Trains.com

PRR Duplexes and Experimental Engines ( S1, S2, T1, Q1, V1 etc.)

89518 views
772 replies
1 rating 2 rating 3 rating 4 rating 5 rating
  • Member since
    May 2003
  • From: US
  • 25,157 posts
Posted by BaltACD on Sunday, December 11, 2022 8:06 PM

Never too old to have a happy childhood!

              

  • Member since
    September 2003
  • 21,602 posts
Posted by Overmod on Wednesday, December 14, 2022 2:22 PM

That video has some of the most splendid video quality I've ever seen in a YouTube video -- even before we consider the subject matter.  It's well worth watching, and I think we should encourage his 'channel' with likes and subscribes.

  • Member since
    July 2016
  • 2,607 posts
Posted by Backshop on Thursday, December 15, 2022 5:27 PM

You do know that was CGI?

  • Member since
    May 2003
  • From: US
  • 25,157 posts
Posted by BaltACD on Thursday, December 15, 2022 6:31 PM

Backshop
You do know that was CGI?

CGI can illustrate things that no longer exist as well as things that only exist in the mind of the creator.

Never too old to have a happy childhood!

              

  • Member since
    March 2016
  • From: Burbank IL (near Clearing)
  • 13,524 posts
Posted by CSSHEGEWISCH on Friday, December 16, 2022 10:18 AM

BaltACD
 
Backshop
You do know that was CGI?

 

CGI can illustrate things that no longer exist as well as things that only exist in the mind of the creator.

 
It would be interesting to see some of the more outlandish proposals that were never built but turned up in Wiener's "Articulated Locomotives".
The daily commute is part of everyday life but I get two rides a day out of it. Paul
  • Member since
    September 2003
  • 21,602 posts
Posted by Overmod on Saturday, December 17, 2022 3:57 AM

Backshop
You do know that was CGI?

I was referring to the video clips at the beginning.  The MSTS was typical 2011 quality, no more, no less.

An actual render in modern CGI would be much closer to 'photorealism', both in ray-tracing lighting effects and resolution.  See the 3D derived-pointcloud models that were produced for the T1 Trust about a half-decade ago, but in color...

  • Member since
    January 2019
  • From: Henrico, VA
  • 9,688 posts
Posted by Flintlock76 on Saturday, December 17, 2022 4:59 PM

Backshop

You do know that was CGI?

 

I enjoyed the video!  Authentic footage where available (and the rendition was excellent) and CGI where needed.  A pretty good balance and a great end product.

I definately gave it a "Like!"

  • Member since
    May 2003
  • From: US
  • 25,157 posts
Posted by BaltACD on Monday, May 15, 2023 7:51 PM

Question - On compound non-mallet engines.  Are the low pressure cylinders quartered on the same phase as the high pressure cylinders?

Never too old to have a happy childhood!

              

  • Member since
    March 2016
  • From: Burbank IL (near Clearing)
  • 13,524 posts
Posted by CSSHEGEWISCH on Tuesday, May 16, 2023 10:06 AM

BaltACD

Question - On compound non-mallet engines.  Are the low pressure cylinders quartered on the same phase as the high pressure cylinders?

 
It would have to be that way on Vauclain compounds and probably that way on cross compounds and D&H 1403.
The daily commute is part of everyday life but I get two rides a day out of it. Paul
  • Member since
    September 2003
  • 21,602 posts
Posted by Overmod on Tuesday, May 16, 2023 11:31 AM

Thing is that it would require 180-degree opposition on each side to get the engine to balance, and that is manifestly not true of a Vauclain compound (I presume you mean type 1, with the high-pressure and low-pressure cylinders outside, driving on a common crosshead).  There are illustrations on the Web of the special piston valve and convoluted port and passage arrangement that is at the heart of this method of compounding, and although the porting and operation are complex to analyze, the system was certainly capable of developing near- if not actual world's-fastest speeds in the early 1890s.

I think the inside connections on a Cole balanced compound are quartered at something like 135 degrees relative  to the (quartered) outside.  Early compounds did not use the analogy to loop scavenging to lower effective HP backpressure going into the receiver, and I'd think dynamic balance would be a more important concern that equalizing MEP... especially with the actual condensation of LP steam in the receiver and then during expansion in the LP cylinders, which was often far more abysmal than manufacturers and designers seem to have realized.

