LithoniaOperator Flintlock76 "Catechism Of The Locomotive." Was it published in Baltimore, by any chance? (Inside joke for all you Catholics out there! ) And is there a section with questions and answers? And if you don't know the answers, you have to kneel in the corner for hours?
Flintlock76 "Catechism Of The Locomotive." Was it published in Baltimore, by any chance? (Inside joke for all you Catholics out there! )
"Catechism Of The Locomotive."
Was it published in Baltimore, by any chance?
(Inside joke for all you Catholics out there! )
And is there a section with questions and answers? And if you don't know the answers, you have to kneel in the corner for hours?
JMJ, don't remind me!
Lady Firestorm and myself, 100+ questions and answers for Confirmation!
We shudder at the memory!
Still in training.
Seriously, I stand in awe of men like Forney and his colleagues. All that math to make the magic, and NO computers to help them, just a pencil and paper. Maybe slide rules, if they had them in the 1890's. Maybe.
And then putting the plans on paper with T-squares, French curves, triangles, steel compasses, India ink bow-point pens, and a lot of patience. No CAD, not even dreamed of. I took a mechanical drafting course in high school in 1971, darn near the end of the era and used all those implements, so I know wherof I speak.
That's how that mind-blowing schematic drawing of the UP 4-12-2 Mr. Jones posted on April 3d was done.
Awesome men.
seppburgh2As a companion book, recommend M.N. Forney, Catechism of the Locomotive, revised 1890 edition.
1891 edition available for download or online reading, here:
https://books.google.com/books/about/Catechism_of_the_Locomotive.html?id=jBk9JFrrT50C
Overmod Likely what you have is an artifact of high-speed slipping, producing (perhaps little more than momentary) rotational speeds affecting the valve gear equivalent to 140mph or greater. Since the T1s were not equipped, even in testing, with something like Valve Pilot, and the Jones-Motrola speedometer worked only off one 'engine,' this effect would have been easy to miss in service, but it sure isn't difficult to predict once you understand how easy it is to momentarily unload one engine of an unconjugated four-coupled duplex...
Likely what you have is an artifact of high-speed slipping, producing (perhaps little more than momentary) rotational speeds affecting the valve gear equivalent to 140mph or greater. Since the T1s were not equipped, even in testing, with something like Valve Pilot, and the Jones-Motrola speedometer worked only off one 'engine,' this effect would have been easy to miss in service, but it sure isn't difficult to predict once you understand how easy it is to momentarily unload one engine of an unconjugated four-coupled duplex...
The reports of 120mph (and higher) speeds could well be some sort of CYA by Franklin. A perhaps related issue, I remember reading about problems with high speed slip with the FM C-Liners, which led to commutator flashing on the Westinghouse electrcal gear. IIRC, the C-Liners were slipping on NYC track in Indiana, although that may have been in a different part of the state as to where the T-1's were slipping.
As a companion book, recommend M.N. Forney, Catechism of the Locomotive, revised 1890 edition. What I like about Mr. Forney's book is the first 40 pages is devoted to getting you up to speed on the concepts and formulas used in his book. Along with others such as force and motion, principles of the lever, and the forces of air and steam. O, and a 14 page deep dive into work, energy and the mechanical equivalent of heat. After the basics, he gets into the steam engine. The book is written in a Q&A format. Example Q # 85 "What is the motive power employed in ordinary steam engines?" Answer "The expansive force of steam." If you have been paying attention, M.N. Forney happen to build steam locomotives as a side gig to his writting. So, folks that is your lesson on steam. Read chapter XVI, the valve-gear and pay attention on question 359, how is motion of either eccentric communicated to the vale? There be a quiz on link motion next class.
LithoniaOperatorI recently bought this extremely impressive, and very informative book by the late William Withuhn:
I purchased a copy shortly after publication as well. Definitely a worthwhile investment
Google has a suprising number of pages for "review" here:
https://tinyurl.com/y4zrupct
For those who may be "on the fence" about buying the book.
Regards, Ed
Flintlock76Certainly Drury's book is a very general overview, but maybe there's something to this?
The problem really is that poppet valves contributed too little, too late, with too many late-breaking problems to matter in North America. Their advantages, for the relatively high price and added complexities, came in two general areas: a meaningful saving on water rate and fuel consumption, and much higher 'volumetric efficiency' at higher cyclic rpm (for the T1, above 90mph or so) than the best equivalent piston-valve equivalent.
Neither of these mattered after 1948 or so, and even as late as 1952, a "superior" installation (on ATSF 3752, as good a testbed as you likely could find) didn't make any damn difference.
