Duplex Steam Locomotive / Steam discussion

[Murphy Siding]

QUOTE: Originally posted by Overmod

(I should also mention here that it would have been interesting to see what might have been done with a Centipede that had decent build and detail quality, 'standard' implementation rather than being piece-built like steam locomotives, and actually had enough power to hit the 'magic' 6000-hp-per-unit number in the late '40s. We know Baldwin had this specifically in mind with the modular 408-engine "genset" locomotive;

I never quite understood what Baldwin must have been thinking of when they built the Centipedes, until I ran accross a bit about them in Black Gold,Black Diamonds. It seems that Baldwin was trying to build the diesel locomotive that PRR envisioned would be the answer to their dreams. The idea kind of makes sense-lots of powered wheels on the rail, and somewhat *modular* construction. You might say they were thinking dash-2 thoughts many decades early. That Baldwin seemed to be working on what they perceived their customers needed,makes sense. That they failed,maybe says that their customers' perceived needs may have been incorrect. Did the big locomotive customers of Baldwin really have that much influence on Baldwin?

[Murphy Siding]

What does "free steaming characteristics" mean, in relation to how a boiler is designed? In this context, it was used to describe the design of Beyer-Garratt locomotives. Thanks

[Kevin C. Smith]

I've always taken "free steaming" to mean that the locomotive(s) in question kept up boiler pressure easily, without too much firing-but that's just me thinking, again.

Also, re you previous post about Baldwin's Centipedes-just off the top of my head, if your big customer was PRR, that HAD to have no small influence...

[feltonhill]

C&O operated the turbines west out of Charlottesville over the Mountain Subdivision which has a profile that matches its name (see parts of the C&OHS article cited above). However, the lightweight cars coming on line at the time didn't require any more than the already present 4-8-4's and 2-8-4's in use (1948). Seems the three turbines operated for a day or two and disappeared. But their low speed TE could have been put to some use over that segment.

Most recent information seems to lay the blame for the H8's weight problem on the AMC. They apparently wanted to best N&W's Class A and weren't too particular about how they did it. So they kept adding features and iron, making it bigger and badder I suppose, without allowing for anything to be removed to compensate for the weight gain. Lima got stuck building the beast as specified and when it came time to weigh it, reality struck. Operating crew's wages had a component that took weight on drivers into consideration. So that's where the tale of secrecy in the scale house comes from. Most of this is covered in the book, The Allegheny, Lima's Finest and an article in Trains IIRC

Overmod, you mentioned S3 on C&O, I believe it's the last J3a 4-8-4's you were referring to as initial candidates for poppet valves. Would have been interesting, but it was probably best not to bother that late in the game.

As far as the double Belpaire goes, I thought I read that feasibility plans had been done for an installation on an NYC Hudson, but nothing came of it. Again can't lay my hand on the source at the moment. Too many books and files here to find that detail fast enough.

[Overmod]

Interesting about the Hudson. Do I presume that the proposal was for one of the J2s, rather than one of the Alco-built classes? (And were they going to fix the cab sag as part of the installation??? ;-})

I have no idea where that "S3" came from... or what I might have been thinking when typing it. 614 is one of that class. The point I was thinking about -- which may have been part of the distraction -- is that classes built previous to the J3as already had poppet valves, and extensive testing (and the valuable experience that came from that testing), as well as promotion and sales activity, for Franklin type B was already well in hand. I concur about it being seen as 'best not to bother so late in the game', BUT it seemed a bit odd for C&O to be building new big steam without poppets when they had been so recently willing to inve$t in the M1s, had poppet-equipped locomotives already, and presumably were aware of developments at Franklin to address the issues of the variable-cam rotary gear. (Is any of this correspondence or technical review still extant? I miss Vernon Smith more and more!)

