Extreeem Steeam ! About Unconventionals - Part II

[Overmod]

Come on, guys -- where is everybody?

 

Juniatha

...in my 'walk on the wild side' type of private (pre-)engineering me too I had put up a concept of a Triplex , and just because of this very point - supplies vanishing & and tender adhesion mass alleviating - I had carried it just one step further , figuring if you can have two axles coupled in front of the main driver you could as well have the same on the back side of it - makes for a 10 coupled unit ...

This is not all *that* wild.  The basic Triplex idea can be 'better' than the original motor-tender-with-2-8-8-0-stability if you attend to some proportioning (and used some care at the joint under the cab): it might be thought of as one-and-a-half Y6s, with only the leading and trailing two-wheel trucks at the ends.  Give that locomotive Chapelon-style proportional-IP injection (US jingoists, think of this as a more controllable version of the booster valve Louis Newton described) and there is no particular reason why, compound, it couldn't be worked at normal 45-55mph freight speed.  Principal issue for such a thing would be a tendency to pick the 'leading' driver flange of the rearmost engine, but if you were really concerned there are approaches that could be used to address this with minimal loss of potential adhesive weight.  (And in case you start fretting about water and fuel mass on drivers, there are ways to address that too...)

You'd think I configured it to be a 2-8-8+10-2 ?  Naw-naw-naw , sir - not with my way of thinking ! 

Actually, if any engine were to remain eight-coupled, it would be the one at the 'wrong' end of the locomotive (no good practical way to use the exhaust even with GPCS steam injection; trouble with the longer rigid wheelbase; relative loss of usable TE, etc.)  What you'd do would be put two ten-coupled engines up front, like the infamous Virginian locomotives, but taking advantage as you note of the relatively balanced thrust distribution from five axles with drive on the center one.  If you need inspiration on the possibilities of this, look no further than the PRR I1, which would spin happily up to 50mph although admittedly with oxcart ride...

Ten coupled once introduced , I immediately thought "Why not have it on all the three engine units ?"   Mind my previous mentioning preference for identical sets throughout . So I ended up with a 2-10-10+10-2 - all simple expansion. 

There are a couple of thorns in this particular idea, the most obvious being the lateral-motion implications for the two rear engines.  Too much indeterminacy in guiding with lateral-motion devices on adjacent sets of driver axles... and you dare not have too much lateral in those outer rods, too.  (There is also some phase-breaking advantage to  having more sets with a smaller number of independent drivers, as in the theory of the duplex locomotive, but we can take that point up later).

The real problem is: Where would you find a single train, in that era, which would take advantage of so much over-the-road power?  Haul it, and drawbars start popping like corn, or you start to get stringlining fun.  Push it, and... you thought Big Liz pushed cars all over the place?

And in any case, six simple-expansion cylinders are going to give you nastily peaky power, and require great care and perhaps great effort in assuring the 'right' amount of steam-generating capability, while the rear-engine exhaust is not terribly useful BUT represents an enormous water-rate loss.  You might think you can condense this steam in the tender and canteen water... but you'll be aghast if you actually run the numbers, even with saturated exhaust at atmospheric pressure, and see how quickly the water heats up.  The problem with the original Henderson Triplex wasn't that it was compound, just that it wasn't proportioned correctly... make arrangements to keep the LP chests proportionally pressurized, not an impossible exercise even with '40s tech, and you have no need for fancy throttle artistry... even though every engine pulls its full share of the weight. 

Sure enough there was that vanishing adhesion mass problem again.

Well if it troubles you so much, use the Concorde solution.  Use the 'tender' as adhesion mass, and pipe the actual water (and convey solid fuel, if necessary) to 'reservoirs' on the tender proper (where the injector feeds, stoker worm, etc. are located).  You don't even 'need' to do the transfer until you have an actual requirement for full adhesion...   

... Ah , but not with young June :  "Let's fit individual throttles , one for each engine unit , three in a bunch to be grabbed together as the throttle levers in a BUFF yet individually adjustable to make the most of adhesion ..." 

 

I suspect this isn't exactly the right 'metaphor' for the kind of throttle action you'll actually need.  Throttles on a '52 are made as they are because the differential action is clear and fairly easy to comprehend.  Just try, though, having to pull back ONE of the fistful, in the middle, to an exact proportion you don't know and can't really sense through the seat of the pants... particularly when the 'right' way to do traction control via the steam is to modulate the valve gear, not the throttle.