The 'answer' of course can be seen first in effective steam-streamlined passages and superheating in modern compounds, and then in the 'booster valve' applied to some of the N&W Y-class engines (which of course were only circumstantially and accidentally 'in pnase' HP to LP).  I still have no hard information on whether the LP reheat 'superheater' on 160 A1 was actually useful or not.  In my opinion the 'best' approach is still that proposed by Chapelon, which is like a modulated version of the booster valve: high-pressure saturated or superheated steam is preferentially injected into the receiver at acceptable determined HP back-pressure excursion, so that not only the MEP but the instantaneous pressure on the LP pistons over the effective range of their stroke 'matches' what the HP cylinders are producing.  That does not require that the HP and LP be co-phased at all; in fact my RSR engine design phases the front and rear soft-conjugated engines (via a detent) at 135 relative (both engines being simple 2-cylinder DA in quarter) so that there are eight controlled power impulses per revolution for the engine as a whole, which also should help eliminate objectional high-speed surge effects.

  • Member since
    February 2005
  • 2,360 posts
Posted by timz on Wednesday, May 17, 2023 4:04 PM

BaltACD
On compound non-mallet engines.  Are the low pressure cylinders quartered on the same phase as the high pressure cylinders?

Guess a four-cylinder would always have the low-pressure 180 degrees from the high pressure on that side, wouldn't it? And the left and right sides the usual 90 degrees apart. Presumably that's why the main-driver counterweight is at the bottom on this 4-6-0

https://babel.hathitrust.org/cgi/pt?id=uva.x002211444&view=1up&seq=604&size=125

Next question: what was more common, four cylinders all driving one axle or low-pressure driving the lead axle and high-pressure the second?

  • Member since
    September 2003
  • 21,602 posts
Posted by Overmod on Wednesday, May 17, 2023 6:10 PM

Look up 'Plancher System' for a different approach to four-cylinder compounding (e.g. the early Italian cab-forward 4-6-0 design)

The de Glehn-du Bousquet design, one of the more successful compounds, had the HP drive on the main with a conventional axle and the LP on a cranked leading driver axle.  Four sets of valve gear with the LP fully adjustable separate from the HP -- a reason the French called engine-drivers 'mecaniciens'.

Most of the balanced compounds (such as Cole and Vauclain type 2, the latter being the "Baldwin balance compounding' applied to the New Haven engines) had a cranked main-driver axle and all four mains bore on the same driver pair -- in other words, not divided-drive.

An interesting type that you'd think couldn't work all that well was the von Borries, which is a normal quartered 2-cylinder DA... run as a compound with HP on one side, LP on the other, and asymmetrical counterbalancing.  The original PRR T1, with the 84" drivers, was made this way and there is a photograph of it at high speed with a considerable train -- the trick was that the two sides had separate cutoff so it was relatively easy to adjust running balance at a given throttle opening and HP cutoff.

  • Member since
    February 2012
  • From: CAPE CORAL FLA
  • 507 posts
Posted by thomas81z on Sunday, May 28, 2023 6:58 PM

I have read & reread this thread since 2018 & it still blows my mind well from what i understand , wow just wow . thank you guys so much for deep diving for us that arent as knowledgeable  on this subject

  • Member since
    February 2012
  • From: CAPE CORAL FLA
  • 507 posts
Posted by thomas81z on Saturday, June 15, 2024 6:42 PM

just bumping this to the top to reread

  • Member since
    February 2012
  • From: CAPE CORAL FLA
  • 507 posts
Posted by thomas81z on Monday, June 24, 2024 5:29 PM

I WANT TO BUILD THIS IN HO

 

My fantasy Steam turbine locomotive PRR V2 #6600 Smile, Wink & Grin

 

  • Member since
    September 2003
  • 21,602 posts
Posted by Overmod on Tuesday, June 25, 2024 10:10 AM

Tell me more about this.

Original V1 - two Westinghouse turbines, 1 impulse and several reaction stages (similar in principle to the unit on PRR 6200) each driving four axles connected by Cardan shafts (see the Krauss-Maffei Amerika-Loks for some representative detail).

Nominal 8000shp, using a slightly-modified Q2 boiler.  The original design was approved for production in 1944 (when fast, efficient power for long heavy westbounds was becoming a priority) but was 'not proceeded with' because the water rate was abysmal -- you'd have needed multiple water tenders, not just the one in the design.

Subsequently the design was revamped to utilize the Bowes drive, which would have permitted 120mph+ road speed -- this is at the same time the suggestive 'design patent' by Loewy et al. for the streamlined 'passenger' body was filed.  This was stillborn when all the rest of big complicated steam development was abandoned on PRR, between 1947 and 1948.