There is a bit more in the picture, too: the RC shifting-cam system in type C was something of a maintenance disaster waiting to happen: an expensive ground continuous-contour camshaft using spherical roller followers was sure to start grooving and preferential wear around the 'favorite' cutoff positions. NYC thought so little of the observed saving from a type A installation specifically optimized for fuel and water consumption that it took the very expensive engine out of service long before the comparable 'sisters', by 1950 if I recall correctly, and most importantly didn't apparently even consider modifying to RC despite PRR having done all the heavy lifting on the "B-2" package that would have been required.
For what it's worth...
In his book on North American steam the late George Drury said the problem with poppet valves was they came along just a little too late for American steam designers to master them. And, by the the time poppet valves arrived on the scene the steam era was coming to a close anyway.
Certainly Drury's book is a very general overview, but maybe there's something to this?
SD70Dude Overmod rcdrye "Like having an expensive mistress at every division point!" Famously said of the Southern Pacific "three barrels of monkeys" (alright, it was actually a little different) 4-10-2s... I recall reading a story somewhere that one of the SP shops had a mechanic who also happened to be a terrible alcoholic, this fellow would regularly miss work or cause other problems. But they kept him on the payroll because he could properly set the valve timing on the three-cylinder locomotives, which no one else seemed to be able to do.
Overmod rcdrye "Like having an expensive mistress at every division point!" Famously said of the Southern Pacific "three barrels of monkeys" (alright, it was actually a little different) 4-10-2s...
rcdrye "Like having an expensive mistress at every division point!"
Famously said of the Southern Pacific "three barrels of monkeys" (alright, it was actually a little different) 4-10-2s...
I recall reading a story somewhere that one of the SP shops had a mechanic who also happened to be a terrible alcoholic, this fellow would regularly miss work or cause other problems.
But they kept him on the payroll because he could properly set the valve timing on the three-cylinder locomotives, which no one else seemed to be able to do.
Erik_MagThe reports were that trains were often exceeding 120mph and on rare occasion would approach 140mph.
This keeps coming up. And being disproved, I think. The story originates, I think, with Vernon Smith, but it has acquired the verisimilitude of 'timing with a stopwatch' to get around the perceived speedometer problems (T1 speedometers only went to 100mph, a somew)hat surprising oversight considering the locomotive's intended speeds, but perhaps squarely in the domain of 'plausible denial'.)
Incidentally, the short-term solution was to modify the return springing, with the effect of debouncing the valves at some critical speeds in the range, and the longer-term solution from 1948 (only partially implemented) was to improve the metallurgy of the valves, including centrifugal casting. No particular reason I can see why this wouldn't have fixed the issue satisfactorily...
Greetings from Alberta
-an Articulate Malcontent
The chapter on the T-1 and poppet valves may toke the fires about the T-1's top speed. Withuhn states that Franklin was wondering why the poppet valves were breaking on the T-1's, when a similar installation was rock solid on the K-4. He went on to state that the valves were breaking on a specific division, so they had a "mystery rider" on the trains with a stopwatch to time the milemarkers. The reports were that trains were often exceeding 120mph and on rare occasion would approach 140mph.
rcdrye"Like having an expensive mistress at every division point!"
OvermodYou can get some idea of the maintenance issues with the Gresley gear from the UP having adopted this somewhat extreme expedient.
My copy arrived in the mail today, been too busy reading to do much posting.
Update: It's a good read, and brings back memories of my earliest reading about steam locomotives, particularly the June 1967 article on the D&H high pressure compaounds and the June(?) 1968 article on the "Big Engines", with Withuhn's book explaining the math behind the '68 article. One interesting tibbit was learning that D&H 1403 was credited with a 10.5 to 12% thermal efficiency.
Overmod Jones1945 Something that has not been brought up explicitly with respect to these multiple-cylinder engines -- but that ought to be -- is the arrangement of valve gear to run those internal cylinders. We have just had an interesting failure involving one such, on the replica British locomotive Tornado. Note in the above pictures of the Nine there is no room for inside valve gear to work the center cylinder's valve 'from the rear'. What's used instead is Gresley's adaptation of Holcroft's 'conjugated' valve gear, which uses a combination of levers to derive valve motion from the two 'conventional' gears on both sides of the engine. There are some practical issues with doing this and maintaining it long-term, but it should be recognized and remembered that the official world-record speed was reached on a locomotive using this arrangement. Later, some Nines used an alternative arrangement which was a full third set of valve gear mounted behind the regular arrangement on one side, with a long rod and crank arrangement to reach the center-cylinder valve location. You can get some idea of the maintenance issues with the Gresley gear from the UP having adopted this somewhat extreme expedient. Other locomotives, notably in Poland, use the same general approach. Note the conjugated valve-drive arrangement in the YouTube clip for the four-cylinder locomotive in Jones1945's post. This is technically simpler in not requiring the 2:1 lever-in-a-lever arrangement...
