Trains.com

american four cylinder locomotive

1874 views
12 replies
1 rating 2 rating 3 rating 4 rating 5 rating
  • Member since
    May 2011
  • 122 posts
american four cylinder locomotive
Posted by IA and eastern on Sunday, January 31, 2021 2:18 PM

On an american four cylinder would the cylinders be on the outside or would two cylinders be on the inside as on a three cylinder locomotive. Gary

  • Member since
    January 2019
  • From: Henrico, VA
  • 6,402 posts
Posted by Flintlock76 on Sunday, January 31, 2021 4:57 PM

Omitting the duplex-drive concept where all four cylinders would be on the outside (but not in the same locations) or other experimentals that didn't amount to much I don't see how an American four cylinder engine could have ignored the European approach and have not located the two additional cylinders inboard of the other two. 

I can't think of anything else that might have worked with any degree of practicality.  

  • Member since
    December 2001
  • From: NW Wisconsin
  • 3,784 posts
Posted by beaulieu on Sunday, January 31, 2021 6:08 PM

IA and eastern

On an american four cylinder would the cylinders be on the outside or would two cylinders be on the inside as on a three cylinder locomotive. Gary 

They were mounted in just about anyway you can imagine. There were "Tandem Compounds" with the cylinders mounted one in front of the other sharing the same piston rod. Another type of four cylinder steam locomotive was the "Vauclain Compound" named for Samuel Vauclain the head of locomotive production at Baldwin Locomotive Works. These had the high-pressure cylinders mounted below the low-pressure cylinders outside the frames.

  • Member since
    January 2002
  • 4,246 posts
Posted by M636C on Sunday, January 31, 2021 11:12 PM

Flintlock76

Omitting the duplex-drive concept where all four cylinders would be on the outside (but not in the same locations) or other experimentals that didn't amount to much I don't see how an American four cylinder engine could have ignored the European approach and have not located the two additional cylinders inboard of the other two. 

I can't think of anything else that might have worked with any degree of practicality.  

 

There is always the "L.F. Loree" arrangement.

It could be set up as a simple rather than a compound...

American High-Pressure Steam Locomotives (douglas-self.com)

(scroll down)

I could imagine the "George H Emerson" with the duples pairs coupled by conventional rods, and converted to burn oil to save the abrasion from coal dust at the rear cylinders.

Duplex Drive Locomotives. (douglas-self.com)

(scroll down)

You have a four cylinder 4-8-4...

Peter

  • Member since
    September 2003
  • 15,997 posts
Posted by Overmod on Monday, February 1, 2021 9:14 AM

M636C
I could imagine the "George H Emerson" with the duples pairs coupled by conventional rods, and converted to burn oil to save the abrasion from coal dust at the rear cylinders.

That is not dissimilar from the ACE3000 approach (with the 'coupling by conventional rods' following Withuhn's approach to conjugation) -- that design retained both 'run-of-mine-coal' firing and compounding.

The only real 'abrasion' issue on the B&O duplex was piston rods, and my suspicion is that it was nicking of the rods rather than progressive wear to the glands.  The solution in the '80s was to have been to use a 'boot' (like that on a long-excursion 4x4 shock absorber) with appropriate condensate and lubricant handling.  As noted before, SP used 'back-pedaling' engines for many years and didn't report particularly showstopping problems with piston rods -- of course that was at much slower speeds.

I was an early fan of the de Glehn-du Bousquet arrangement which divides the drive longitudinally, with the HP normally on the outside (where inspection is easier) and the LP displaced forward (which makes it easier to fit larger cylinders) acting on the forward driver axle.  This was tried early, but the engines were apparently built too light for our service, and no one chose to build one 'strengthened'.  

The original Vauclain compound (as used on the fastest locomotives in the world just before the turn of the 20th Century) was a mechanically clever arrangement that put the HP cylinder above the LP cylinder (very seldom was the HP under the LP, generally for reasons of space, as on Manitou & Pikes Peak), driving an outside main through a common crosshead.  This was an early use of a piston valve, with preternaturally contorted porting to do the job of admission and release to both cylinders on one side with a single spool; the cylinder blocks are a triumph of casting.  In practice the differential pressure on the LP in service 'cocked' the crosshead against the guides and the piston rods against the glands, etc.  which in that era of tribology meant serious maintenance work.