The reason I brought up the Chessie trains was that, even though the cars were lightweight, I recall the trains themselves were intended to be very long and equipped with amenities like large fi***anks and beaucoodles of domes that were going to take the train weight up substantially compared to a typical 'streamlined' train. From the article in Trains lo these many years ago, I got the distinct impression that the marketing intent was to produce something similar to the California Zephyr experience. That might explain why C&O felt the need to go to something with the very high starting tractive effort of electric drive, but cost-effective high horsepower output at speed that would NOT have been achieved -- for high trailing load -- with either E's or passenger F's at comparable capital cost. (Don't know, but would be interested to know, the extent to which burning coal was a 'political' item of importance for this project, as we know it was over at N&W)

I think that between the V1 and M1 designs, it has pretty clear that there was a minimum size of locomotive that was going to be practical -- and I use the term in a purely technical sense! -- as a unit, and only a train capable of 'sinking' the effective power of such a locomotive would make 'sense'. Almost by definition, that train would be not only fairly heavy and long, but also need fast running -- both in relative peak speed and acceleration. (Now, I could reiterate my point about C&O using the Alleghenies, which were explicitly higher-speed freight power, in what were either by intent or result slow-speed heavy service, but it would be a cheap shot and unjustified slam at C&O to say that that was an 'institutional blind spot' that carried over to other designs of modern motive power.)

On a public computer now, so I apologize if parts of this make even less sense than "S3"... ;-}

[feltonhill]

The double Belpaire was to go on a 1938 NYC J3a Hudson. This and other information about the double belpaire and the 4-8-6 are in Eric Hirsimaki's book, Lima The History (pg288 in first edition). I believe this book has been revised and is now out as a second edition.

The first C&O poppet valve installation was on the L1 4-6-4 rebuilds and they had oscillating cam (Type A) gear. The later L2a Hudsons had rotary cam (Type B) gear, which was more accessible for maintenance than the Type A.

For those who have not heard of Vernon Smith, his views were published in his book, One Man's Locomotives. There a section in the book where he discusses the different types of poppet valves and the locomotives on which they were installed. All of us in the locomotive perfomance game owe him a lot!!

[nanaimo73]

Was the 4-8-6 just a Lima idea that they were trying to sell to C&O, or did the C&O have an interest in the double Belpaire before this proposal ?

Would Mr Woodard have been in on planning this boiler, or did the idea come along after he passed ?

Is there a website that covers Mr Woodard ?

Thanks

[feltonhill]

As far as I can gather, the 4-8-6 was a Lima idea. The RC valve gear would likely have involved Franklin Railway Supply because they were the developers of the valve gear. I never read that C&O had any specific interest in either the 4-8-6 or the double Belpaire firebox.

Woodard lived from 1873 to 1942.

[germanium]

Feltonhill, Overmod et al. - with all the experimentation going on, were there, in your collective/individual opinions, any really worthwhile improvements on the basic Stephenson machine resulting from this experimentation ?

[Overmod]

Short answer: Yes, but no.

One of the better examples of late steam design is the Duke of Gloucester, BR 71000. Bears noting that several details of this locomotive appear to have been intentionally misconstructed (cf. the firebox and front end) as built -- I stop short of saying 'intentional sabotage' but am reluctant to attribute it to negligence -- which when corrected have given what appears to be a good-running locomotive.

There is no doubt from the surviving evidence that some of the late improvements in valve gear led to better operation (cf. the ATSF testing of Franklin type B on one of their large 4-8-4s) There is equally no doubt that regardless of how 'good' these improvements may have been on the basic machine, they were sufficient for the railroads to consider them 'worthwhile' enough to retain steam. While I think a more wholehearted embracing of those 'improvements' that increased reliability and reliable performance might have kept steam around somewhat longer, it's only in the absence of the development of high-horsepower diesel-electrics that one could imagine mainline Stephensonian steam surviving in general service until now, no matter how effective it might have been made.

Might bear noting that many of the 'sophisticated' improvements that worked well on locomotives with high power that were well-maintained were not at all suitable when things began to degrade. The 'electric' Cossart gear on the Algerian Garratts is a particularly notable example imho. Easy, cheap, and familiar usually wins out in the long run, and in such a framework only the innovations that last long enough to become 'familiar' are the ones that stay adopted.

I'll be very interested to see feltonhill's take on this question!

[feltonhill]

The final improvements proposed for US steam were likely not sufficiently significant to change history.