Jones1945
Something that has not been brought up explicitly with respect to these multiple-cylinder engines -- but that ought to be -- is the arrangement of valve gear to run those internal cylinders. We have just had an interesting failure involving one such, on the replica British locomotive Tornado.
Note in the above pictures of the Nine there is no room for inside valve gear to work the center cylinder's valve 'from the rear'. What's used instead is Gresley's adaptation of Holcroft's 'conjugated' valve gear, which uses a combination of levers to derive valve motion from the two 'conventional' gears on both sides of the engine. There are some practical issues with doing this and maintaining it long-term, but it should be recognized and remembered that the official world-record speed was reached on a locomotive using this arrangement.
Later, some Nines used an alternative arrangement which was a full third set of valve gear mounted behind the regular arrangement on one side, with a long rod and crank arrangement to reach the center-cylinder valve location. You can get some idea of the maintenance issues with the Gresley gear from the UP having adopted this somewhat extreme expedient. Other locomotives, notably in Poland, use the same general approach.
Note the conjugated valve-drive arrangement in the YouTube clip for the four-cylinder locomotive in Jones1945's post. This is technically simpler in not requiring the 2:1 lever-in-a-lever arrangement...
Thanks a lot, Overmod. The following video is a 3D demonstration of 3 Cylinder Locomotive: Union Pacific series 9000, it renders how the "conjugated valve gear" works and the rod connecting the middle cylinder to the driver. For the "double Walschaerts" valve gear, it doesn't show much of it. The 3D model is a railfan-made model from more than 10 years ago so the smoke effect doesn't work anymore.
It is an unlisted video, please tell me if it doesn't work. Thanks a lot.
Regarding the new book by William Withuhn, Pennsy fan will find a new-old pic of the de-streamlined PRR S1 #6100 posing for a publicity photo after her second overhaul. Preview is available on Google Books.
Jones 3D Modeling Club https://www.youtube.com/Jones3DModelingClub
Note the conjugated valve-drive arrangement in the YouTube clip for the four-cylinder locomotive in Jones1945's post. This is technically simpler in not requiring the 2:1 lever-in-a-lever arrangement.
Australians came up with what they perceived as a nifty way to get around the whip and bearing-wear issues with Gresley timing: they modified the geometry to work with shafts in torsion. All would have been well if they had used shafts that were adequate in torque stiffness... for example, using a large-diameter relatively thin-walled hollow shaft... but this nicety escaped them, the shafts they used whipped as badly or perhaps worse than Gresley gear at high speed and load, and the diseasels came, etc. etc. etc.
LithoniaOperatorOn the 3-cyl engines, did the middle cylinder connect only to the main driver, or was there a connecting rod to other axles?
On almost all multiple-cylinder locomotives, the inside cylinder(s) only connected to ONE driver pair. A range of good reasons accounts for this, starting with recognition that even one crank in a driver axle may significantly weaken it (or require very stout additional structure). In general, American practice went toward keeping rod bearings outside where they could be easily inspected, and this is one reason why the late American four-cylinder locomotives, in the era that divided-drive began to become essential for high-power reciprocating locomotives, were duplexes (and standard 90-degree quartered DA duplexes without inside cylinders, at that!)
I think it has been noted that you do NOT have to have all three or four cranks on one axle; the de Glehn-du Bousquet locomotives, one of the most successful approaches to compounding in the world, specifically divided the drive with the inside cylinders driving on the lead coupled axle and the outside ones on the second. (There is no formal reason why a locomotive needing to preserve good rod angularity couldn't divide the drive between the first and third axle, or even the second and third, the idea being to minimize weakening or other compromise of working strength while preserving adequate access for maintenance)
You wouldn't get an inside rod 'forward' from a cranked axle without at least one other crank throw, and a matching offset crank on the leading axle(s). It makes far more sense to increase the strength of the side rods slightly to take the additional thrust of the third cylinder, which is the method used on the UP Nines (and the Baldwin 60000 'compound equivalent' 4-10-2 preserved in Philadelphia) and by far the preferred solution in the post-1928 era of lightweight and stronger rod materials and fabrication.
