks (with heat shielding and better ash and blowdown management used as necessary to give them long practical life.) Any alignment problems are covered with the cannon box; any longitudinal alignment issues are handled with hardened liners in the front pedestal jaws and Franklin wedges for the rear.
... bearings on powered wheel sets of steam locos are submitted to variable horizontal forces potentially very much larger than mass load on them , especially at high speeds
Timken actually noted that proper 'absorbance' of thrust and shock from any angle was an important characteristic of roller bearings on driver axles. Even with a fully circumferential plain box -- which is a difficult and somewhat dangerous thing to experiment with on a large locomotive -- there will still be problems with hydrodynamic film characteristics that simply don't cause trouble for good rolling-element bearings.
I hope I am correct in assuming that 'horizontal' means fore-and-aft, and you're discussing the longitudinal inertial momentum in the rodwork as well as normal surge. I think the Franklin wedges can adequately handle these forces (because the forces reverse at each revolution, and cannot build up net pressure against the rear jaw even at high force, as the frequency increases 'enough' to make the duration of the actual peak force against the wedge slight, while still allowing time for the wedge to accommodate during the 'unloaded' half of the cycle).
I also see issues with lateral (which needs to be provided on multiple axles) as some of the 'usual' methods of providing compliance (for example, the Fabreeka composite springs on pedestal tender axles) would not be workable for driver axles in our period of interest, for many of the same reasons you mentioned in the difference between freight-car and locomotive bearing loads. Something more than the 'standard' spring centering of contemporary lateral-motion devices may need to be provided, and I'd welcome suggestions as to what would be best.
#319
Some of the comments over in the N&W threads reminded me of something that might be relevant to this discussion.
I had a copy of a blueprint elevation, from the North American Locomotive Company (dated 1979!) and mentioning Riley Deem as a consulting engineer, showing a "Chapelon Modification for American Railway Service". This has some very interesting notes and details (e.g., it is fully bidirectional and equipped with the 'Aubert' system of push-pull control, and is indicated as having a 'full condensing' tender.) It has an induced fan for draft, with the note "This area SAR design" in the region of the smokebox just ahead of the main steam pipes to the cylinders, and "Exh. turbine geared drive" under the front of the smokebox. Locomotive is a three-cylinder compound, doubtless similar to the arrangement on 242 A1, although my copy is now heat-damaged and does not show the arrangement of the center cylinder's drive. The outer cylinders (labelled "BP cylinders"!) are slightly inclined and drive on the third coupled axle.
It might be interesting to consider whether this design accurately reflects what 'might have been' the state of contemporary development if reciprocating steam had continued to develop through the 1970s as first-line power in the United States.
# 320
Yes , I have heard of that - it was one of seemingly several attempts to 'market' the 152 type of the 1950s family of high performance locomotives by the DEL / André Chapelon . This was because the locomotive was fully designed to the point ready for construction - one was actually started to be built when the decision to abandon steam came down on it like an executioner's ax ( some people in SNCF HQ must have sighed audibly muttering something like “Boy , that was close !” ) it was all-important this engine was not going to be built ! In the 1970s it seems a group of Englishmen have toured around with the plans trying to find a place for this – still quite advanced – design of steam locomotive , allegedly they even tried China . Plans failed , no wonder if you come to think of the design's pretty special tailoring to specific conditions and requirements of the SNCF , such as low loading gauge , light axle load as related to American standards and many more . What adaptations they subjected the original design to - such as condesing SAR type as Overmod has mentioned - may or may not have been adequately up to the general level of design of this locomotive , more likely it would have been some sort of tampering as the railways usually have submitted their steam locomotives to which more often than not didn't improve anything to put it mildly .
The 1950s family of types were three cylinder compound , I have so far not read or heard a conclusive explanation for Chapelon’s change over from four to three cylinder type , except for a rather general notion that a single crank axle was easier to manufacture and more solid than a dual crank axle . While that is true , French locomotives never featured one-piece forged dual crank axles such as the Baden IVh Pacific . The latter had its dual crank axle last over 1.5 million km ( near 1 million miles ) – better than ten times (!!) the regular life span of built up crank axles used on the SNCF . In my view , a large part of the problem with French dual crank axles was due to
* composite construction
* too light construction with too little rounding off at critical points and
generally insufficient adaptation to actual stress flow –
consequence : stress concentration
* webs with integral counter weights , causing proper frequency vibration at
lower frequency than without – in combination with staggered rails this seems
to have caused regular running with crank axle at or near proper frequency vibration
for extended part of regular travel , sharply aborting to longevity
* insufficient guiding by front bogie , loading excessive lateral stress on first coupled axle ,
which was dual crank axle in many Pacifics
* insufficient rigidity of composite plate frames , not enough flange thinning on
crank axle to prevent lateral flange loads
I believe it had not been impossible within 1950s technology of forging to produce one piece dual crank axles of solid , stiff construction with proper frequency beyond such reached by excitation in traffic and especially in combination with large scale welding of rails then in fast progress would have allowed to fully address this problem , i.e. have axles regular life exceed 1 million miles mark .