The double belpaire boiler discussed so far may have been an improvement over a conventional firebox, but how much of an improvement? Overmod mentioned this idea earlier - would it have been enough "better"? For example, would it have been worth the manufacturing problems brought on by a new configuration? Would it have improved combustion enough to be worth the cost? Would there have been enough clearance for driver spring rigging underneath the flat bottomed combustion chamber?

Same with the 6-wheel trailing truck and the 4-8-6. Yes, it would have allowed slower burn rates and more time for heat transfer to take place in the firebox (which is where most of the evaporation took place). These are both good things. But it added weight and length and didn't add more drivers to get the power to the rail. The booster added complexity and maintenance. So, would it have been better when measured on the basis of coal and water usage per unit of output? Maybe, but how much?

The 4-8-6 would have had tough competition between the Niaraga and the N&W J which, on a quick review of available test data, seem to have somewhat similar, and low figures for coal and water usage. This is a very preliminary opinion because I just started to look at all the data that's available. It will be a real minefield to make any equitable comparison between the two. The engineering issues are bad enough, but the fans of both these locos are knowledgeable and vocal. In some respects I look forward to the project, but in others, I don't.

Poppets (RC only) would have been a theoretical improvement but again, how much better were they than a well-designed locomotive with conventional valve gear? There's no really good A vs B comparison. The ATSF 4-8-4 in Vernon Smith's book was an older, low pressure loco which was considerably improved by rebuilding. More than just the valves were changed. PRR's K4 experiments also involved not just poppet valves but also revisions to the steam circuit. So although there was substantial total improvement, which revision caused what increase?

On the other hand, the only valid A-B comparison of poppet valves vs conventional valves would have been the NYC 6023 vs 5500 tests. Problem with that is that 5500's poppet valves were allegedly undersized for the amount of steam the boiler could generate. This is not my opionon, but was covered in the Third Quarter 1993 issue of Central Headlight, published by NYC Historical Society. The results of the test showed only a very modest (almost negligible) increase in power and a 16 % reduction in coal consumption per unit of output. So that opportunity for comparison doesn't work.

Believe it or not, there is a memo indicating that Franklin Railway Supply was considering making a proposal to N&W for poppt valve installation on a J, but nothing came of it. Whether the outcome had been good or bad, it would have interesting so see what happened.

All in all, some of the improvements actually installed or proposed were probably in the right direction on a theoretical basis, but were they good enough to prolong steam use in the US? I doubt it. The final users of steam in the US had one thing in common . NKP and N&W matched locomotive capability to service requirements and operated them at their most efficient power output most of the time. They managed to stave off the early diesels the longest.

Even if final holdout N&W's management hadn't changed in 1958, the die was already cast. N&W knew it had taken steam as far as it could in their type of service. Sure, maybe the Y6's and A's may have continued on another year or so to 1961 or 1962, but IMO that would have been it. In the end, simple straightforward locomotives used at peak efficiency, serviced well and operated on a good physical plant seemed to have proved that the basic steam locomotive concept was pretty good. Any further improvements would be mere refinements that wouldn't hold off the major technological change facing the railroads in the postwar period. After the war, operational economics was the yardstick, and diesels carried a bigger stick.

[Overmod]

Two bitty points:

Seems clear to me (from circumstantial evidence) that the NYC poppet-valve test assumptions were geared toward having the same output power with better economy, not higher power with a 'more free' steam rate. I may further speculate that some of the high-speed slipping problem observable with Franklin poppets may have been worrisome enough to contribute to a decision like this. (I find it very difficult to believe that the NYC steam guys would have made a 'mistake' in steam-circuit design, just as I start looking at the lack of a significant combustion chamber on the production Niagaras as intentional rather than misguided maximization of convective vs. radiant heating (cf LeMassena in Trains in the '80s) or excessive conservatism, etc.)

