A good read for steam heads…
A Pennsylvania Railroad Class T1.
LAST CHANCE for a Pennsylvania Railroad Class T1
I loved this story from the first time I read it … and am still looking forward to some Q2 tales. Gil Reid bought into the story ‘big time’, and the result is something we should all have on our walls:
But, just as there is no ‘real’ Santa Claus for Virginia, there is no getting around that 5536 had no unusual mods to the speed-recorder system documented anywhere in the (rather extensive) material so far identified. I am still looking to see if 5500 (which got the rotary-cam setup) might have had a higher encoder and speedometer display put on as part of the conversion … but for reasons previously and exhaustively described, I do not think there is, or was, any reason for PRR to put more than a “100mph” speedometer on any of its steam locomotives, with the possible exception of the S1 (which I think might have been 110mph, still well short of what was reported on 5536 - I am looking for my backhead shot of the S1 locomotive to confirm this).
Note that Crosby is not a T1 expert - he claims that all the T1s had ‘rotary cam poppet valves’ which is a strange thing to make it past the Trains Magazine editors. I continue to think that this is akin to the situation where the 161mph Hudson slip tests morphed with the ‘serum to Nome’ railroaders’-boyhood stories to produce that story about a Hudson rushing medical relief across New York State at 157mph … a lot of verisimilitude with a few little mistaken points that spoil the tale a bit, like some of John Barnes’ early science fiction.
Note that I say this all with extreme regret, as I think the T1 design was easily capable of performing as reported.
Thanks for the Gil Reid artwork. immediately downloaded to my hard-drive. Thanks!
I remember the article and especially the Gil Reid painting. I saw it in a short-lived magazine called “Vintage Rails” which I wished I kept, dammit!
It may have originally appeared in “Trains,” but I’m not sure about that.
Thanks David and RME!
I don’t have the T1 print but I do have a similar one of an NYC J3a Hudson framed print entitled “79 MPH”.
Another really beautiful painting that is on my hard-drive is Rose’s Walking through Texas. I really get a lump in my throat every time I look at it. Two Riply Northerns heading into the sunset with SF refers.
Some day I will find the right story to attach to it. Railroading has insipired some terrific painting as well as photography.
One that I am particularly fond of, as telling a story that unfolds as you look at it, is John Winfield’s take on Scheeler’s Rolling Power called “Three Hours to Amarillo” (I’m not posting a thumbnail, but here is a place where you can see a large preview by clicking the thumbnail (and a great many fine other paintings as well).
John Winfield is a great, great artist. I am fortunate to have prints of his “Extra 3801 West” and “Cab Forwards On Tehachapi”. Looking to give myself a Christmas present of “Storming Curtis Hill” showing ATSF 2926 on a reefer block. All his work is excellent.
A great story indeed! Almost made me feel like I was in the cab.
Here’s a question for one of you with an engineering degree. This T1 was pulling 14 cars, which I would estimate to be about 1200 tons. Please feel free to amend that estimate if you wish. Add to it another 450 tons for the locomotive and tender. Now tell me how much horsepower would it take to move that load 120 mph on level ground. Also is there a formula one can use to determine HP? In my foggy memory I recall the T1 to be capable of 5000 to 6000 HP. Is that enough to pull off the alleged performance?
No one here knows. We can guess at it, and maybe be right within a factor of 1.5 or so.
Horsepower of the locomotive, you mean? There are formulas, but no one would promise they’re right. Only way to learn the power of a steam locomotive is to test it.
The T1 may have been capable of 5000 dbhp, but not at 120 mph.
There is no ‘one true’ formula for calculating dbhp (which is the only thing that really counts on the road). Chapelon says PRR observed 6760ihp from the instrumentation on one run on the test plant at 100mph indicated, but he does not say what was recorded on the roller brakes for that run (which would be close to wheelrim HP). Machine losses on the T1s were remarkably low, so I would not think dbhp would vary very far from wheelrim output; there should be a correction factor for the high-speed slipping but that’s probably impossible to model deterministically, let alone get meaningful numbers from; it should be possible to model the aerodynamic drag under straight and quartering conditions using a second locomotive and a dynamometer car able to record accurately in buff.
Meanwhile, the practical falloff of developed HP at high speed is probably far less on the T1s than most other locomotives, as both admission volume and timing precision will be maintained by the poppet valves up to very high cyclic rpm (as observed on 5399) and the superheat will ramp up to ridiculous levels (as on 614) without producing catastrophic valve issues (as on the high-speed class J testing) if the 1948 valve improvements are used. So the falling ‘tail’ of the horsepower curve, already relatively broad and flat, may be sustained far longer than anticipated at speeds above those PRR actually tested on the plant. I expect the actual limiting factor to be backpressure in the front end (and characteristics of chassis stability) and both of those to be well above 120mph on level, straight, ideal track.