I’m not sure if the whole project of marketing this locomotive type to foreign railways had been to Chapelon’s contention ; since he was an exceptionally bright man , he may very well have seen his engine had a dim chance to succeed under such conditions and I’m not sure if he was happy about perspective of likely failure . There would be more I might want to write on this , however it would tend to be more about emotions , less about facts and I don’t think it’s wanted or warranted .
# 321
Here is one of those I feel might have rolled out of ALCO yard -
the NYC S-2a , improved 84 in drivered Niagara :
Hope you like it .
# 322
I really really really like the Niagara. Because of my limitations with the engineering in steam locomotives, I find myself wondering if a T1 or a Niagara would have been 'improved' in some acceptable way by increasing the diameters of the drivers. I have also suspected that I was barking up the wrong tree in that regard, but never found an opportunity to ask my question. So, J, thanks for providing it. What should I understand, apart from balancing issues in the smaller diameters?
Crandell
# 323
Hi Juniatha! Wow, looks like you've done it again with your re-design magic, that 84" hypothetical Niagara looks like a real racehorse, as opposed to the standard Niagara which reminds me of a battering ram with a bad attitude!
I don't know how it would have handled the NYC clearances, the original Niagara was pretty tight in the tunnels, but your re-design's fun to speculate on.
# 324
Hi Crandell – Hi Wayne
84" drive wheels :
While the Niagara's boiler was pitched up as far as can be within the loading gauge , there was space below the boiler to install slightly ( + 5 % ) larger wheels .
With the Niagara as built having easily reached and passed 100 mph you may ask why larger drivers ?
While the cylinders did pass the steam to rev up enough to make it happen , there still was throttling - to be seen as a falling line above some 80 mph in the power over speed graph . To fight that , I would have proposed to use both a freer steam circuit *and* somewhat larger wheels to contain rpm speed a little . Both together should have made enough of an effect to trim down specific steam consumption (heat consumption , actually – but that’s another story ) at 100 mph by some 12 % . Add to it a freer exhaust trimming back pressure at cylinder head down from ~ 30 psi to but 12 – 14 psi , and increased superheating for having 3350 kJ/kg live steam heat content and at but 77 % maximum point of thermodynamic efficiency to keep it conservatively ( DR standard light Pacific 03 class of 1930 : 75 % max thermodynamic efficiency ; Chapelon rebuilt two cylinder 141.E.113 : ~ 80 % max thermodynamic efficiency ) ssc would be trimmed down to 7.79 kg/kWh or 12.80 lb/ihph . At 100,000 lb of live steam , net to engine , which should be well within grate limit of a re-draughted Niagara boiler ( fed suitable grade of coal ) this would release 7800 ihp in the vicinity of 100 mph or over . With a little fine tuning here and there 8000 ihp around 110 mph should reasonably be within reach retaining given mechanics of the two cylinder engine .
If you would ask for more , you would have to change design for a three cylinder SE engine unit of much the same individual cylinder volume as given in the original Niagara as built – i.e. 50 % larger total cylinder volume - otherwise you would see the engine reach her top performance level only at speeds well above what was commercially practical in ( extended , that is ) steam times .
As for the Duplex engine : *any* ( free running ) four cylinder engine of boiler capacity and engine output equivalent to a large two cylinder engine with which it is to be compared , by default has a higher rpm ceiling and tends to outrun the two cylinder engine in the upper speed range while being somewhat less efficient and thus less 'punchy' in the lower speed range .
Thus , whatever an improved Niagara could be tuned to do in the range of 80 – 120 mph , a likewise tuned T1 could easily do and then some . Clearly , such a 'race-happy' type of locomotive would only have made sense in combination with a billiard table smooth and laser straight race track of a mainline to run around 110 - 120 mph for miles and miles on end , making super-fast start-to-stop timing , for cities connection back then able to compete with air travel over medium distances .
# 325
Thank-you, J. I actually understood that.
