That's a good reference, which I had not seen before. It is peculiar that Holcroft makes no reference at all to American practice, where some of his 'desirable reasons' for adoption were observed nearly a decade earlier, and where three-cylinder construction would become a major design point for one builder (Alco) stadting right about the time of this paper.
Note the stated difficulties with three cylinders on reciprocating engines are not so much with the inside pistons, but with the inside piston valves. Were a Budd-Besler truck to be made with three (very small!) cylinders, the power-piston access would (I think) be little worse than for the outside cylinders (the heads of which face each other). Presumably the inside valve would be driven by an eccentric as all three valves, being Stephenson, could have common lead change as mentioned.
The chief problem I see -- and I suspect it is related to the real problem with the press crank fits -- is that the axle drive is unconjugated. This is a common issue in the original Besler constant-torque, and that its effects were under appreciated can be seen in the example of the PRR duplex 'double Atlantics' which were notoriously slippery with twice the conjugation.
Personally I think there was not enough consideration of the practical unloading of axles due to shock or track geometry. In the PRR duplexes it was found that an event like a low rail joint or momentary adhesion fault could greatly decrease the effective adhesion of an individual drive wheel. Were this wheel conjugated (via rods) with 5 or 7 others this would be comparatively less of an issue, but when there are only three, with high horsepower capability in the engine, the chance for a slip even at ridiculously high nominal FA is still extreme, and if the valve gear gives accurate events up to high cyclic you'll have the risk of high-speed slipping comparable to those Franklin results of '140mph' or higher reported in Vernon Smith's One Man's Locomotives. The risk is worse in locomotives with individual axle drive (see the complex arrangement on a Roosen motor locomotive to see how the Germans addressed the issue). In that age, effective detection of slip for traction control was difficult and fairly expensive, and response to detected slip usually could not involve proportional throttling -- note the critical mistake on the slip-reduction system applied to the Q2s, which was implemented as bang-bang witb a variety of poor results and still had maintenance issues with sticking.
I think a dominant part of the Budd-Besler power truck was cost. The very logical solution to some of the issues was to modify the motor design for the constant-torque to be geared-horizontally-opposed, flanking the axle with its outside frame and bearings as in the W-1, with the mass being spring-borne at front and rear to relieve some of the effect of unsprung mass. This would act much like a geared version of the Paget locomotive; the balancing concerns are very much like those described for the constant-torque, and at railcar scale the two motors could be reasonably self-contained without sacrificing the ability to drop either the axle or the whole engine unit easily for service. The catch is the plethora of small parts, most requiring some attention in lubrication, and the number of expensively-ground gears. Nonetheless I think the design would be at least comparable to the Sentinels of this period.
I am as yet ignorant of most of the technical details of the Doble high-pressure work in Germany in the latter '30s; it is not well covered in the SACA volumes on Doble, which I found unfortunate. If you want to look over adapting the de Dietrich steam plant to railcars, I have copies of all the fabrication drawings...
This brings us to the somewhat poignant issue of the power trucks on Bulleid's Leader. The valves here were perhaps the ultimate in clean port area, quick effective opening and closing, and low dead space; I think American practice could have addressed most if not all the tribology issues reported (although the cylinder block might have required jacketing or tracer-line heating for the valves to run effectively). Note that Bulleid had all three cylinders plus sprocket for the timing drive arranged on one axle between the wheels, thanks to outside bearings.
American loading-gage clearances would easily permit outside chain conjugation, overcoming some of the issues with how Bulleid had to arrange things on Leader. Whether chains would be a fully-workable solution over time would remain to be seen; it would be interesting to see other approaches to the issue.
The Besler Budd steam truck was designed for low center of gravity and motorcar application...
On another note, seem to remember a 3 piston locomotive design, having both inside and outside pistons. Things were fine until having to work on the inside piston...
http://kesr-mic.org.uk/resources/The+Modern+3-Cylinder+Locomotive.pdf
Former Car MaintainerAlthough the W1 was never built, it appears moving the pistons inside the wheels were an improvement over the Budd Besler unit that had exterior pistons and weak pressed on cranks...