So Vauclain's second compound was a balanced compound, like Cole's (and a number of others) where all four cylinders were in 'line abreast'.  There was less limitation on cylinder diameter in this arrangement than there was in Britain for four-cylinder simples, often favored for balance (as civil had far more influence on management over 'permanent way' issues with management there).  I recommend if you're interested in these that you download Colvin's "American Compound Locomotives" via Google Books or Swingle's "Standard American Locomotive Engineering" from the Library of Congress.

If there was a horizontal tandem compound in this country that worked, I'm unaware of it.  The idea works quite well in steeple-compound high-speed engines and the like -- not so much on large external reciprocating cylinders, where the whole mass of the two piston heads and extended rod counts in the reciprocating mass, all the thrust has to be taken through one main and pin, and the sealing and head arrangements between HP and LP will reduce you to tears -- even if great cleverness is used (ATSF actually mounted a small crane on one of theirs to facilitate dealing with this!)

Not all four-cylinder engines in the United States were compounds -- one notable attempt was the Shaw locomotive, which (like some British designs) used external cranks and a pair of parallel cylinders side-by-side (and nominally 180 degrees opposed) with an outside frame for the crank center bearings.  I had some fun with a low-augment high-speed design that doubled the cylinders vertically (so you had eight per side, 16 DA) -- this was a bit like Satan's answer to the Besler W-1 or the Paget locomotive, but I thought at the time that both the vertical and transverse augment would be very small, easily handled with active mass in the suspension.  (This was about the time I discovered jacketing with overcritical water and use of thermal barrier coatings, if you're wondering how such a thing could survive its own differential thermal expansion.)

There's a lot to be said for noncompounded four-cylinder conjugated duplex drive if you want a high-speed express locomotive, but then again a modified Turbomotive 2 or asynchronous compound accomplishes the same thing with far less fuss and more available 'top end'.

  • Member since
    February 2005
  • 1,979 posts
Posted by timz on Monday, February 1, 2021 10:42 AM

Guess the first 4-cyl engines in the US were Vauclain compounds, with low-pressure cylinders above or below the high-pressure, on the outside of the engine's frame. Does the name mean Baldwin was the only one that made them?

Maybe tandem compounds were next, with the axes of the HP and LP cylinders coinciding. HP was always ahead of LP? They were always slow freight engines?

Balanced compounds (HP cylinders inside the frame) started ... around 1905? A few 4-cyl balanced simple engines were built; no one got more than a couple of simples, but 4-cyl balanced compounds weren't rare -- SFe had hundreds: 4-4-2s, 2-6-2s and 4-6-2s.

Wonder when the last 4-cyl engine in the US ran -- were they all scrapped, or converted to 2-cyl by 1929?

  • Member since
    January 2019
  • From: Henrico, VA
  • 6,402 posts
Posted by Flintlock76 on Monday, February 1, 2021 11:44 AM

I don't think the Vauclain compounds lasted too long after the concept of superheating was proven.  While those compounds worked they were pretty maintanance intensive, and American railroaders were alway very averse to maintanance intensity. 

  • Member since
    September 2003
  • 15,997 posts
Posted by Overmod on Monday, February 1, 2021 1:30 PM

Flintlock76
I don't think the Vauclain compounds lasted too long after the concept of superheating was proven.

The original kind were on their way out long before Schmidt figured out practical superheating; I suspect this was at least part of the reason Sam went to balanced compounding at Baldwin and did not change the name.  

