Pennsylvania RR T-1 modification?

feltonhill wrote:Q2 used about 50% more steam [than the A] for same output. Big difference was the short stroke against the longer stroke of the A.

...When you had the longer stroke engines, they behaved very much better in cylinder efficiency and that’s the reason the A’s and the rest were as high as they were overall.

Obvious question: why did the N&W use 30-inch stroke on the A? Wouldn't 32-34 have been better? Maybe even better with 38-40?

timz,

Can't answer the question on the A's 30" spec. It was likely the best compromise based on info available in 1935-36 when the A was designed. A 34" stroke at that time was used on C&O 2-10-4's and the first NKP 2-8-4's, maybe others.

There are three calculation books that were found in N&W papers at the NWHS archives, but nothing that far back at this point. The oldest page is dated 2/15/37 in Voyce Glaze's book. McGavock's first page is dated 4/1/43. We may find something else as time goes along, but it's a slow process sorting through all the paper.

During the 1975 interview Mr. Pilcher offered the opinion that the A may have been improved with a 32" stroke, but modifications were not pursued. I'll try to listen to the interview tomorrow because I believe he gave a few more details. Just got back from three days in Roanoke and everything's piled up here.

Typewriters,

I’ve read Don Ball’s description of the C&O tests in his book and it seems he very likely had access to the test report. However, it also seems he just couldn’t resist editorializing to make the story a bit better. This is where his account came apart and deviated from events as stated in surviving correspondence from both C&O and PRR personnel who were involved in the tests.

C&O’s Chief Mechanical Officer C. B. Hitch observed the Waynesboro situation this way in correspondence with PRR(C&O letter, Hitch to PRR’s Cover, October 7, 1946):

“The 5539 did not slip but failed to start the train at Waynesboro after taking slack several times. It was finally necessary to call on a yard engine for assistance, which engine was cut off as soon as the train was well in motion.”

PRR’s Special Duty Engineman Brown was assigned to the tests and confirms Mr. Hitch’s observation that the T1 did not slip in its attempt to start the train. His report back to PRR covered the same event as follows (PRR memo, Brown to Warren September 18, 1946):

“Stopped at Waynesboro, Virginia, regular stop, at 10:01 A.M. Departed at 10:07 1/2 A.M. and after trying to start train seven or eight times, was assisted by pusher. Maximum drawbar pull when trying to start train was 45,500 pounds. Grade at this point was 1.25. Engine did not slip and had full head of steam.”

At Craigsville PRR’s SDE described the situation this way:

“Lost about 20 minutes at Craigsville account of waiting on eastbound C&O #6. 15 minutes of this time was lost in trying to start train on grade which was 1.44. Engine was hot and did not slip. No time was lost at Clifton Forge as engine was serviced while train was switched.”

The repeated attempts to take slack would cause the bumping and jostling Ball described at both Waynesboro and Craigsville, but that’s where his facts seem to end. Neither he nor Tom Harley were there, so Hitch and Brown, both of whom were involved in the tests, would be more credible.

For many years, I believed Ball’s account to be the best available, but after I started to look at the surviving documents myself, I was disappointed. He came very close to getting the story straight, but either he or someone else couldn’t resist embellishments and they were included in his book.

feltonhill wrote:"Maximum drawbar pull when trying to start train [at Waynesboro] was 45,500 pounds. Grade at this point was 1.25. Engine did not slip and had full head of steam".

What was the train tonnage again?

Hard to see how it could refuse to start if it was being tugged on that hard.

According to the COHS article, the train was 13 cars and 1,098 tons. In 1946 they were all heavyweight cars with solid bearings. There was considerable starting inertia, even when the bearings were warm. If you couldn't "bump" them into motion, sometimes it was a no-go. Waynesboro was one of those times.

Questions:

Granted that there were slipping problems with the B&O duplex drive locomotive as well as the more famous PRR T1 slipping.

British locomotive designs include three and four cylinder simple locomotives often driving on different axles; all driving axles were connected. Apparently they did not have slipping problems with such locomotives. You could say that these were duplex drive locomotives with all driving axles connected.