The outboard cranks avoided the need to machine (or worse, fabricate) a cranked axle for that service. In my opinion, the cranks could have been given offset keys secured with an analogue of Superbolting if the issue with pressed cranks could not be resolved with better shop practice. As you renember with the dry ice and heat, the right combination could be applied to axle fit and crank, with the key serving as an effective guide holding theoretical correct quarter while the crank was driven on to correct depth, then the pins finish-machined "on a quartering machine" to be right.
There were detail-design flaws in the railcar set, perhaps carried over from its pedestrian beginnings. From what I remember of the Besler truck, its 'advertised' 70mph capability was right about what the suspension could safely handle; personally, I think much better secondary suspension would have been needed, perhaps with better lateral control as well, over that power truck.
Bought a copy of Swanberg to check; we'll see what he says. Literally 5 minutes after I placed the order, someone listed on Amazon for 25% less. Ah, the pains of research!
Nice pic....always have trouble posting pics to the site.
The purported breakdowns for this train were outlined in the book, "New Haven Power, 1838-1968" by JW Swanburg....of which I don't have a copy...can anyone help out with what it says...?
To visualize what is being talked about
Never too old to have a happy childhood!
Great info...Besides design errors, I am still seeking info on what kept breaking and the resultant repairs to the Budd Besler unit...I seem to read a lot about them continuously being broken but not much about why...
Although the W1 was never built, it appears moving the pistons inside the wheels were an improvement over the Budd Besler unit that had exterior pistons and weak pressed on cranks...
The idea of having exterior pistons, valve gear and gearing to make it "easier to service" may be an oxymoron for being "serviced more often" ....
Former Car MaintainerBTW, this is the W1 "V" design to which I referred
The patent number I provided, as well as the illustration, are taken from an online reference quoting the article -- I did this to facilitate locating the reference, as I don't maintain a Web site or image stash that furnishes URLs the current Kalmbach software will recognize. I doubt, with their often expressed concern over unattributed images and material, that their 'new and improved forum experience' will feature any inline embedding at all.
I have not looked carefully at the rail car patent drawings yet, but it was my impression that the powered axles in it were not conjugated mutually some multiple of 45 degrees apart. If not, it's likely to me that slipping would tend to put the engines in phase (comparable to the effects with larger Mallet-pattern articulateds) and much of the eight-pulse-per-revolution smoothness then reduced in practice. I do not know if that was observed in operation of the railcars, but it would have been difficult to check both by inspection and in operation.
If you have detail drawings for the flexible connections between the W-1 motors and the throttle arrangements, I'd like to see them. That would be a critical detail, and while I can speculate on ways it could be done, I don't recall seeing how they intended to build it.
Here is the Google Patents reference to the later patent (for some reason Google won't provide the whole set of images from its PDF download).
https://patents.google.com/patent/US2235957A/en
Note the very careful discussion of balancing, and the very sensible handling of the torque moment in a way that does not interfere with suspension action in fixed pedestals (the original Arlen patent for what would become the PRR S2 thinks it has done this by the same method used for nose-suspended motors but it doesn't quite work geometrically)
For some reason I strongly remember the W-1 engines being single-acting, likely to eliminate condensation issues on the 'top' side of the cylinders where it might be difficult to provide cylinder-cock or relief arrangements. Facilitating condensate clearance would also be a beneficial consequence of inclining the cylinders in the 'block'. Note how cagily the '957 patent avoids showing this (although DA is suggested by how the steamlines, probably coaxial in these drawings, go into the valve chests).
You have the wrong patent, wrong design
https://www.google.com/books/edition/Railway_Age/a3lCAQAAIAAJ?hl=en&gbpv=1&dq=besler+steam&pg=PA581&printsec=frontcover
https://video.search.yahoo.com/search/video?fr=yfp-t&ei=UTF-8&p=Besler+steam+video#id=6&vid=3e5dd1fa3d16c6f685ad1ad05b8559c2&action=click
https://patentimages.storage.googleapis.com/74/a3/bc/73987f08b268c0/US2155781.pdf
in any event, suspect the pressed on cranks were problematic as well as the frequent stops and grades the New Haven required. Also the pneumatic controls for the Stephenson valve gear were sketchy. However the 8 strokes per wheel revolution helped equate to the action of a 8 pole motor, helping to aid smooth acceleration, adhesion....
