Jones1945If PRR S1 was the "Big Engine", ATSF's 3776 class and 2900 class are "Mega Engine"...
The S1 is over 608,000lb. That is far beyond any of the modern ATSF engines.
It would be interesting to see data for the proposed ATSF 6-4-4-4 cab-forward duplex, which surely reached the point in design where initial weight balance would have been computed. This was designed at a point where B&O 5600 George Emerson would have been a significant influence, and (hypothetically speaking) since long high-speed runthrough would have been an important requirement for this design, we may presume that the largest size tender would be adapted for it when available... now with some additional streamlining weight. (You will recall this came to over 17,000# for the 3765 class, although I don't remember how that broke down between engine and tender).
Overmod Jones1945 If PRR S1 was the "Big Engine", ATSF's 3776 class and 2900 class are "Mega Engine"... For some perspective, the heaviest ATSF 4-8-4s got was a smidge over 510,000lb (the 2900s, built with wartime materials). The 5011s, with the same 464,700# tender, had over 25000# more engine weight. The S1 is over 608,000lb. That is far beyond any of the modern ATSF engines.
Jones1945 If PRR S1 was the "Big Engine", ATSF's 3776 class and 2900 class are "Mega Engine"...
For some perspective, the heaviest ATSF 4-8-4s got was a smidge over 510,000lb (the 2900s, built with wartime materials). The 5011s, with the same 464,700# tender, had over 25000# more engine weight.
You are absolutely right, Overmod. The S1 was still so much heavier than the heaviest ATSF 4-8-4s, though the TE of Santa Fe's 3776 and 2900 class was much higher. I don't know if 3776 and 2900 class were as slippery as the S1 since the Factor of Adhesion of them were all below 4.0.
Overmod It would be interesting to see data for the proposed ATSF 6-4-4-4 cab-forward duplex, which surely reached the point in design where initial weight balance would have been computed. This was designed at a point where B&O 5600 George Emerson would have been a significant influence, and (hypothetically speaking) since long high-speed runthrough would have been an important requirement for this design, we may presume that the largest size tender would be adapted for it when available... now with some additional streamlining weight. (You will recall this came to over 17,000# for the 3765 class, although I don't remember how that broke down between engine and tender).
The proposed ATSF 6-4-4-4 cab-forward duplex would have been one of the most interesting passenger steam locomotives west of Chicago. It is a reasonable inference that the tender of it would have been at least as large as the 3765 class or even the 2900 class. This abandoned design also reminds me of PRR's Triplex, which would have had a much better streamlining design because of the long fuel tender on the front end.
Jones 3D Modeling Club https://www.youtube.com/Jones3DModelingClub
I see to my interest that steamlocomotive.com lists the 3776-class tenders at an even 466,000 loaded. That is a nontrivial increase over the previous 464,700.
The relatively smaller FA of the late Northern classes will not necessarily make them 'slipperier'; compare the even smaller FA of the N&W J class. This is an artifact of the high boiler pressure and formula; in principle control of these engines would follow the "usual" idea of rapidly shortening the cutoff after starting. To my knowledge these did not have a reputation of being slippery locomotives when run properly.
The S1 is a bit of a different case, in that it shares with the other duplexes the unfortunate lower 'quantization' of adhesion should there be a problem with contact patches. I think the presumption was that with the smaller piston dimensions the propensity to slip would be reduced; the problem was that the degree of propensity to slip was much greater with an unconjugated four-wheel engine.
Having said that, I'd think it would be much likelier for this engine to stall rather than slip in most cases, and I think that is borne out in some of the "slipping" reports: there is enough power to spin one set of drivers, but the other set never breaks away and yet the locomotive fails to accelerate. We now know that stalling much more than slipping characterized T1 road 'failures' (e.g. in testing on C&O) and it would not surprise me to see difficulty in starting a consist that the locomotive would likely have little difficulty accelerating to the high-speed slipping region (which is adhesion-limited in a different way).
Note that the AEM-7s (derived from Rc4s) had a ridiculous FA at starting, far in excess of what they could use -- normal practice was to accelerate slowly up to the point the motor torque could be fully applied. In my opinion it would be difficult to run the S1 this way, with no slip mitigation and no separate throttles or trim for the two engines. It certainly did not help that many starts would be expected in station trackage.
OvermodThe S1 is a bit of a different case, in that it shares with the other duplexes the unfortunate lower 'quantization' of adhesion should there be a problem with contact patches. I think the presumption was that with the smaller piston dimensions the propensity to slip would be reduced; the problem was that the degree of propensity to slip was much greater with an unconjugated four-wheel engine. Having said that, I'd think it would be much likelier for this engine to stall rather than slip in most cases, and I think that is borne out in some of the "slipping" reports: there is enough power to spin one set of drivers, but the other set never breaks away and yet the locomotive fails to accelerate. We now know that stalling much more than slipping characterized T1 road 'failures' (e.g. in testing on C&O) and it would not surprise me to see difficulty in starting a consist that the locomotive would likely have little difficulty accelerating to the high-speed slipping region (which is adhesion-limited in a different way).
Thanks, Overmod. Do you think the Triplex (using duplex engines) would have suffered from the same stalling problem?
Jones1945Do you think the Triplex (using duplex engines) would have suffered from the same stalling problem?
It doesn't appear that the engineers at Baldwin were much concerned with starting conditions on the S1; the slipping, which is a kind of opposite to stalling, seems to preoccupied their concern. That this is an important concern can be clearly seen in the T1 (which had an exaggerated FA to start with which only became higher with early modifications) suffering repeated stall events -- an added issue being that some of the typical ways one might have run the engines to overcome these, for example a technique I've read about which involved setting the independent, opening the throttle, then releasing the brake to 'start with a bang' () will give a duplex an induced propensity to slip one or the other engine.