However, at the end, 'someone' in PRR publicity ghostwrote an expansion to "9000hp" (literally; the locomotive is in ghostly outline in a PRR brochure about modern steam power!) which might have been interesting to see attempted.  It is possible that something like a good double Belpaire with Cunningham circulation and Snyder preheaters could produce the required mass flow... if Operation Downfall had followed the expected course.

Even with the Bowes drives, there might have been a need for Ferguson clutches between the axles, to prevent the wheel diameters needing frequent dressing to keep them commonly profiled.  Note that the early work in magnetorheologics was in the late Forties and would have been highly attractive for this purpose.

You could, for fun, design a larger nose-mounted coal bunker, and adapt the FM-TV system used for the Navy's version of Aphrodite as better view of the road...

  • Member since
    February 2018
  • 37 posts
Posted by RailfanGXY on Sunday, June 30, 2024 5:57 AM

Hermann

 

 
Jones1945
The S2 also hauled lots of crack trains that the S1 seldom pulled, like the Broadway and Manhattan limited. I read somewhere that people saw the S2 once appeared in Pittsburgh, the "forbidden city" of the S1. On the other hand, I have seen pics of the T1 phototypes powered the Trail Blazer.

 

 

Hi Jones1945,

interesting thing that for three long years, the two prototypes ran into and through Pittsburgh without any reported troubles.

Only in Summer 1945, the first derailment took place im Pittsburgh. Then, just one month after delivery of the first serial T1s, 5502 derailed on December 1, 1945, and another serial T1 the very next day.

Did somehow, tragically, the two prototypes have the edge over the 5500s in curves?

 

 

The performance of Flash and Buck seems to vary compared to the production T1's, but they certainly didn't have an advantage in lateral motion! They did run Harrisburg to Chicago for several months, but suffered repeat derailments at Federal Toward heading east. Not every time, but often enough to be confined in 1943 to the Ft. Wayne Division like the S1. They ran towards Pittsburgh a few times between then and '45, as did the S1, but would've been cut off before the tower and get turned and/or serviced at Scully Yard across the river.

Still find it humourous that the Q1 had more clearance into Pittsburgh (and more range overall) than the passenger duplexes for a year or two. It might not have been in service as much as it could've, but it was a Panhandle regular, plus runs on the Sandusky Branch. Well over 600 miles if I did my math right. Also did some math regarding its lifetime mileage and for kicks compared it to the early records of the L-4 Mohawks. The Q1's annual average is lower that of the L-4's within their first four years, but it's higher than their mileage in passenger service during that period. Quite impressive given the lower range and speeds. The design says Shop Queen, but the mileage says Large Cruiser.

  • Member since
    September 2003
  • 21,602 posts
Posted by Overmod on Sunday, June 30, 2024 8:48 AM

Most of the early duplex designs (B&O, ATSF, PRR Q1) were built with the assumption of minimizing 'rigid' (I.e. driver) wheelbase.  The Q1 is a peculiarly interesting case, as it was specifically intended as a 5/4 expansion of the M1/M1a 4-8-2, with 77" drivers so clearly intended as dual-service, not just M&E.  You would not have gotten this engine through any restricted trackwork if it had the rear cylinders a la Q2 -- even necking the cylinder support.  Even a "duplex" version of an ATSF 5001/5011 class would have had a shorter driver wheelbase!

The real trouble with the Q1 in my opinion is that PRR had no idea what to do with it, certainly after 'standardizing' on T1s for all the fast passenger trains, and keeping the 50mph freight restriction postwar.

  • Member since
    February 2018
  • 37 posts
Posted by RailfanGXY on Friday, July 5, 2024 8:10 AM

Might be a wild take, but I think the best thing for the Q1 would've been to put it in passenger service at that point. Sure it lags behind the T1 in most aspects, but it has a bit more range and availability compared to the S1 and S2. Its power curve is seemingly ahead of most 4-8-4's, and doing better than a Niagara until 45 mph while suggesting it could catch up to an N&W J at 70 mph if not a little later.

I typically think the PRR could've done with a 4-4-4-6 if they wanted to make a step up from the M1, but the fact the Q1 could still generate that much power as a 4-6-4-4 with its "restrained dimensions" is still quite impressive. And to its credit, not only does its tractive effort eclipse all the 4-8-4's, it also has an ace or two above the J and the Niagaras. As well as having higher drivers than a J, and not needed tandem rods like a Niagara, the duplex layout lends the Q1 a bit better to high speeds. Firstly in safety; as much as people like to point out the duplexes propensity to slip, none of them broke large components on the running gear. The T1's broke valves often from excessive speed, but the gear itself wasn't damaged too often. Compare that to the numerous 4-6-4's and 4-8-4's that bent or threw their rods at least once. Secondly, there's a perfomance benefit.