Opinion, likewise: the 'weak point' of the N&W J at high speed was lateral deflection of parts of the valve gear at high rpm, often encountered as a problem with 'conventional' piston-valve drives. Part of the problem that can cause this deflection (high steam-chest back pressures including those attributable to compression; relatively high inertia of the valve components) can be solved with poppets; more importantly, use of a RC with or without continuous variable cams (or, for that matter, a properly-arranged OC drive using roller crank) remove the deflecting components themselves. (The next thing to go would probably be main pins a la MILW F-7, but it would be easy to address THAT question in turn with bigger better bearings in the lightweight rods...)

I'd expect N&W would also choose to go with some form of disc driver (or, at the very least, something like WebSpoke) if going to a higher rotational speed and power with 70" drivers. I'd also expect that some of the exhaust-stand and nozzle developments done in the '50s could have been adapted to make a poppet J run with very little backpressure problems (but proper compression for minimized shock loading on the rods, etc.) given the capabilities of good RC gear. If I may be so bold as to predict, the problem would have been 'grooving' wear of the variable cams at the most obvious 'notches' used in service, with a possible 'fix' being the servomotor varying drive that oscillates the follower slightly about an 'optimal' setting -- another possible solution being either facing or refacing of the cams with metalsprayed hard material. Doubt that it would have improved the situation with N&W steam a bit, because the decision to throw it over was made by lawyers up at the top of the company, not with reference to the effectiveness of steam power in service (not that I'm saying I would have arrived at any different conclusion at the time!)

[feltonhill]

The referenced article the Central Headlight puts it this way:

"There was a choke in the live steam passages to the poppets. This minimum cross sectional area was only 52% of the minimum area of a 6000."

It goes on to reference two drawings of the cylinder saddle casting. It further references the location of the choke to the poppet inlet valves on page 35. You can guess which page I'm missing. Time to e-mail NYCSHS and see if I can round up the missing page (s).

The choke point may have been an aw-bleep on Franklin's part ,or GSC's part, or NYC's part. The article also mentions that 5500 couldn't be run out of steam but it just wouldn't burn the coal when pushed, so there must have been a restriction somewhere.

The fuel savings were about 16% according to the article. My earlier comments were not intended to belittle this percentage, but it was not significant enough when faced with the operating cost reduction of a pair or trio of then-available EMD E7's. I changed the earlier post to reflect this recorded figure.

[germanium]

Overmod, Felton Hill - thanks for those informed answers and comments, and the effort obviously put in. I found them most revealing and of great interest, as I'm sure others will.

[Murphy Siding]

Overmod & Felton Hill, you point out some of the *issues* inherent in some of the later steam designs. Could you elaborate on some of the *cause & effect* of some of these design problems? For example, how would " lateral deflection of parts of the valve gear at high rpm's " affect performance & maintenance of a locomotive? Thanks

[Anonymous]

Quoth Overmod: "Opinion, likewise: the 'weak point' of the N&W J at high speed was lateral deflection of parts of the valve gear at high rpm, often encountered as a problem with 'conventional' piston-valve drives."

Where did that opinion come from?

The highest rotational speed credited to the J was undoubtedly the 115 MPH the 610 attained during tests on the PRR. This was just prior to the failure of the engine caused by the left valve freezing in its cage due to lubrication breakdown. If there was, as you opine, lateral deflection of the valve gear, it wasn't sufficient to cause problems at any RPM.

When I worked at Shaffers Crossing 1959-1961 there was a Night Roundhouse Foreman there named Ed Payne. Payne had been on hand with the 610 during its tests on the PRR, and described the engine failure to me. He said that when the engine was brought back to Roanoke Shop and the valve heads removed, the valves and cages were discolored (yellow) and the conclusion was that something in the PRR's water treatment cut the lubrication of the valves.

Bear in mind that the Js operated at 100 MPH every day (with, say, an inch or so of tire wear, that's 500+ RPM), but for relatively short distances. It was never necessary for the engine to be faster than it was; with the schedules and the territory on the N&W, speeds faster than that, sustained for long periods of time, were not necessary.