This isn't trying to argue you shouldn't have separate throttles for multiple engines -- something like that would have been useful on, say, a four-coupled duplex to allow 'derating' the relatively slip-prone front engine at All Those Times It Wanted To Cut Loose.  But making the engineer, who remember is no mecanicien, have to dance with three of 'em all the time isn't going to make you a popular girl ...  ;-}

The throttles were to have on spot hydraulic power operation so moving the levers didn't take but light effort as you effectively just opened the power valves of hydraulic lines.   The idea was, you would start opening the combined levers as one yet just to an extent sure to be supported by adhesion and then as the engine began to move you'd open up further on the second and first engine units...

Let me interrupt momentarily: there's a reason for long throttle travel completely distinct from effort; it's to give a sense of fine control with fairly coarse physical haptics.  What you're doing is far better handled with the analogue of aircraft trim: allow adjustment from 'baseline' for one or more engines, after which moving the single lever automatically gives you the proportions.  (Slip being handled several ways I'll take up in a bit...)

You also will have a certain amount of fun with the three reversers, which is where I think I would put the actual proportional power adjustment used for running trim.  All three of THOSE would be ganged, but with the ability to tell 'at a glance' if one set were slipping (this can be as simple as a differential with switch running off the relevant valve-pilot sensor, illuminating a slip light over the relevant reverse lever) and then ease back the affected lever in the 'gang' as needed... without compromising what the other engines are doing, or requiring foreground-attention tinkering with settings after the slip has been arrested to keep the engine at full achievable power.

 

Since the locomotive was purely intended as a [grade] pusher , tender capacity wasn't all that large anyways while no attempt at low empty tender mass was to be made.

This in an age where large pushers weren't particularly useful (see above, but also recognize that there are very few grades where a normal, union-sanctioned crew would have a large enough road engine to pull something requiring so very large a pusher).  Three eight-coupled engines is overkill already.  On the other hand, there MIGHT be a potential place for something that size as a road engine, on consists that a given railroad could optimize for great size...

...but now we get squarely into the fact that contemporary trains weren't up to that kind of TE.  And even before we get into doing DPU with steam locomotives, we have to look at something else: siding and block length.  Even before we get into yarding monster trains effectively. 

Arrangement of live and exhaust steam piping was a point I was particularly content with:  it featured a central live steam pipe with short ball type articulated sections with pipes branching out to cylinders ( it was a slightly more complex system than my words describe it here)

I will tentatively, very tentatively note that late articulated practice was to keep the steam feeds to the cylinders separate, not derived off a common pipe.  See how and why there were so many pipes on an Allegheny, for instance...

 Exhaust lines from cylinders would combine in reverse to form a steam jacket around the central live steam pipe thus insulating it against the outside...  

Now this is a nifty idea UNTIL you look at the actual volume that needs to be provided to make the trick work properly.  And the heat loss between a fundamentally unshielded steam line (with high superheat required for all the usual reasons) and exhaust that is already starting to think about condensing, certainly enough to provide a nifty amount of evaporative transfer from the hot line.  So yes, you need to maintain insulation on your main steam lines inside the 'jacket' pipes, just not as much as you'd need to lag them.  But those pipes are going to be VERY fat on the outside, perhaps getting increasingly large the closer to the physical exhaust you go.  In the '50s we didn't have simple ways to figure out how to fill available volume with a properly-area-ruled duct... or to fabricate pressure pipes that were noncylindrical in cross-section inexpensively.  Not saying this was a 'bad idea' -- just that you have to watch the detail design and decide if it's worthwhile....

Each of the short articulated sections also was to contain the throttle to those very cylinders.

But these are hydraulic, aren't they?  Especially so if you have Wagner throttles at the ports.  It was a reasonable analogy to the way throttle rods were run through boilers, but running the rods through variable exhaust steam isn't going to get you the same inherent compensation for temperature length change... and you now either have the entire linkage to the throttles buried out of sight and in steam flow (not an optimal lubrication situation, and sticking any part of that linkage might be particularly dangerous for control purposes).  This to me makes the maintenance of something as elegantly simple as the Q2 butterfly valves look simple by comparison [/sarcasm].  Perhaps best to co-locate the throttle hydraulics with the control for the power reverse, along the underside of the boiler or running boards in a well-protected location, and handle trim and adjustments with servos.  (If you haven't researched Meier-Mattern valve gear, you should do so now...)  

You know , at that time I was not all too much concerned about how such an assembly could be put together in actual construction practice and what it would mean to extract one unit for an overhaul in case of wear or breakage.