Rather than raising the pressure and stretching the cylinder (smaller bore, longer stroke) the duplexes use shrunk cylinders; comparitively short strokes. I recall one advert of the S1 using "switcher cylinders!" Closest comparison that comes to mind are the Bulleid Merchant Navies, quite similar to the T1's in their original form (for another time haha) but in their rebuilt form retained the quite small 18x24" cylinder dimensions, as well as a free-steaming boiler tied to a Belpaire firebox. The next closest Pacific with the same pressure and wheelsize (and cylinder count) was the 30 "shared" LNER A2's. The Peppercorn iteration managed 101 mph with a Double Kylchap and 19x26" cylinders, while the earlier Thompson A2/3's (identical TE dimensions) could match the timings of the A4-hauled Coronation to the minute, so they likely could make 101 mph as well. However, 3 of the rebuilt Navies were clocked at or slighlty above 105 mph! They're probably the highest speeds recorded for 74"wheeled locos, as both the authenticated speed records (streamlined and not) and the next loco after them to break 100 mph had 80" drivers.

So back to the Q1, even with the crowded space of its cylinders and firebox, it would be a better option for running west of Pittsburgh compared to a 4-8-4 of similar power. Could've made a good pinch-hitter on trains a T1 wasn't ready for, rather than repeatedly scrambling to hustle K4's to the task and explain to management why the Blue Ribbon trains didn't have their newest investments on the head end. At least 6130 would've only had 3 years under her frame instead of...15 at the minimum

  • Member since
    September 2003
  • 21,602 posts
Posted by Overmod on Saturday, July 6, 2024 7:23 PM

Two initial points:

The Q2 was just too big for any conceivable passenger train PRR could have run cost-effectively... and it still would have had issues above 85mph 'making up time' -- duplex drive isn't magic -- unless you rebuilt it to zero overbalance.  You also have to deal with the higher water rate of the relatively large boiler at 'passenger' mass flow with excessive dead space and 69" drivers.

The better answer across the board for passenger work was to use the Q2 boiler as the V1 with Bowes drive -- even if it involved far more independent-brake shoes and foundation!

There was no need for a 4-4-4-6 on PRR to make the greatest single-unit horsepower practically necessary.  The Q2 didn't have the Lima-style double Belpaire (which was the thing most 'necessitating' that six-wheel trailer on Townsend's stillborn 4-8-6) and easily made over 7600dbhp with a four-wheel trailing truck.  I doubt that the added weight of Snyder preheaters and Cunningham-circulator manifolds would constitute 'too much back-end weight gain' to require the extra axle.

What you WOULD want on a double-Atlantic duplex is the general firebox size and construction Baldwin originally intended -- 102 to 104 square feet.  This was very pointedly included in the May 1945 C1a design -- which let's face it would have been a lower-augment locomotive using a little-modified Niagara boiler and good optimized piston valves.  No need for the cost, weight, length, etc. of a six-wheel unpowered truck.

  • Member since
    February 2018
  • 37 posts
Posted by RailfanGXY on Monday, July 8, 2024 4:07 AM

I follow most of these points though I'm baffled by bringing the Q2 into it. It had power for days up to medium speed, but I wouldn't consider it for passenger work, even as a last resort. The boilers had trouble holding water on freight runs, heaven forbid how much water they'd use on passenger work. Crews worried about the J's rpm over 100 mph, no way the Q2 would be better. Only the Q1 should've been "bumped up" after the war was over, the Q2's power was far more valuable on freight, even if  they didn't quite run fast enough to use all of it.

I want to be clear though, I mean 4-4-4-6 not in the same way at Lima's proposed 4-8-6, but as an alteration to the Q1's 4-6-4-4 arrangement. For an enlargement over the M1, a duplex could've done just fine with 8 smaller but sizeable drivers. The C-1a's firebox seems to be about the same 101 sq.ft. as the S-1b/S-2a, which makes sense given...everything else. I can only assume Altoona learned their lesson from the S1, as the Q1's firebox is just over 98 sq.ft, and the extra trailing axle is more to keep the engine balanced base on those dimensions. Narrower than the S1 and T1 fireboxes but between them in length.

But even with a firebox a few inches longer than a Niagara, not only is the Q1's trailing truck strangely spaced, but the rear trailing axle also has the "9000's" issue of carrying far more weight than the other axles. Most of that is probably the fault of the booster, but the Q2's and even 6111 don't have such an imbalance.