But the 70" wheel produced a horsepower curve ideal for the N&W's topography and heavy passenger trains. The engine was capable of running track speed (curvature limits) UP any of N&W's mountains with any of the normal train consists. It was necessary for the J to start and accelerate these trains in difficult situations, on curves and grades, in all kinds of rail conditions - Northfork eastbound, on Elkhorn Mountain; Marion, Va. on the Bristol Line, Ada, W. Va., etc. There were sharp curves on Bluefield Mountain (Jug Neck) and Elkhorn (Bottom Creek Curve at Landgraff) where the J would have to accelerate its train back to track speed after a 25 MPH slowdown. The engine was never found wanting.

I don't know enough about the Niagara and NYC's use of them to know what sustained speeds were expected of them. Did they operate at 100 MPH for miles on end?

If such speeds were not expected of them, if NYC's 80 MPH limit was strictly enforced (or even with a 5 MPH tolerance), then one can ask the following questions: 1) Could the J have done everything expected of the Niagara on the NYC with NYC's traffic and topography? 2) Could the Niagara have done everything expected of the J on the N&W with N&W's traffic and topography?

My feeling is that the answer to #1 is yes, and the answer to #2 is no.

In other words, the J's 70" drivers would not be as much of a liability on the NYC as would the 79" drivers of the Niagara on the N&W.

Old Timer

[wallyworld]

This is a fascinating discussion and I have an appreciation of the amount of thought that have gone into these comparitive discussions on late steam design. The PRR, N&W and C&O are well known for their innovations and philosophies in design but recently I saw a photograph (which I cannot relocate) of surprisingly, a B&O engine that had (from what I recall) a duplex drive. Anyone know anything about this engine?

[feltonhill]

Overmod, were you referring to the article and painting of a MILW F7 throwing a main rod? I believe it was in Trains or Classic Trains not too long ago.

From what I know, NYC was very restrictive about speeds except on tests. They used a device called the Locomotive Valve Pilot, and IIRC there was a speed recorder incorporated in this system. Most NYC locos were designed to cruise at 80-85 maximum on virtually level track (except for Albany Hill).

Some iconoclasts have argued that the Niagara was too big for Central's operations because they spent a large amount of time at part throttle. They were powerful enough that at full throttle and 80 mph, they couldn't be run at sufficiently short cutoff to maintain that speed. They contend that the Niagara didn't need to be as large as it was for Central's operations. The Niagara posted low operating costs and high mileages as it was designed (so did the N&W J) so the argument is more academic that anything else.

Although the N&W J may have been able to do anything required on NYC, it would not fit Central's load gage or weight restrictions. Maximum dimensions allowed by the load gage were 15'-3" high, 10'-8" wide. Locomotive diagrams indicate that the Niagara carried 68,750 lbs per driving axle, the highest of any NYC locomotive, so that may have been near the upper limit allowed. The Niagara was 15'-1 3/4' high, 10'-7" wide over both the cylinders and running boards and carried 68,750 lbs. on each driving axle. The J was 16' high, 11' wide over the cylinders, 11'-2" over the running boards and carried 72,000 lbs on each driving axle. So it would have take a considerable amount of "scrunching" to get the J shoehorned into NYC's clearances.

On the other hand, no 4-8-4 could do as well as the J on its home turf. It was an uncompromised design directly aimed at N&W operations. Trying to compare a 4-8-4 that actually registered drawbar pull readings (not TE, not a typo) around 80,000 lbs under 8 mph (Chart 1, readings taken from 1945 tests, highest three data points, 81,500, 78,800, and 78,000 lbs) with a locomotive that develops 62,500 to 61,000 lbs drawbar pull in the same speed range (from graph in test report of 6023) is not equitable.

See what I meant earlier? Trying to compare the J and Niagara is very difficult!!

[feltonhill]

The B&O 4-4-4-4 was mentioned early in this thread by Old Timer. It was an experimental design, class N-1, four opposed cylinders (two under the smokebox facing forward, two under the firebox facing backwards), 18" bore x 26" stroke, 76" drivers 350 psi watertube firebox. It was built by B&O in 1937 (see B&O Power by Lawrence W. Sagle, published 1964 at pg 289-290) and scrapped in 1950.