I think you are vastly overestimating the train weight. If you consult PRR equipment diagrams for the lightweight equipment (for example the POS211a observation) you will find a light weight under 60 tons, and I believe this may cross-relate well with 14 such ca
This story and the painting ran in Trains Magazine back in the 60’s or 70’s - likely I have the issue and can find it.
Doc
Are there any usable sources from machine losses – friction between the power in the cylinders and power at the wheel rims?
John Knowles had posted a long report on this subject, but it looks to be that his Web site is no longer up. What was interesting about his work is that this machine friction varied with load and speed in a way that made it more complicated than a simple “mechanical efficiency” number. Another point was that the machine friction of a locomotive was much higher than the locomotive’s Davis formula rolling resistance, to the point where the machine friction of the locomotive was roughly comparable to the (non-aerodynamic) rolling resistance of the entire train it was pulling.
I am interested if there are other sources going into comparable detail.
Wish to correct my Johe Rose SF ReferBlock painting posting:
Racing into the SunRISE is correct!
There are some discussions relevant specifically to the T1 in the Engineering Committee sections on the T1 Trust site.
Part of the reason for the machine losses being (relatively) low was the use of roller bearings, and supposedly-more-than-desultory wheel turning to profile. I’m sure the burns from the various kinds of slip did not help the situation, but the magnitude of the difference was probably only a fraction of a percent, the difficulties being much more pronounced in terms of reactions at high speed and increased propensity for very-high-speed slipping.
The difference between Franklin type A and piston valves is substantial, both in relative friction and inertia. Typical piston valves for a locomotive this size would consume something like 35hp around diameter speed, and of course much more at higher speed or if unanticipated high superheat degraded the tribology. Franklin indicated that the hp to run type A was about 3hp (much of which was in compression of the springs and lubricated contact of the little dwarf radial gear in the box). “Standard” type B RC was supposed to be less, but I don’t have any material on the specific 4-valve RC conversion (“B-1”) that the only RC T1 had, which had more cams and linkage at all ‘eight corners’.
ISTR some correspondence at the Hagley that indicated the hp to drive the valves on the T1a was about 22hp, but I don’t really know if that was for both or for ‘each engine’ - the valves are not that much smaller than typical.
As a perhaps-interesting aside, there was some discussion about flexion in the Timken lightweight rods being a source of loss (Chapelon mentions the effect in conjunction with lateral-motion devices and roller-bearing rods). I’m not sure I believe this, but
Unless I’m sadly mistaken you need a little more correction - isn’t it TED Rose?
While we’re on the general subject, here’s Ted ‘on topic’:
and for a couple of you who are enamored of the Milwaukee F7, here’s something T1s for all their slipping didn’t do:
And, for Mr. Klepper, here’s something akin to ‘Racing into the Sunrise’ that he might appreciate…
I am guessing that is a representation just a second after broken rod?
It is one of two possible things. The likelier one is the event Scribbins described where (if I recall this correctly) the crosshead seized in the guide due to tribology problems at speed. The train momentum (and thrust from the other side) bent the main, which then windmilled knocking off the shrouding but somehow not digging into the ties and lifting that side of the locomotive.
The other possibility, which I saw documented in a thumbnail picture in Trains years ago without a full narrative, involved a broken main pin. I suspect those were starting to become relatively frequent in the latter-'40s experience with big steam running lots of effective horsepower through just two pins, especially in arrangements like the ‘revised’ system on N&W Js that put the bending moment on the main far out on the pin (somewhat ironically to relieve a chronic fracture problem on the extended #4 driver pins in the inline tandem-rod original arrangement). I do not know how much actual R&R of roller-bearing rods and magnafluxing of pins actually took place, but have my suspicions that there is a relatively narrow critical range of load in which the failure of the pins would rise dramatically…
I found this info from a thread from 2012.
http://cs.trains.com/trn/f/741/p/2917510/reply.aspx
[quote user=“NM_Coot”]
Santa Fe test results showed that the 3765 class with limited cutoff could maintain full boiler pressure at all speeds and cutoffs. Test runs with 3766 showed 5450 MIH at 55 to 70 mph and over 5000 IH from 35 to 90 mph. Given the Santa Fe penchant for long runs at high reliability, this high HP over a very wide speed range makes for excellent performance with heavy, fast services where maximum output is required over wide speed ranges. The concept of limited cutoff was recommended for implementation on the 3460, 3776, and 5001 classes. In passing, the test of 3766, as equipped with roller b