Note that you can't adjust a Vauclain compound the way you can a de Glehn-du Bousquet engine, to equalize thrust in the HP and LP.  Theoretically (very theoretically) you could rig up some sort of riding cutoff that would vary admission on the LP ports to match what the HP was developing (and this would be done 'by feel', so changes in HP performance would be balanced too) but I could not figure out a way to do this with the valve as Baldwin constructed and placed it.  While the task may be no easier with the revised arrangement (I believe the common piston valves and complicated passage coring were retained) certainly the differential-wear issues were solved.  

While those compounds worked they were pretty maintenance intensive, and American railroaders were alway very averse to maintenance intensity.

Famously, compounding in the presence of superheat, which seems almost like a no-brainer for economy, was tried very little in the United States in the decade after superheat was demonstrated to 'work' -- which is great support for your position.  And there does not appear to be a lack of experimentation with both superheat and compounding in the early years of the Superheater Company ... perhaps the combination of the licensing fee and cost of the superheater plus the added cost and complication of misunderstood compound expansion did the business effectively enough!

Even more, very little of the use of three cylinders, even simple, even though as stoutly advocated by Alco a few years later as 'the answer' to increasing engine size and power, did not; certainly Baldwin's bold bet in the late '20s that Caprotti poppet gear would be the wave of the future almost completely tanked.  

Where the game starts to look interesting is in maintenance issues involving rolling-element bearings.  As far as I know, anywhere Multirol bearings were tried in place of plain bushings, e.g. in valve gear, they succeeded.  But comparable use of roller bearings is not as clear-cut:  NYC, which understood them well as a significant part of the Niagara's success, only tentatively applied them to any L-class Mohawk ... and never repeated the experiment at any point during the succeeding years, either for performance or lower maintenance.  N&W likewise put Timken rods on the class As but never considered the arrangement for the compound 2-8-8-2s.  Union Pacific, which tried the Timken rods early (on a Pacific and a Mountain type) never really considered roller rods at all for their high-speed 4-8-4s which were operated as fast in service as any of that wheel arrangement.  Interesting conclusions on perceived cost/benefit can be drawn from these and other examples.

  • Member since
    September 2003
  • 15,997 posts
Posted by Overmod on Monday, February 1, 2021 1:58 PM

timz
Guess the first 4-cyl engines in the US were Vauclain compounds, with low-pressure cylinders above or below the high-pressure, on the outside of the engine's frame.

The Shaw, and I think some other kinds of balanced engine design, considerably predate the Vauclain compound.  You might look to see if the PRR experiments with de Glehn-du Bousquet effectively predate extensive use of the Baldwin arrangement -- but there's little doubt that the perceived advantages of the Vauclain arrangement, while they remained obvious, would make the much greater investment of machinery and training to use it effectively for a de Glehn type less attractive.

Does the name mean Baldwin was the only one that made them?

Baldwin had a number of key patents on the arrangement, notably on the valve design.  Making that cylinder arrangement work with slide valves would be comical in a number of respects, especially some of the passage lengths; the situation is not much improved if you allow slide valves working in other than horizontal seating (as many English locomotives of the period used).  Whether Baldwin would have licensed the arrangement to other builders at all is unclear; in the event, the amalgamated American Locomotive Company pursued its own distinctly different solutions.

Maybe tandem compounds were next, with the axes of the HP and LP cylinders coinciding. HP was always ahead of LP?

This may depend on other things.  One of the engines tested on the original PRR plant as erected at the St. Louis exposition was a Santa Fe 2-10-2 with the tandem-compound arrangement and considerable test data were recorded for it -- I don't have a copy of the results but I believe it's now easily obtained on line for study.  It was not considered either a failure or particularly ominous maintenance nightmare in 1904.

Where you put the HP vs. LP probably involved the steam circuit more than having the smaller mass 'outboard' -- the great trouble as I imperfectly recall being the gland between HP and LP when the cylinders were 'combined' to save mass.  There is relatively little additional mass in having a longer piston rod and two heads; the 'catch' is how you make them both lightweight and strong and still have ease of access to the rings and bores when you have to, and the 'problem' is that it's all reciprocating weight, which can't be easily balanced at all speeds.

They were always slow freight engines?