Did the fact that all axles were connected prevent slipping, or was the factor of adhesion sufficient to prevent slipping?

Finally, were there differences between the Q1 - Q2 and the T1 other than the factor of adhesion that explain why the former didn't slip but the latter did?

VA RAIL FAN

The greater number of connected driving axles, the less the locomotive is PRONE to slip, all other factors such as FA being equal. However, any locomotive will slip under the right (wrong??) conditions.

The only over-the-road test results I've seen for the Q2 indicated that in heavy coal service on N&W, the front engine slipped repeatedly even on sand. Front engine had two coupled axles, the rear had three. May have been the difference, although this was not the type of service the Q2 was designed for.

I'm repeating myself here, but IMO, duplex drives as realized in the US were better at high speed work. PRR generally used the Q2 in relatively fast service, so the slipping problem noted on N&W was not a factor on PRR. However, with a 50 mph freight speed limit, the J1 2-10-4 was a better bet. They were simpler, easier to handle, and capable of sustained running at 45-50 mph any day. The Q2 would not have come into its own until about 50 mph or so. If it had been given 60 mph time freights on N&W instead of 11,000 ton coal trains, it would have probably been something to see!! However, on PRR, it wasn't worth the extra operating expense, and the class was retired early.

Also repeating, the T1's slipping problems were the result of poor handling and slipshod maintenance. Given careful preparation and a good crew, they could get a sizable train over a district as well as any other. There are many documented trips with regular trains of 19-21 cars out of Harrisburg with no problems reported.

Nice to see continued interest in this topic!
As for the front engine that the Q-2 was reported as slipping most... I don't have my reference sources, but I have a feeling that I remember that the cylinder swept volume for the rear engine was less that 60% of the total, and of course the steam pipes to the rear engine would be longer than to the front. So, at a guess, there would be a tendency for the rear engine to be putting out a bit less power per axle than the front. Could this also have been a factor?

Thank you, gentlemen.

Let me ask whether the maintenance and handling was superior with, say, the K-4's and M-1's? I have not heard of slipping and other problems with postwar steam locomotives; one might suspect that on the same railroad the skill of locomotive engineers and the level of maintenance would be similar for all locomotive types. To attribute the T1's problems to locomotive engineers' skills and the level of mantence one would expect either similar problems with the older types of locomotives or a preferential level of skills and maintenance for the other types of locomotives.

In particular, did the J1's have similar problems with slipping? They were roughtly the same vintage.

VA RAIL FAN

Skills sets from a standard (simplex) engine might not translate directly to a duplex.

Speaking of skill sets...
I have read that, on the Altoona test plant, it was found possible to run the two engines of one of the PRR Duplexes at different speeds at the same time...
From the engineer's end, what did the controls look like? I would guess there must have been some way of controlling one engine separately from the other: if nothing else, if the front (say) engine started slipping while the rear was still o.k. it would have been nice to be able to reduce power only to the front unit?(*) Was this in fact possible?
And if it was, then right theree you have a way in which the Duplex's controls would have been much more complicated than a conventional locomotive's, and it's not surprising AT ALL that experience with conventional locomotives wouldn't "translate" immediately.
---
(*) Maybe, though, this just reveals my amateur status! Once one engine unit started to slip the other, I'd guess, would be likely to slip in a few seconds (it would still be trying to turn its wheels at the same speed as the train slowed from the reduction in power caused by the first engine's slip), so maybe PRR would have decided it was better to set up the controls so the only way (for the engineer-- reducing power to one unit was the jop of the automatic butterfly valve) to respond to a slip was by reducing power to both units. Which way was it?

Best guess? I'd say it had to do with valve timing.

By changing the valve stroke, you increase/decrease the amount of live steam admitted to the cylinders. More steam means more exhaust steam. It also means more inertia for the valve gear to deal with.

Inspection and maintenance example - PRR K4's had two sanders, one each in front of the lead drivers. T1's had either four or eight sanders depending on which modification had been installed. So a T1 required a lot more inspection and possibly maintenance (cleaning and/or adjusting the sanders) than did a K4. Lots more room for cutting corners. Unfortunately, non-working sanders on a T1 were not a good thing.