BTW, this is the W1 "V" design to which I referred
https://patentimages.storage.googleapis.com/pages/US2235957-1.png
Also in my opinion, the 'better' use of a Besler high-pressure approach eould be to frame-mount the motors, with rigid steam and control connections, much as the engines in a RDC are mounted, with comparable drive (ideally via the Bowes-drive railcar variant as seen in the collection at ISM in Philadelphia) using Cardan shaft to axle drives with spiral-bevel, hypoid, etc. optimized to use counter pressure braking in backdriving. That would give you high-speed and high acceleration torque with very little complexity other than the self-contained Bowes unit (which also coordinates reverse outside complicating the steam-engine valve gear) together with very effective dynamic snd blended braking to simplify issues with friction brake proportioning, integrity, overheating and shoe/pad and wheeltread life; not incidentally also allowing both easy towed operation and electric station access with the flash boilers idling, both of which were important potential considerations in the East.
Former Car Maintainer The W1 used a "v" piston design mounted above the wheel, to transfer the downward thrust to the wheel - rail surface.
I think you're remembering the Roosen motor locomotives, which used comparatively large V-2 engines alternating outside the gauge entirely, acting inward on adjacent wheel spokes in a one-ended version of quill/cup drive. In my not-so-humble opinion this was a much more promising approach than the Beslers' even if it needed better conjugation of the physically-independent axles to get the effect of the 'constant torque'. One of our great national shames was cutting up the locomotive we brought over here as 'reparations' after the War, merely for political Korean-war 'optics'...
I have wondered whether the Paget locomotive drive (with better materials and a workable valve system) would be a better thing to try in a constant-torque express locomotive. That is a comparatively very-short-stroke design with relatively little rod angularity (and the angular vertical moments on the axles balanceable) with most of the thrusts longitudinal and affecting augment comparatively little. With four axles and somewhat larger wheels, and maintaining the outside-bearing frame, a very interesting and stable locomotive with great modularity in the drive and steam connections could have been developed, and worked at as high a pressure as the Emerson watertube box could cost-effectively utilize...
i wonder if the problems you mentioned would have been eliminated if the railroad had accepted a Zephyr like installation for the Besler as originally proposed by Budd? The W1 used a "v" piston design mounted above the wheel, to transfer the downward thrust to the wheel - rail surface. The Besler Budd used a horizontal thrusting pistons. And as I recall a sealed, oiled gearbox that should have kept clear of rail dust...
Part of the issue with it, in my opinion, was that it had a very long wheelbase and relatively primitive suspension and guiding. I will have to see if I can find good illustrative pictures of the arrangement -- they do exist -- as it has been many years since I looked at this.
As I recall, the little engines in the trucks shared some of the detail design with the engines for the B&O W-1 constant-torque locomotive, and would have been susceptible to a range of shock and contamination issues in service on typical dusty jointed rail and environmental conditions.
There seem to be many designs of railcar and lightweight train construction (the original McKeens and the Baldwin RP-210 for 'Train X' being early and late examples) that carried complex reciprocating machinery on a truck frame with only relatively hard primary suspension. I'm still looking for the actual Locomotive Drive 1 and 2 Westinghouse patents for the PRR S2 chassis, as there were detail design changes from Alben's original that may have produced similar unsprung-weight concerns in the inner-axle gearcase design as built; a critical issue of concern being whether counter-load springs were provided at the outer ends of the 'balanced' trunnioned gearcase to lower its vertical effective polar moment of inertia and limit the effective unsprung mass effects on the suspension of the middle axles (equalization of which might be an interesting dynamical design exercise in itself, as the locomotive was not initially designed with Timken rods between #2 and #3 driver pair, and therefore some of the suspension 'lessons' from the first T1s might have been seen as relevant even with effective 'conjugation' of the driver groups).
Remember the streamlined train, that was an original railcar, converted to a streamliner by Budd, with a Besler Steam truck? Besides being too heavy, what were the other major maintenance problems with it? Not too much has been mentioned about the flex coupling between the flash boiler and steam truck. The short 7" stroke of the compound steam pistons, and the 8 stroke per wheel revolution showed promise for speed, tractive effort , etc....
Our community is FREE to join. To participate you must either login or register for an account.