And, oddly, aside from the considerations above, there is little additional difficulty in 'extracting one unit for overhaul'.  Let's look at this:  (1) you wouldn't be extracting the center engine for any reason; it's attached to the boiler.  (2), the forward engine is just like a Mallet engine, pull the flexible steamline joints and disconnect (and cap) the various control lines... there isn't any need to maintain geometric precision, or require careful measurement of rods or linkages; (3) the trailing unit goes with the tender, and again involves little more than taking pipes and ducts loose and disconnecting lines.  You are sooner or later going to have to determine how to handle the bearer-plate issue (it was solved on high-speed simple articulateds only by machining the forward bearer PRECISELY so that the locomotive acted as a long rigid base in the vertical plane -- see Bruce's book if you don't believe me) for all three engines, on a railroad that doubtless contains a plurality of vertical curves... if you don't know where they are, this locomotive will unfailingly find them...

Now, in practice, are you going to have large numbers of 'spare' reciprocating engine units ready to be swapped out at a moment's notice?  Of course not, for much the same reason as you wouldn't shut a motor tender off when not needed for starting TE -- it's just too much capital to be tied up, when it could instead be assembled into more locomotives and used to produce revenue.  Everyone that ever ran those numbers came to a similar conclusion.

The effects of time did not loom too large in my thoughts back then - yet it was not without charm and I think it wasn't a bad idea in principle - although it clearly would have demanded more in-depth attention to details , mostly about manufacturing , design and shaping of parts and ease of mounting / dismantling

.

I presume this is servicing time, and you're trying to limit the duration of any required service -- in many respects a noble and important goal.  I am not sure, though, that making ease of engine replacement a high design priority is necessarily the 'best' way to go.  Better, I think, to make it less likely for the engines to wear or fail, in the first place, and then do all the rebuilding and PM in parallel when the engine goes in for 30-day (like X-day) servicing, when it can be dismantled and the parts repaired or replaced in parallel.  You may have larger service-team personnel requirements, and perhaps a larger shop than you'd need without 'triple quantization' -- but the service time requirements, and hence ability to schedule work better, would be better known.  A useful thing to look at in this respect is the difference, the actual important difference, between the Y6 and the Y6b with respect to reliability and maintenance...

Maybe I'll put up that old side view picture of back then , if you'd care to see it?

Yes, please do -- and while you're at it, put up pictures of all this more current stuff, too.

Regrettably, we have all too few of Porta's brainstorming and sketches where they can be seen and enjoyed.  Don't you make that same mistake!

Now let's see some more discussion on this topic.  I haven't seen anyone take up the issue of Besler tubes in a practical boiler, for example.

 

RME

 

[Overmod]

I'll get to the power throttles -- btw having decoded Porta's 'Waggoner' throttle references -- in a bit.

Forced (as opposed to induced) draft has been tried on locomotives, most notably in the Velox boiler.  Principal technical issue is fuel feed; secondary issue is the incredible mess that results from any leak in the pressuretight casing (the amount of blankety-blank language from engineers on power-station boilers with pressurized firing can be frightful!)

Biggest thing about  use of forced draft is that it's only economically 'optimal' for long, sustained runs at high output.  Rail service isn't like that; you're constantly varying steam demand even in general train-handling, and things go quickly out the window if you regularly take siding or experience slow orders.  The armchair thermodynamicists never seem to grasp this point adequately.

A better case might be to use a small measure of variable forced draft in conjunction with a GPCS-style firing system (where the gas-tightness of the combustion spaces is required by the system).  One thing here is that you would need proportionally larger radiant absorptive surface to take best advantage of this (although I grant you the situation isn't as bad as methods that rely on increasing gas speed)  Question is whether the relatively small gains in firing efficiency warrant the capital investment, training requirements, and so forth in general service, as well as the perhaps-severe problems if the added equipment or systems break down or their performance degrades 'ungracefully'.  (Hint: not likely in the early '50s, and even unlikely in the relay-logic era of the TE1)

You still need to maintain induced draft at the front end for a fairly wide variety of reasons -- if you do this with a fan or blower, place it where the South Africans did, whether or not using the 'turbine drive' methods in the ACE 3000 patent.  This simplifies the forced draft to simple servo following (and not incidentally allows economical 'automatic action' over a wide range of speed and load if the front end is done correctly).  I won't go into technical front-end optimization here, but you should note that there are reasons, especially with coal firing, why the convergent-divergent design of the gas passages from the front tubeplate/superheater headers up to the screens  is arranged as it is on modern American power.