It is a question to which I wander, which needs not be answered, but I sometimes wonder if the steam era had somehow been extended by even five years, would the Chesapeake & Ohio or Santa Fe have found the need to bring their excellent engineering & design to build truly breathtaking speed demons? Maybe not so much the C&O with their needs, but the Santa Fe had some good long runs of track that could have been put to use with high-speed passenger trains.
For now, I am quite happy with the standing of the three, the T1, the S1, and the J. The Americans should be proud of those achievements.
# 326
Hi Crandell
What the Santa Fe might have come up with .. hmm .. interesting thought .. with my East cost preponderance I don't feel like the right one to answer . We may speculate they might have conferred with Baldwin about some very specific ideas they had developed on a top class locomotive and have them built it . Since their unique high drivered 2-10-4 obviously operated well on their mainline , what about if they had combined 10 drivered with 4-8-4 front end to realize a 75" 4-10-4 dual-purpose engine - notably in view of their very severe mountain passes ? Only , I hope they had chosen driving the *mid-way* axle , not second coupled as in a Northern Type - ouutch , please no-no-nooh !
On the other hand , maybe they’d put right wanting points in steam circuit of their big 4-6-4 and that would have set them free on the long level mainline sections .
The Chessy ( or the Virginian , rather ?) would be my favorite for finally asking to have my 2-8-8-6 Grizzly Type built - the w/a I would practically consider about last word in development of the classic line of the North American Simple Expansion Mallet . And maybe they'd convinced Lima to seriously keep within axle load limits , too . Oh , I guess if the Chesty Railroad had ordered some , the Virgins , too , would have considered it a *must have* - *g* .
#327
Juniatha What the Santa Fe might have come up with .. hmm .. interesting thought .. with my East-coast preponderance I don't feel like the right one to answer. We may speculate they might have conferred with Baldwin about some very specific ideas they had developed on a top class locomotive and have them built it.
What the Santa Fe might have come up with .. hmm .. interesting thought .. with my East-coast preponderance I don't feel like the right one to answer. We may speculate they might have conferred with Baldwin about some very specific ideas they had developed on a top class locomotive and have them built it.
We actually have little reason to speculate: there are sources. One is Worley's book "Iron Horses of the Santa Fe Trail" which mentions the cab-forward duplex and steam turbine. Brashear's "Santa Fe Locomotive Development" alludes to a 1934 proposal for a cab-forward, watertube 4-6-2; this was specifically abandoned with the understanding Diesels were more promising for that service.
Since their unique high drivered 2-10-4 obviously operated well on their mainline , what about if they had combined 10 drivered with 4-8-4 front end to realize a 75" 4-10-4 dual-purpose engine - notably in view of their very severe mountain passes?
Why only 1" difference, with all the costs involved with extra parts and procedures? 74" was plenty high for a mountain engine (I will not go into why I think it would be 'enough' here, again). I'd put the improvements into other places.
... I hope they had chosen driving the *mid-way* axle , not second coupled as in a Northern Type
I would agree here. I'd presume that the rod dimensioning and geometry for this 4-10-4 would be common to, say, that of the 5001 class, with longer piston rod and crosshead guide structure (as on the T1) to allow driving, as the 5001 class did, on that center axle, without compromise due to that lead-truck wheelset between cylinder block and lead drivers...
On the other hand , maybe they’d put right wanting points in steam circuit of their big 4-6-4 and that would have set them free on the long level mainline sections.
Maybe so... but I sure wouldn't waste much time on a Hudson when the later 4-8-4s were already demonstrably faster -- put the design work on eight-drivered locomotives instead. For example, put 84" drivers on THAT size locomotive... and, well, that explains ATSF interest in divided-drive without any further speculation, doesn't it? ... and you will have a better answer to ATSF passenger consists than any six-coupled locomotive.
In my opinion, a Grizzly would be wasted on either C&O or Virginian -- that's a high-speed, high-power freight locomotive, not a strictly coal-drag-oriented locomotive, and should be used as such. (Neither of those roads was particularly known for its fast-freight traffic, I think.) I for one would like to see whether a Grizzly 'cub' could be built within B&O clearances, like an uprated EM-1 (on the short list of most beautiful articulateds already). This might be more than usually relevant to the 'dieselless alternative history' as B&O in particular would have needed high-horsepower steam for the grades on its principal Allegheny crossing...