Likewise, any arrangement to take a 'third rod' back to the following driver pairs would involve expensive cranks. Here you could 'technically' use a method like fork-and-blade (with removable caps or straps at the back of the main and the front of the inside coupling rod) but this would significantly weaken the most vulnerable part of a three-cylinder engine, the rear of the inside big end, and make for interesting stress management in forging, machining, and maintaining the inside main rod. I won't go into the shenanigans that would be associated with providing good roller bearings either on cranked axles (it's no fun!) or on split-cap rods (it's far less than no fun!)
Of course there is a famous theoretical (and with no disrespect meant to Mr. Withuhn, best left theoretical) example of inside conjugating rods on a locomotive with four outside cylinders. That is the ACE 3000 implementation of the Withuhn conjugated duplex principle, which uses a pair (!) of appropriately-phased inside rods between the two mains (!!) of a small* 4-8-4-sized chassis instead of the outside rods a 4-8-4 would usually have between the main and third driver pairs. Somewhat amusingly, the patent discussion (which is very specific in some other respects) gets all handwavy when it comes to showing the structure that would support the two cranked mains in the actual locomotive -- all we get is the next worse thing to a stick drawing, showing the geometry in a frame about as sophisticated as the one on an 1850s 4-4-0. What would actually be required to provide effective inside-rod conjugation in that area is not very pretty even before you start having to account for suspension action ... and then cross-level accommodation.
There are ways around this, and some of them are surprisingly early: several attempts to use gears to get around the 'rod problem' go back right to the earliest days of practical steam locomotives (and, in fact, this kind of approach factors definitively into most of the designs at the Semmering trials). Riley Deem (of Lima) rather famously if somewhat mysteriously designed geared conjugation for the Q2 locomotive, and while I have not yet seen specifically what he intended, it is relatively easy to assure the necessary mechanics to give all the benefits of the inside conjugating rods with few, if any, of the drawbacks.
Bulleid's Leader, which we have already brought up in a different context, had a three-cylinder crank and a sprocket to drive the valve gear, on the center axle in each power truck, and there was no room for inside rods even had someone wanted to fit them. Outside rods would have worked, albeit crudely. but Bulleid rather sensibly proposed outside roller chains on sprockets. The problem was that he needed sprockets on each end of the leading and following axles, and that meant four sprockets abreast (two each side) plus the width of the oil-bath enclosure around them, outside the roller bearings and hub liners and all in the sideframes, to fit inside the stringencies of the British loading gage. The consensus was there was only room for one chain each side, but the suspension didn't much care for that (and neither do I); if you had to run much more horsepower than that of an elderly tank engine through that arrangement it would not end well.
LithoniaOperatorOne really cool engine shown is a 2-8-8-8-2 with the aft set of drivers under the tender!
If you can find a copy of Fryer's book on experimental British steam, he goes in some detail into the question of 'motor tenders'. The Triplexes (and Henderson's proposed 'too many legs and not enough steam' Quadruplexes and Quintruplexes actually fall in this category (as do the Southern's experiments) -- if you think of them as 'tenders with a steam-locomotive's engine underneath' instead of a long chassis carrying coal and water on its rear part (as in, for example, a Beyer-Garratt) you will understand what's going on a bit better.
The situation is a bit complicated on the 'built' Triplexes because of the (frankly, rather sensibly derived) arrangement of compounding. Ideally a given high-pressure cylinder (at reasonable American boiler pressure) "wants" to feed somewhere between 2.3 and 2.7 times the low-pressure cylinder volume to equalize contribution to tractive effort. The Triplexes had all common-diameter cylinders, with one HP feeding the compound cylinders 'in front' and the other the cylinders on the tender engine. Only the residual exhaust pressure on the forward engine contributed to draft (!), the LP tender engine exhausting through an escape-pipe at the rear (there are reasons why it was not blown down to heat the tender water, as you might think would be a good use of that otherwise-'wasted' heat and water mass).
Strangely, lack of draft on the fire does not seem to have been an issue with them; operation at necessary long cutoff with inadequate boiler size was the real problem. A major issue was that, in this era, the contribution of radiant heating to boiler effectiveness wasn't that well-realized (the theory still being more that the water around the firespace was 'renewable insulation' and the steam mostly generated by the tubes and flues).
Thanks, BigJim.
Unfortunately, I just don't have a good enough eye for such drawings that I'm sure I can discern that. By looking at photos elsewhere, I can see that the outer (regular) main rods connect to the third set of drivers; so you are saying that the middle main rod connects to the second set, right?