The 'practical' ones certainly were.  I have seen at least one faster locomotive designed to use the arrangement, and certainly no few engines, like the Heilmann locomotives, used steeple compounding (which is just tandem or triple compounding reduced in size and turned upright) so there's nothing inherent in the balance that would preclude use for 'somewhat' higher speed.  I suspect the great issue with them involved necessary phasing of HP vs. LP admission and exhaust and passage length, but this was scarcely worse than the situation for Mallets.

Balanced compounds (HP cylinders inside the frame) started ... around 1905?

I believe earlier -- see Colvin (at least one edition is 1903 and has some extensive discussion of balanced designs at that time)

A few 4-cyl balanced simple engines were built; no one got more than a couple of simples...

That's no different from what Shaw encountered earlier: why spend the additional money if you don't have powerful allies mandating smoother performance on the track and higher speed together?  

... but 4-cyl balanced compounds weren't rare -- ATSF had hundreds: 4-4-2s, 2-6-2s and 4-6-2s.

And not all of them used common compounding systems -- some of them being highly imaginative.  (Interestingly I believe they actually had two Player compounds ... from different people named Player; how many of those could there be!)

Note how fast all of them disappeared after practical superheat was introduced ... this in the presence of ATSF experimenting more, I suspect, than anyone with different arrangements of "superheating" and steam-drying in the years before and even during the early Schmidt adoption.

Wonder when the last 4-cyl engine in the US ran -- were they all scrapped, or converted to 2-cyl by 1929?

For the balanced compounds, I think the answer is in Worley's Iron Horses of the Santa Fe Trail somewhere -- I'll have to look when I have the time.  I can't imagine a road other than Santa Fe retaining balanced compounds longer, and I think 1929 is very conservatively late.  Of course the date of the last 4-cylinder engine running would probably be the last service date of a T1... but that's not what you meant.

  • Member since
    July 2004
  • 2,582 posts
Posted by Paul Milenkovic on Monday, February 1, 2021 2:44 PM

Owing to current circumstances, I have been hitting the Renew button on my loans from the "U", and the Alfred Bruce book remains on my nightstand.

Bruce explains that the "tandem compound" became maintenance expensive, both from the need to renew the gland isolating the HP from the LP cylinder and also the need to take the whole, fine thing apart to do that.  Bruce points on the "crane attached to the smokebox" (essentially a metal arm that pivots outward) to aid in that task.

If GM "killed the electric car", what am I doing standing next to an EV-1, a half a block from the WSOR tracks?
  • Member since
    July 2004
  • 2,582 posts
Posted by Paul Milenkovic on Monday, February 1, 2021 4:26 PM

There is that sole-remaining Swedish steam turbine locomotive that uses some form of mechanical jackshaft drive from turbine to siderods to wheels.

M3t nr 71 - Bing video

Look starting at 16:45 for a view of the drive arrangement.

The problem is that the rod drive mechanism of a steam locomotive is already overconstrained and hence sensitive to binding if the drivers are the least bit "out of quarter" or if the frame is the least bit skewed from either assembly tolerances or frame damage.

The Stanier-designed "Turbomotive" on the LMS railway in England along with the Pennsy S2 turbine had a quill drive, to one drivers for the Turbomotive and to two drivers with the S2?  The other drivers are connected with siderods, which have the further advantage that they don't need (as much) reciprocating balance, although Overmod pointed out that the drive arrangement wasn't in perfect balance owing to the effects of different rotation planes.

The Swedish M3t turbine locomotive, however, uses a jackshaft drive -- the turbine drives a "dummy wheel" (the jackshaft) that is connected by siderods to the actual driving wheels of the locomotive.  To me, such a jackshaft could be turned by a high-speed multi-cylinder steam engine in place of a turbine.

Alfred Bruce (who else?) covers this type of drive used in steam trams and low-speed street-railway switch engines, and his criticism of it is binding and friction and wear in the drive.