# 328
Hey Juniatha---
You said something very important. "within B&O clearances". Several years ago somebody put out a video of B&O steam power and speculated that an EM-2 would have been even more spectacular than an EM-1. Fat chance! The EM-1 was designed to fit within B&O's clearances, and that's why the last (and arguably the best-looking) 2-8-8-4 was also the smallest. Hard to believe that, for anybody who ever stood beside one or sat in the cab! Maybe you could have increased heating area by installing the bottom portion of Lima's double Belpaire boiler, but I question whether the top portion could get through some of B&O's tunnels As it is, the stack was uncomfortably close to the roof in a lot of those bores. But wouldn't a more powerful boiler just worsen the factor of adhesion on an engine that was already a bit slippery at 4.22? At least that was what I was told by the men who ran them. Using a welded boiler might make it even lighter, and I'm not sure that would have been a good idea.
This thread has focused a lot on N&W and I guess that's to be expected, considering N&W's belated dieselization, but the conditions found on N&W were not universal, and if the RR industry were to hang onto steam in general, then the principles we need to address have to be more universal. Most RR operations of 1950 didn't need as much power as has been discussed here, and couldn't have fit these monsters into their loading gauge without rebuilding the physical plant anyway. Yes, we're talking about alternative history (that oxymoron again), but what's suggested has to be kept within the bounds of what's likely or practical.
I used to think a 4-8-6 would be a great thing, but it's been suggested that a 6-wheel trailer would be more than what's needed for an eight-coupled high speed engine. I'm not an engineer, so I don't know. I do know that I have always questioned B&O's decision to build those lanky, ungainly T-3 4-8-2's with stretched boilers from 2-8-2's and 4-6-2's. I'd have suggested using the boilers from S and S-1 2-10-2's, or nearly identical ones from 2-8-8-0's, and put them onto a 4-8-4 chassis instead. The boilers could have been stretched a bit to make room for a combustion chamber. B&O was already convinced of the value of syphons, possibly as a result of their ill-fated watertube boiler experiments. A cast frame with roller bearings on rods as well as all axles could have produced a pretty remarkable engine, I think. Drivers somewhere in the 74-80" range. Come to think of it, didn't AT&SF play around with the idea of using old N&W Y-3 boilers on new AT&SF 4-8-4's? Baker gear would be fine, and if some genius could come up with a Poppet valve system that worked well enough and was easy and cheap to maintain, then so much the better.
No matter what, railroads in general needed smaller locomotives for general utility service to do the things that diesel road switchers were doing in 1950. The last Mikado built by Lima for U.S. service is a good example. Back in the 1920's, Akron Canton & Youngstown decided the light USRA 2-8-2 was an excellent engine for its purposes. Over the years they put improvements on them until no. 406 was delivered from Lima in 1945. She had roller bearings on all axles, thermic syphons, Baker valve gear, and an Elesco Exhaust steam injector. A bigger engine would have required rebuilding the railroad in many places, replacement of a turntable, lengthening of sidings, etc. etc. etc. Number 406 did not have the welded boiler and improved steam distribution that has been proposed in other postings, but she was as up to date as could be expected for the AC&Y at that time. If the diesel had not come along, 406 probably represented the kind of loco that railroads would have wanted for this type of service. But the diesel did come along and AC&Y pounded on Alco's door, wanting to buy RS-2's. Alco couldn't provide them, so AC&Y went to Fairbanks Morse and started buying road switchers in 1948. Number 406 was scrapped in 1954, less than ten years old, and the last AC&Y steamer went in 1955. So steam's fate was sealed, no matter what.
I haven't read every comment in this thread, so I'm not sure how much has been said about switchers, but they would have to have a place in a diesel-less world. The USRA 0-8-0 became the de facto industry standard shortly after it was first built, and copies were built nearly 35 years later with improved fixtures but basically the same engine. Is that a record? Could a better switcher be designed?
# 329
Hi ACY ,
>> You said something very important. "within B&O clearances". <<
Da**** - there I'm decorated with having said something not quite insignificant - and then I didn't say it ...
Never mind , guess it can happen – in the summertime
http://www.youtube.com/watch?v=wvUQcnfwUUM
( btw , Jeezzus imho - these guys *do* look weird , don't they ? )
Well , don’t worry – bee happpyih
http://www.youtube.com/watch?v=L3HQMbQAWRc
hu-hu – hu-hu – hu-hu – hu-hu ..
reggaerds
Ju-un-ià-tha
# *330*
Hi Juniatha! So THAT'S what Mungo Jerry looked like! A werewolf! I remember the song quite well from the radio, no MTV in 1970.
Our community is FREE to join. To participate you must either login or register for an account.