[wallyworld]

This question is directed to feltonhill.This is abit off topic. All of the engines mentioned existed prior to the advent of computer control to coordinate subsystems in an engine. I noticed a common thread in your posts as well as in my reading elsewhere, that optimum operator control of an engine fell in almost direct proportion to the increased complexity of the engine. Through my reading of The Red Devil I am aware of Porta's continuing improvement to combustion and other areas for steam long after they were considered extinct. Outside of the ill fated Ace engine(which those familiar with the book
know that Porta considered too ambitious), has anyone proposed a reciprocating steam engine with all these above mentioned improvements as Porta would have preferred? It's hard to imagine no one has pursued this as the current oil crisis appears to be worsening. Maybe there is but I am unaware of any.

[feltonhill]

Check this link

http://www.5at.co.uk/5AToutline.shtml

for the latest proposal for new technology steam. Incorporates some of the features discussed on this thread but not all. David Wardale, author of the Red Devil... is involved. Status of project unknown, last update to site was a year ago.

[germanium]

Wallyworld -
http://www.5at.co.uk/5ATupdates.shtml shows the latest position. Also contains a link to the engineering specification - no mention of computers to control subsystems.

[Anonymous]

Well, here's another slight deviation from the discussion, but the mention of the "double" Belpaire boiler in a few posts (which needed a three axle trailing truck) and the height restrictions of the Niagra and the J's of 15'+, all brings a new question to mind:

With today's 20' max height on most mainlines, could it have been concievable for a large firebox design such as the Belpair to ride ontop of a second set of driving wheels? E.g. getting the necessary steam quality/quantity necessary for superpower steam, while at the same time keeping most of the locomotive's weight on the drivers. I know this is getting to be an abstract beyond most of your interests, but bear with me for a moment.

Take the J with the 70" drivers. That's a tad under 6 feet, and we'll assume it as the minimum for speed/drawbar pull optimization. For a 20' clearance, that leaves 14 vertical feet in which to place a large firebox, so let's settle for a max height of 19 feet and 13 vertical feet of firebox to be safe.

1. Is 13' of vertical rise sufficient space to place a large enough firebox to produce the desired steaming quality that superpower loco's demanded?
2. Given that a high riding firebox would negatively alter the loco's center of gravity, what more would be needed to keep the COG in an optimum range?
3. Other than for firebox support, did superpower steamers need to have a trailing truck for stability, etc.?

If there is room for affirmation to some degree of those three questions, then why not a duplex 4-6-6-0 with the same length of frame as the J's or the Pennsy T1?

[Kevin C. Smith]

I think the firebox grates are at or below the bottom of the boiler, so the entire boiler would ride above the drivers, rather than nestled in between to some extent. While clearances might no longer be a problem now days I think you'd have quite a difficulty with the center of gravity. Anyone that knows how to calculate that sort of thing have any ideas if/how it would balance out?

[Anonymous]

You'd still run into problems with a shallow firebox above drivers. One of the advantages of the trailing truck was having a deep firebox with more volume which provided better combustion. Good examples of engines in existence today are Southern's K 2-8-0s of 1903 (630 and 722) and the Ms 2-8-2 of 1911 (4501). Both have 54 Sq, Ft. grates but the deeper firebox behind the drivers of the 2-8-2 made it capable of supplying cylinders 5" bigger than those of the 2-8-0 (22" vs. 27" - as built; both the 2-8-0s were rebuilt as class Ks-1 with superheaters, lower boiler pressure and 24" cylinders - the 4501 was built with a superheater).

Look at the difference between UP 3985 and N&W's 1218. The Challenger has a grate area of 132.2 Sq. Ft. and a 106" combustion chamber to give it adequate volume to supply its 21x32-inch cylinders; flue length was an even 20 feet. The grate area approaches that of the Lima 2-6-6-6, but evidently is necessary to get the required volume.

Being able to place the firebox totally behind the drivers allowed the 1218 to use a 122 Sq. Ft. grate in conjunction with a 114" combustion chamber; flues could be 24 feet long. And this combination was completely satisfactory to supply 24x30-inch cylinders.

There were definite advantages in having the deep firebox.