The interesting thing here was that Baldwin would have specified a shorter stroke if they could; in fact on the T1s the shortness of stroke was limited only by the web strength in the cast driver center between axle and mainpin, and indeed was 'cheated' even there by grinding the mainpins to make the circle for the mains slightly smaller than that for the (fully balanceable) side rods -- you will have noted that this makes the stalling problem worse, whatever it does for balance.
This was the place a good high-speed booster design could have assisted materially. There are a number of places this could be done on a Triplex, especially with a design like the Lewty booster that minimized the unsprung mass for suspension and guiding.
As mentioned, the 'backpedaling' engines could have been made less of a road disaster by providing 'boots' for the piston rods and gland/packing box areas on the cylinders. I think it is interesting that the model Loewy proudly showed in his office had 'normal' rear-facing cylinders... not that this changed anything about stalling potential.
Careful control means to produce higher spot effort at starting -- Porta said with Weiss ports; Wardale with Herdner valves -- has to go, on a duplex, hand-in-hand with fast effective traction control acting on the several engines. It is unlikely that the desired precision or speed of action would be served either by a mechanical grapevine linkage to a poppet multiple throttle or a typical reverser control, although with long practice and familiarity I suspect a skilled and sensitive engineer might 'make do'. Even an air throttle actuating a multiple in the usual place will have ghastly response in a duplex slip; I think the likely 'safe' solution is a laterally-acting version of independent on driver cheek plates, away from the tires and treads.
Overmod
I wrote you a pm about this ..
This is like
https://www.youtube.com/watch?v=RZq2JYYKzmo
to me and I think the problem lies somewhere quite different.
I could now write about where and why but first please do my pm
request, thank you.
Sara
It will be a few hours to get to the PM.
See if you can read Vauclain's address to the folks in Atlantic City about the T1 detail design.
It is my opinion that extreme short stroke, especially on a duplex, was a wrong answer on many levels. Same with the 'clever' idea of eccentrically grinding the mains to get less unbalance in reciprocating augment. This as Lima and, to a lesser extent Alco, were going to almost ridiculously long stroke in order to be able to use smaller pistons and other gear.
I suspect the answer to true high speed in that era was turbines through a variable-ratio gearbox... whether or not using a hydrokinetic coupling or Bowes drive ahead of it. But that's just my opinion.
>>read Vauclain's address to the folks in Atlantic City about the T1 detail design.<
>> 'clever' idea of eccentrically grinding the mains to get less unbalance in reciprocating
augmen<< So it wasn't round anymore or what???
More riddles !?!?
No, come on, can't you express yourself in such a way a person not knowing your daily life, prefences and comments and has read all you have read in all details can understand it?????
Prozse bardzo, pan Overmoderem
Dziekuje, dzien dobry!
A bientôt, bis nachat-dann,
Gruezi, hobä de-Eehrä
S1x7
UP Big Boy
Jim K
JMKThe UP BigBoys had very big tenders.
We aren't counting tenders with 'water bottle/A-tank' auxiliary water tenders. I don't offhand know the exact weight of the 'tender' that 3985 toured with, or the one that 4014 currently uses, but the added 'water car' would have substantial tare and loaded weight. Of course if the restored 2926 were to tour, they could use a comparable range-extending auxiliary tank... and remain 'biggest'...
Incidentally I cleared up the 'eccentric main pin' thing in PM, but for anyone puzzled: the main pin on a reciprocating locomotive with the sort of Timken rods on a PRR T1 is mounted to produce the shortest possible stroke before the metal between the main pin and axle holes becomes too thin for distortion or cracking in service -- this is what determines the 'throw' of the side rods at 26". On a normal engine this also determines the stroke of the mains.
For very high speed some designers want the stroke of the mains to be shorter than the throw of the rods (for balance reasons covered in references like Ralph Johnson's). This could easily be done by cranking the main pin... but this would make it impossible to service the side-rod roller bearings without pressing the main pin out of the wheel, not a good idea for a great variety of reasons.
What Baldwin noticed, and Vauclain described to the audience in the presentation from Atlantic City, was that on a duplex engine the thrust from a main is low enough that a smaller main-pin diameter can be used for the corresponding roller bearing. Normally when the main pin is machined, this smaller diameter would be concentric with the inboard bearing diameter. Baldwin instead machined it 'eccentric' so that when the main pin is installed in the wheel, the crank circle of the big end is smaller but the side rod bearing still presses on and off past it. (There are detail drawings that clearly show this but I cannot pull up and post them from here.)
For very high speed some designers want the stroke of the mains to be shorter than the throw of the rods
I believe this was a feature of the Ivatt "Alantics" of the English Great Northern Railway. 1898 was a little before Timken bearings were a consideration, but the Atlantics were the first fast express locomotives on the GNR that had coupled axles, so they may have thought deeply about balance..
Peter
M636C1898 was a little before Timken bearings were a consideration...
Conversely to normal American practice, the eye or at least the bushing bore of the main rod has to be smaller than that of the side rod. This is much less a concern for the Timken lightweight rods than for solid construction with adequate plain bearings.
The Ivatt engines had a 6" diameter journal for the side rods and 5" for the main, implying an axis offset of ½" for a stroke difference of 1" (23" vs. 24", which appears appreciable when augment stresses at high cyclic are concerned)
https://www.gnrsociety.com/locomotive-class/c1/
Going back to the T1s, I'm ticked because there was a drawing right in the coverage of Vauclain's remarks that showed the detail of the pin, and I can't find it to post an URL or cite. There are certainly drawings in the T1 Trust engineering repository but I can't link those here.
Our community is FREE to join. To participate you must either login or register for an account.