I guess at high speeds and as the locomotive body bounces in relation to the drive wheels with its springs, this introduces a small angularity between the jackshaft wheel and the first driving wheel?  If that is a problem, why isn't the angularity of a driver hitting a joint in the rail, displacing it in relation to neighboring drivers not a problem?  Is it that the deflection of drivers relative to each other, for reasonably smooth track, much smaller than the "bouncing" deflection of the jackshaft attached to the locomotive in relation to the drivers in contact with the rail?

On the other hand, is the turbine and its jackshaft on the Swedish locomotive more akin to an axle-suspended traction motor in that it is in mechanical contact with the rails? A person would need to study the video carefully to see what is going on.  Also, this locomotive may be intended for slow-speed operation and may ride on stiff springs?

If GM "killed the electric car", what am I doing standing next to an EV-1, a half a block from the WSOR tracks?
  • Member since
    January 2002
  • 4,246 posts
Posted by M636C on Tuesday, February 2, 2021 1:29 AM

The Swedish M3t turbine locomotive, however, uses a jackshaft drive -- the turbine drives a "dummy wheel" (the jackshaft) that is connected by siderods to the actual driving wheels of the locomotive.  To me, such a jackshaft could be turned by a high-speed multi-cylinder steam engine in place of a turbine.

I guess at high speeds and as the locomotive body bounces in relation to the drive wheels with its springs, this introduces a small angularity between the jackshaft wheel and the first driving wheel?  If that is a problem, why isn't the angularity of a driver hitting a joint in the rail, displacing it in relation to neighboring drivers not a problem?  Is it that the deflection of drivers relative to each other, for reasonably smooth track, much smaller than the "bouncing" deflection of the jackshaft attached to the locomotive in relation to the drivers in contact with the rail?       (Paul Milenkovic)

The Swedes were very attached to jackshaft drive for electric locomotives. Until the 1970s, they preferred a jackshaft drive to coupled axles for heavy freight traffic, particularly on the Narvik iron ore line.

SJ Dm3 - Wikipedia illustrates a sort of electric Triplex which lasted until the introduction of current variable frequency AC traction motor technology.

High speed multi cylinder steam engines tended to use flexible drive to single axles as shown by French High-Pressure Locomotives. (douglas-self.com) (scroll down to 232P1 entry) and Steam Motor Locomotives. (douglas-self.com). The never completed B&O W-1 followed this pattern (page 210 of Edson's Steam Locomotives of the B&O).

It appears that the B&O had jackshaft drive locomotives as early as 1832...

Jackshaft Locomotives. (douglas-self.com)

Peter

  • Member since
    July 2008
  • 754 posts
Posted by Juniatha on Thursday, February 25, 2021 9:05 PM

Hi Gary

You might have mentioned what you want four cylinder for?

One thing is for sure: it won't be easy to fit in!

American locomotive design is so much developed around and for the two cylinder engine that you have to change a lot to accomodate inside cylinders - and we are not talking of double cluster of cylinders outsides asides each other, neither am I talking of cylinders both ends of the coupled wheels. Cylinders belong to near the smoke box - full stop. A three cylinder engine would be much easier to realize.

It is a paradox situation: although heavy engines would in the first place more ask for multi-cylinder designs, they are at the same time the ones where it is more difficult to realize. Also, cylinder size comes to a practical limit and in a four cylinder simple expansion engine it is hardly possible to use identical cylinder dimensions inside and out.

Oh, no, let's drop that thought as fast as it came up: no staggering of cylinders and two drive axles inside - no! Those wierd forces evoked by such a non-symetric design!

It is a different proposal if four cylinder engine(s) are combined with a Duplex arrangement: this way cylinder dimensions come down to handy sizes and piston forces are also milder. Still the old European aesthetical and technical ideal of the deGlehn compound (with inside LP) will hardly be realized.

Except you dare to explore the really high boiler pressures: 400+ psi ...

Juniatha

Join our Community!

Our community is FREE to join. To participate you must either login or register for an account.

Search the Community

Newsletter Sign-Up

By signing up you may also receive occasional reader surveys and special offers from Trains magazine.Please view our privacy policy