Old Timer

[Overmod]

Sorry for dropping out so long -- honeydews, don'tcha know... ;-}

OT as usual makes good points here. For a variety of reasons (e-mail me if you want more grisly technical details of my opinions on this subject) you want as much VERTICAL space for radiant gas path, even folding that gas path as much as reasonably practical via arch structure. For another variety of reasons, you don't want drivers underneath your ashpan, blowdown cocks, etc. if you can avoid it.

Meanwhile, the higher you carry the boiler, the greater the stability issues with carrying a large water mass high up geometrically on a relatively narrow wheelbase...

Where you can DEFINITELY benefit from a 20' overhead height restriction is to do what the Russians did with their relatively huge overhead clearances: separate the steam circulation from the boiler proper so that very little opportunity for water carryover in the steam is possible regardless of virtually any amount of priming, etc. You'd also be able to run the convection section of the boiler entirely full of water, which allows some interesting ideas like inclining the flues and using a separate cross-drum for steam separation -- this can give you better-defined circulation patterns in the convection section to position circulator feeds for the firebox area. You can also keep the drum or over-crownsheet volume in the firebox reliably full, which eliminates much of the traditional problems related to up- and downgrades, and helps greatly with the steam separation (and therefore vertical firebox water circulation) in various sections of the radiant zones of firebox and chamber.

(Incidentally, the additional height would also facilitate packaging of many things like EGR and air-preheater ducting, enhanced economizer (via combustion gas) feedwater heating in addition to 'traditional' ACFI-style fwh, better physical devices for steam separation (e.g. cyclones) and better isolation of many of the flow-streamlined elements of the steam path 'forward' of the steam separators. Not to mention better lagging.

BTW -- returning to those J valves just a moment: How was something in the water treatment at PRR getting all the way through the steam to compromise the valve lubrication? (I presume the substance was a detergent, perhaps of a nature and in a quantity sufficient to give some 'proper' value of boiler-water alkalinity without additional 'ingredient$...?) Seems to me that the only way this could happen would be if substantial -- perhaps VERY substantial quantities of water were making it all the way through the superheaters etc. to the cylinders, but presumably the water treatment used by PRR would also be intended to reduce priming/foaming to a minimum. I do not know whether PRR or N&W personnel were actually at the controls of the locomotive during the test, and don't know if 'whoever they were' were watching for effects of water carryover, etc. at what had to be almost terrifying steam rates (and mechanical action!).

My comment about lateral deflection was not intended as a criticism of N&W design, and I apologize to all concerned if it came across that way. It is, of course, entirely possible that the valve seizure precipitated the valve-gear failure during the PRR test and the failure mode of the gear only mirrored that in the other cases I have heard reference to. The situation is that there is relatively little structural resistance in some valve gears to lateral deflecting forces, the assumption being that there will be relatively little resistance to motion posed by properly-designed valves, but under some conditions (IIRC usually related to drifting or compression braking, presumably without Okadees) on SOME locomotives, there could be very large unanticipated back pressure on the valves which put very high loads on the v.g., causing it at least occasionally to deflect (cyclically!) I had seen some discussion from what I considered plausible sources that this was the cause of the valve-gear failure that terminated the N&W test; if that is in fact wrong (and I think I'd take the word of someone who was there!) I stand corrected.

I also concur that a J could have done essentially everything NYC required of a general passenger locomotive -- I've seen reference to the idea that NYC "should" or "could" have used at least the later (e.g. L4 class) Mohawks on just about any practical service -- and even a J with a smaller boiler, etc. to fit NYC clearances would have been substantially better than a Mohawk... ;-}

No question, either, that N&W had the 'right' power for what they needed, which did involve 100mph operation but only for relatively short sprints, but also considerable periods of easily-controlled high tractive effort. I still think, though, that a version of this locomotive with properly-implemented variable-cam RC, and an adapted waffle-grate or rosepetal nozzle, would have been VERY interesting...

