I’ve long been intrigued by 3-cylinder simple steam locomotives, such as IHB U-4a’s, SP 4-10-2’s, UP 4-12-2’s, etc. What was the theory behind their design and why were they built?
Google, your best friend.
http://www.google.com/search?hl=en&q=three+cylinder+steam+locomotives&btnG=Search
Rich
Interesting. Perhaps some one has this book?
Three cylinder simple expansion steam locomotives were built for two complementary reasons:
-
Smoother torque - think automotive engines, 4 vs 6 cylinders.
-
The ability to fully utilize the output of a large, efficient boiler without huge cylinders and over-heavy reciprocating parts.
The IP’s list didn’t include the most numerous and widespread 3-cylinder steam locos - Shays.
Chuck
ie a power stroke every 60 degrees of axel turn vs 90 degrees for a two cylinder engine. Also not as much pounding on the rails
Yes, a great advantage–until the valve gear on the middle cylinder fails. I believe that some engines built with three cylinders were rebuilt with only two because of the difficulty of maintaining the middle cylinder.
Johnny
I remember reading that Mallard, after setting its speed record, had to be pulled off the train due to damage to the middle cylinder…IIRC there was a design flaw that caused the middle cylinder to do too much of the work at high speed and this caused the big end (whatever that is) to fail or overheat.
The problem referred to with Mallard – and some other types of 3 cylinder steam engines – does have to do with the valve gear on the middle cylinder, but it’s not a design flaw. Some 3 cylinder types used what is termed ‘conjugated’ valve gear for the middle cylinder – that is, the middle cylinder valves were controlled by a rather complex linkage from the two regular valve gears (which could be anything – Walschaert’s, Baker, what have you). Worked fine and got around the problem of where do you put the third valve gear. However… if it got worn, it was quite possible that it would begin to overtravel and the middle cylinder would take much more of the load than the outer pair. This could give trouble with the big end (the bearing where centre main rod joins the driven axle). Other 3 cylinder designs used three separate valve gears, and didn’t have that problem.
one other advantage i suspect is if any cylinder failed a mechanic could isolate it and the engine limp in to a repair roundhouse.
Two cylinder locos could get home on one cylinder as long as care was taken not to dead-center the one active cylinder. A N&W Class J 4-8-4 that threw an eccentric rod ‘limped’ back to Roanoke at a reported 85mph!
As for Mallard, the problem bearing was the main bearing on the cranked driver axle, which couldn’t be properly lubricated. There was actually a stink bomb attached to it, which would let the cab crew know when it was overheating. Fairly early on in the record run, the stink bomb went off. The driver* thought that the record was more important than the integrity of the bearing and maintained full power. Once it had a chance to cool, the bearing seized.
*I know that the locomotive operator is called an engineer in the US - but Mallard was operating east of the Atlantic.
Chuck
All the 3 cylinders locos in Britain designed by Nigel Gresley (designer of “Mallard” and “Flying Scotsman”) used conjugated valve gear, but almost all others, such as the LMS “Royal Scot” 4-6-0’s and “Duke of Gloucester”, the final Pacific built in Britain had 3 sets of valve gear.
During WW2 and after there were a lot of maintenance problems with the conjugated valve and some Gresley engines were rebuilt by his successor, Thompson with 2 cylinders or with 3 sets of valve gear. An example of the latter was Gresley’s original prototype Pacifc “Great Northern” which Thompson rebuilt and became the prototype for the later A1 class of which class the new loco “Tornado” belongs. This too has 3 sets of valve gear.
UPs 4-12-2s were built with Gresley valve gear. Some of the earlier ones, with plain bearings in the third cylinder valve motion, had the conjugating levers replaced with a third set of Walschaerts valve gear. The later ones were built with roller bearings in the Gresley motion, A few of the early ones were rebuilt with roller bearings in place of the original plain bearings. Apparently the roller bearings solved the ‘creeping cutoff’ problem by eliminating the wear, and slack, at the bearings.
Chuck
Speaking of Tornado, has she been run since she was unveiled with her new Green livery? I’ve love to see that, Beautiful locomotive!
I think they may have done something in the 1950’s with the Gresley Pacifics, as their reliablity was improved at that time.
Meanwhile Tornado has now completed her main line trials with flying colours and has now received her coat of LNER Apple green. I hope to have a ride behind her sometime this year - haven’t yet made up my whether to go for a thrash along the East Coast Main Linee (her natural home) or see her do battle with the Settle & Carlisle line!
Besides the 4-10-2 that is on display in Philadelphia (Baldwin # 60000) are there other USA built three cylinder (side rod driven) locomotives in existence at other museums?
Sir Nigel Gresiley (sorry if I spelt his last name wrong) drew up multiple engines with 3 cylinders. His plans were used for many engines such as Peppercorn A1’s A2’s Gresiley A1/3 A4 and many others. Long story short, more power, more speed, and AND OTHERS
First public run is on Saturday (31 Jan)
And to answer the original question on this thread - because four cyliders would be too many 
Hmm - not sure I agree with that. For my money, I think a Stanier “Coronation” Pacific could have beaten Mallard’s record. Certainly the highest power output recorded by a British steam loco was from a Coronation.
As for three cylinder locomotives on display, isn’t the UP 9000 at the LA County Fairgrounds?
Chuck