Are there "side rods" in the middle that connect other axles to that second one?
One thing that strikes me, from the front-view blueprint earlier in this thread, is how high the middle cylinder is. That surprised me. In the drawing above, I think I can see a rod coming down at the correct angle, but I lose it once we get to the front of the first driver. But I think I may see the crank and pin (if those are the right terms) on the second axle, in dotted lines, with the connection-point facing to about 10:00 o'clock. Is that correct?
LithoniaOperator On the 3-cyl engines, did the middle cylinder connect only to the main driver, or was there a connecting rod to other axles?
On the 3-cyl engines, did the middle cylinder connect only to the main driver, or was there a connecting rod to other axles?
.
It turns out that my above “remembered” (not) description of a locomotive pictured in the book was really a conflation of two or three engines, and not a Glehl design . . .
Sorry. The jury will disregard that remark.
BUT, there are lots and lots of photos and illustrations of really fascinating engines. Great book.
Overmod You could, with money enough and time, do multiple expansion on a locomotive. The last of the D&H compound drag locomotives was this: triple-expansion from something around 450psi throttle pressure, four cylinders (at the corners of the driver wheelbase!) and remarkably efficient operation... at about 5mph. Nifty for Red Edge firing while operating a long, long string of anthracite hoppers, but perhaps not the best for making actual ton-miles with expensively engineered and complicated equipment. (Compare the four-cylinder replacement class, which were Challengers ...)
You could, with money enough and time, do multiple expansion on a locomotive. The last of the D&H compound drag locomotives was this: triple-expansion from something around 450psi throttle pressure, four cylinders (at the corners of the driver wheelbase!) and remarkably efficient operation... at about 5mph. Nifty for Red Edge firing while operating a long, long string of anthracite hoppers, but perhaps not the best for making actual ton-miles with expensively engineered and complicated equipment. (Compare the four-cylinder replacement class, which were Challengers ...)
Ah, the D&H 1403, named L.F. Loree as one of the subjects in the last installment of "Consolidations, Inc" which appeared in the June 1967 issue of Trains. Also famous for being the only steam locomotive built in 1933. High pressure cylinder was under one side of the cab, medium pressure was under other side of the cab and the two low pressure cylinders were up front.
Boiler ws rated for 500psi, along with the 1402, with 400psi on the 1401 and 350psi on the 1400. Not sure what the 1403 was capable when "steam was used extensively", but the 1402 was held back by a merchandise train while climbing a hill, while the 1402 was hauling a coal drag.
With regards to the book, my copy should arrive Saturday and am looking forward to it.
UP 9000 was built all right, the first of a series of three-cylinder 4-12-2's the Union Pacific purchased from ALCO. In the 1920's ALCO was very enthusiastic about three cylinder locomotives and pushed them hard, but most, if not all railroads were less than enthusiastic about them, to say the least. The UP 9000's were the most successful American adaptation of the three cylinder concept, but in the end the UP didn't bother with any more three cylinder engines.
UP 9000 survives today in the county park that UP 4014 was removed from.
Thanks so much, Jones1945. That's all quite interesting. I've learned a lot today. That final photo confirms my suspicion that there probably were no rods to other drivers. Two more cranks would be required, I would think, for balance, and that sure would get tight in there. Although I guess if you offset the middle cylinder and its main rod a bit, then you could use one only one side rod and achieve balance. ??? I wonder if that was ever tried. I can't tell if UP 9000 has a "middle side" rod, but I am guessing not. The middle cylinder is indirectly connected to other drivers already, via the normal side rods. Was that engine actually ever built, or was it just a an idea?
Thank you LithoniaOperator for the book recommendation and Overmod's thorough explanation.
The LMS Coronation Class is one of the examples of the 4 cylinder steam engine, which was the most powerful streamlined engine in the UK at the time. This class had only 40,000 lbf tractive effort, which was 11.15% less than the PRR K4s.
http://wondersofworldengineering.com/streamlined-expresses.html
The following video is a 3D animation simulate the mechanism of GWR King Class steam locomotive, another 4 cylinder locomotive in the UK:
Perspective view of Union Pacific 9000 series, a 3 cylinder engine by Alco:
https://www.etsy.com/listing/522314595/union-pacific-4-12-2-type-3-cylinder
"A driver set from a 3 cylinder locomotive"
Not sure if this is the driver of UP-9000 series...
Login, or register today to interact in our online community, comment on articles, receive our newsletter, manage your account online and more!
Get the Classic Trains twice-monthly newsletter