The idea of a 4-4-4-6 stemmed to make the Q1 more available quickly without changing anything above the frame. Placing the rear cylinders where the 5th drive axle was leaves it far too long to ditch one driving axle and still have only two trailing axles. Might not have a Double Belpaire, but that long combustion chamber...Fig. 4 is the sort of arrangement I envisioned for the truck, but any 6-wheel trailer works for this. Not to support a larger firebox, but to balance out to overall engine and not leave overhang on either side.

The aforementinoed wheelbase reduction kind of deterred me from thinking of the more conventional arrangement. Flipping the rear cylinders and making a slightly longer T1 (rear cylinders placed where the 3rd driving axle was) would make things simpler. There's still the overloading to worry about, but I recognize there's far less need for an extra trailing axle at that point. A new pair of cylinders in the same dimensions of the rear pair would put its tractive effort slightly ahead of the T1 and M1 as well. Hmm...might need to try that for a redesign.

  • Member since
    September 2003
  • 21,602 posts
Posted by Overmod on Monday, July 8, 2024 7:28 AM

That's a beautiful drawing, but didn't your momma tell you that putting pin-guided Adams trucks on the rear of a locomotive is a formula for disaster?

The Reading tried it, in an effort to improve a (good) 4-4-2 into a 4-4-4.  The tales of how bad the result was, alone, should dissuade you.  The Germans did it at high speed and found, at least according to one account, that you had to modify the position of the pivot point of the truck frame relative to direction to get even ~80mph out of the things.

In any case, Withuhn conjugated duplexing is not something you really want to perpetuate in practice (compared with Deem conjugation).  With proper lateral motion, a double-Atlantic with acceptable main-rod angularity on both engines doesn't have that much disadvantage with nominal 'rigid wheelbase' and a whole lot fewer issues with steam-line routing and clearance.

Something else to do (although it would be a bit tedious) would be to use the Bissel formula combined with weight distribution to figure out exactly where the pivot point, steering arrangement, and axle spacing of a proper Q1 trailing truck might have been.  Normally on a two-axle Delta trailer, you'd have the opportunity to equalize the axles and connect with the levers forward to the engine equalization.  That means at least in theory you can space the axles correctly to get proper wheel-to-rail alignment for both axles in curves.  Then figure out how to get the longest steering lever arm back to the rear corners that you can shoehorn into your loading gage (it may no longer be fully outboard at the chassis corners).

The firebox on the C1a is exactly that of a Niagara, because the boiler is in all possible dimensions the same as a Niagara (this is specifically mentioned in the May 1945 discussion.  The chassis is very similar to the PRR T1 in many dimensions, except that it uses piston valves and Walschaerts gear -- I don't think development of the C1a detail design included whether some dimensions were adapted from what wound up on the PRR T1a, but it "could have been used" and at least for purposes of modelmaking can be.

All you need is proper viscous-clutch conjugation, implementation of the independent via floating lateral calipers and cheek plates (for traction control) and some other tinker to get rid of the effect of high- and low-speed slipping entirely with practicable 1940s technology...

...but the stroke is still too short to prevent stalling, which was the real problem C&O had with the T1.  If you have a 6400ihp locomotive, the temptation is going to be to load it to suit the horsepower at your anticipated operating speeds.  That will bite you if you combine fixed cutoff with very short stroke and no booster.

The 'more correct' answer in any case if you want side-rod drive is going to be the PRR R2 (which is the S2 with proper suspension to the gearcase, and the planetary transmission patent-specified by Westinghouse in 1948 which gets rid of the ghastly low-speed mass flow and the reverse-turbine shortcomings together.)  I don't think there is a showstopping reason you couldn't use this drive on a ten-coupled chassis if you wanted the incremental adhesion. 

  • Member since
    February 2012
  • From: CAPE CORAL FLA
  • 507 posts
Posted by thomas81z on Wednesday, July 17, 2024 4:33 PM

Jones1945
those seats look ergonomically correct

 

 
Overmod

http://digital.hagley.org/PRR_11454

Knew if I dug into some of the records, I'd find it.  This is the S1 backhead view.

 

 

 

Such delicate interior! S1 is like a 1925 Rolls-Royce Phantom 1 in my heart. 

CoffeeSmile, Wink & Grin

 

SUBSCRIBER & MEMBER LOGIN

Login, or register today to interact in our online community, comment on articles, receive our newsletter, manage your account online and more!

FREE NEWSLETTER SIGNUP

Get the Classic Trains twice-monthly newsletter