[feltonhill]

I have some information on the PRR tests of N&W J 610 from a couple of PRR memos that were written just after the tests were concluded. Here are some exerpts from a memo from L. B. Jones to H. W. Jones (PRR CMP) on 1/6/45;

Speed - "The locomotive is euipped with a speedmeter which we did not calibrate but speeds of over 100 mph were reported. On one trip an average speed of 94 mph was maintained for 45 miles. This locomotive has valves of large diameter and long travel, giving large port openings at short cutoff. It is this feature that made these high speeds possible in spite of the fact that the diameter of driving wheels is only 70 in."

Valve failure - "On December 10th [1944], the locomotive was cut off at Fort Wayne due to failure of the valve motion when the gear connecting rod and the eccentric rod were bent, the eccentric crank loosened, and the reverse gear quadrant damaged. Cause assigned for the failure was poor lubrication but there is no doubt that excessive contributed."

610 made 24 runs total, 12 westbound and 12 eastbound. The locomotive was out of service from 12/10 to 12/18/44 to repair the valve gear.

PRR was very complimentary of the J, in contrast to some stories that they criticized the loco just because if wasn't one of theirs.

There's not much direct evidence on this series of tests ( I have only six pages, five PRR and one N&W), but there's enough to get the picture and support what Old Timer has relayed from Ed Payne.

[Overmod]

OK, utter humility time here (and how appropriate, on the Lord's Day...)

ASSume I know nothing about the actual PRR testing:

1) If the valve gear failed multiple times -- once on the test plant, and once on the road, what might the dates be?

2) Was there 'pre-existing' damage from running on the test plant (to high cyclic rpm at presumably high load) that caused the observed subsequent valve-gear failure?

3) Was the SOLE cause of the valve-gear-parts failure, reported immediately above by feltonhill, the seizing of the valve in its cage (as seems likely from the crank being loosened) -- this seems to be the thing missing from the anecdotal accounts I've been hearing in the past...

4) How long was the interval (if any) between the testing on the stationary plant and the time of the (reported) road failure? What sort of feedwater treatment was in use during that interval, and what kind of care against carryover, etc. might have been used (or ignored) while the engine was in that road service?

5) I presume that the word "speed" is missing from the 'there is no doubt that excessive contributed'. If it's something else, tell me. (94mph average for 45 continuous miles is speaking for itself!)

I do not find it surprising that 'long-travel large-diameter' valves that did not have tailrod support might encounter difficulty when running with compromised lubrication. What I'd like to see is the cause of the 'yellow' color on the components, if that is known. If it's 'overheat' oxide (thinner interference layer of 'bluing') then the oxygen would have to come from somewhere, ideally NOT the steam flow. Would the steam be dissociating excessively (to hydrogen and oxygen) at these temperatures and conditions? How did the class J handle snifting?

[feltonhill]

No N&W J was run on the Altoona plant. If it had, several of us on this thread would probably have heard about it. I'd have spent a lot more time and $ than I did to find what tests I have just to get my hands on that one. But, AFAIK no J at Altoona. Valve gear only failed once - on the road.

[Anonymous]

AFAIK, there was only one other valve gear failure on a J, and that was due to shop maintenance, as follows:

The exhaust ports of the J's power reverse were choked down so that the engineer could not make a quick movement with the reverse lever. No matter how hard you tried to pull or pu***he reverse lever, it would only move so fast (having had the privilege of moving J's reverse levers around the roundhouse, I know this from experience).

The J in question, the 600, was handling train #45 (the TENNESSEAN) and was passing through the dip at Hayter, Va., (a little less than 19 miles east of Bristol) at the speed limit of 65 MPH (according to the crew who, of course, wouldn't admit to violating the limit, but Hayter was good running ground and the speed was probably more than that by maybe a good ten to fifteen MPH). Unbenownst to the engineer, the choke in the front of the power reverse cylinder had dropped out due to not having been tightened properly at the last inspection. The engineer sought to drop the engine down a couple of notches to come out of the dip, and when he unlatched the reverse lever and applied a little pressure, she went right into the corner with him. The left eccentric couldn't take it and departed the locomotive, taking with it the turbogenerator (located on a bracket above the left drivers) and about half the running board skirt. They found the turbo 3000 feet from the point of the failure.

This is the only other VG failure I've heard of on a J, and it wasn't the engine's fault.

Old Timer