It wasn’t that the maintenance on PRR's standard classes was superior. It's just that they were simpler and more tolerant of neglect than the T1. One of the bigger problems with the T1 was that it demanded perfection to achieve best performance. PRR didn't supply this on a daily basis.

The T1's had very low machinery resistance and back pressure. They also had very large steam passages. When a slip did occur, the engine set usually wound up to fairly good speed even if the throttle was shut off immediately. The problem is that there was very little to counteract the slip once it occurred. The steam pressure just had to play out until the offending set (s) caught hold again.

There was only one throttle and cutoff on the T1. The two engine sets could have been run at different speeds on the Altoona test plant, although considerable effort was taken to make sure they ran at the same RPM. There is a story in The Keystone (PRRT&HS magazine. Vol 36, #4) recounting the events that followed the failure of the regulating system on a set of water brakes. These set the load for the locomotive and were manually adjusted by operators. When the resistance was suddenly decreased on one of the engines, it accelerated to something over 120 before the throttle was closed. No damage was done, but there must have been a lot of lockwashers on the test plant floor afterwards!

Gentlemen:

Thanks for the information - I'm learning. I think there's a good case for the Duplex drive locomotives requiring more/different maintenence and operating skills than the older locomotives.

Still remaining is the issue of the Q1/Q2 versus T1: Was the difference in performance merely the factor of adhesion?

VA RAIL FAN

I delayed getting into this because I’m not sure the FA comparison of the Q1, Q2 and T1 will answer the question. Like most things re: steam locos, it gets complicated, and this explanation will be way too long.

The Q1 had an FA of 4.34 and had 3+2 coupled axles
The Q2 was 3.90, 2+3 coupled axles

The prototype T1's had an FA of 4.15 as built, 2+2 coupled axles
The production lot T1's had an FA of 4.33

Nine T1's were later modified with 18.75" cylinders:
Nos 6110 and 6111 had an FA of 4.60
Production T1's modified (5521, 5511, 5524, 5531, 5532, 5536, 5540) had an FA of 4.80.

So all the T1's had an FA greater then the Q2's.

For comparison, the J1 had an FA of 3.99 to 4.05 and five coupled axles.

FA is supposed to be 4.00 according to hearsay, but in reality, a fairly broad range was applied by the locomotive builders. Also, recent designs had both similar figures and some different trends. Based on a large sample of locos, average FA and recent trends look something like this when arranged by the number of coupled driving axles:

FA2 - 4.48, trend to 4.63
FA3 - 4.43, trend to 4.30
FA4 - 4.14, trend to 4.14
FA5 - 3.91, trend to 3.92
FA6 - 3.67 (this is UP 4-12-2!!)

The FA increases as the number of coupled axles decreases. As built, the T1s FA’s were less than the average for 2-coupled engines. However, the few modified T1's were above the average and contributed to the trend toward other 2-coupled locos.

The Q2 should have been a major handful because its FA (3.90) was the equivalent of a 5-coupled loco. Note the similarity to the J1 2-10-4 FA of about 4.00. Yet its reputation on PRR was marred by its operating expense (maintenance and fuel usage), not its adhesion. At this point, I don’t have an answer for this. Could have been the type of service, where perhaps the Q2 spent little time going through transition speeds on its way to operating speed.

For the Q1, there is only one source of information that I know of - the article by Neil Burnell in The Keystone, Vol.39, #2, pgs 7-30. This is truly a monumental work because, as the author says, “...As best I could only describe the Q1 as a ghostly image momentarily emerging from a fog, arriving unannounced and unheralded, and just as quickly vanishing, scarcely leaving a trace.” He put together a lot of little known but entirely factual information about this one-off experimental. He notes that the Q1 had problems with the rear, 2-coupled engine slipping. Although modifications were made to the equalization system, the problem was not entirely resolved.

All this seems to indicate that 2-coupled engines on modern steam locomotives were prone to have adhesion problems, unless the FA was higher than PRR considered, but not necessarily higher than the late trends set by the locomotive builders.

Hope this helps!