Part of the issue of stoker fires, etc. would have been handled the same way firing efficiency was traditionally maximized: crew knowledge and experience.  Men who 'knew the road' and the characteristics of train handling would understand when to start reducing their firing in anticipation of cresting a grade, or responding to a train order.  At least theoretically, a simple slide at the top of the elevator would provide enough of a heat barrier to prevent a stoker fire, particularly if there was a working arrangement to reverse the elevator screw direction.  Other approaches, such as nitrogen or CO2 pressurization, would have been applicable in the historic period (diesel design, in particular, had already provided standard CO2 bottling for fire suppression by the late '40s)

On the other hand, it certainly seems to me that chain-grate firing would have been tried to more extent on reciprocating power, and even the arrangement on the TE1 neatly covered the issue of firing control during rapid unpredicted power excursions.  (Yes, there's a 'sluicegate' slide at the rear of the firebox that closes down to restrict coal feed...)

With all due respect, you're going in orthogonal wrong directions with the water issue.  Yes, you want to restrict the effective water rate.  The PRR V1 turbine with Bowes drive was the right answer to almost all the motive-power issues in its era... except that the water rate was so high at its 8000-odd hp that even a large tenderful was only expected to run about 130 train-miles.  In the era where all trains had to stop at all division points, and all division points had water, that was not SO bad as it would be today, but imho this was one of the most significant reasons that PRR went with F units rather than turbines...

But on the other hand, there isn't any real reason to throw away water with cylinder cocks and blowdown.    I know that you understand boiler-water treatment well enough to know how to keep blowdown minimized on working locomotives -- certainly continuous blowdown was a stupid thing even in the era of relatively well-provided, relatively continuous track maintenance.  My own opinion is that better steam separation is a better answer, where it can be done, than Porta-McMahon-style antifoams, when you want to keep a high continuous level of TDS and perhaps emulsified oil in the boiler to preclude scaling and other maintenance horrors.  But blowdown of sludge isn't the 'right' answer: keeping sludge from settling out is.  Reserve the blowdown for maintenance stations!

Condensing on reciprocating locomotives is, in my unhumble estimation, not warranted in the absence of effective recompression (which is largely intended to benefit the pressure cycle in the engine, and not for 'thermodynamic' reasons alone).  Steam-to-air heat transfer at the sheer power level required of a modern reciprocating engine -- even one with steam-motor drive -- just isn't worth the problems it costs to implement.  Get the water in liquid phase and the thing becomes more 'do-able'.  (I'll keep the ultrasupercritical-motor design for a different thread, as I believe the appropriate alloys and techniques hadn't been commercialized cost-effectively in the '50s...).  And therefore we need to fall back on getting better effective heat transfer to the various flavors of 'economizing' to decrease the water rate on the horsepower side, and minimize the required fuel burn on the gas side.  I suggest examination of the Snyder combustion-air preheater, the Cunningham circulator, and some flavor of Franco-Crosti-like economizer FWH, at a minimum, before even starting to look at condensing -- any energy you can recover in a Rankine cycle is better than throwing it away -- and when you do start with condensing, start by maximizing effective back-pressure reduction for high mass flow at the cylinders.

Likewise, I'd have expected you, of all people, to have gotten the lesson about heated cylinder blocks from Chapelon.  Take the approach a stage further, and use overcritical boiler water in a pumped circuit through comparatively thin lines to keep the cylinder and heads at 'equilibrium' temperature with the boiler water.  This is well above the phase-transition temperature for nucleate steam near the end of the stroke (where nucleate condensation produces its evil effects) and hence there will not be any condensate to blow out of cylinder cocks when starting.  Now isn't that a more elegant solution?

Now, turning to the individual port throttles -- as it turns out, these are Wagner throttles, and you can save countless hours of trouble by simply reviewing Fritz Wagner's various patents (which basically apply principles of fluidic amplification, to use the more modern term for the principle, to proportional throttle control).  There is a certain amount of fun, and a somewhat greater amount of concern about plaintiff's bar, in coordinating four separate Wagner throttles for HP admission, but this is less of a problem than you might at first think because there is little effort required to move the actual linkage, and techniques for coordinating multiple throttle lengths existed perfectly adequately before the '40s.

The short answer, of course, is that the principal steam throttle doesn't have to be via the Wagner throttle. For starting, as an example, you want to restrict steam pressure at the ports (by modulating the throttle) rather than trying to tinker with the cutoff (which will aggravate slipping as well as noncircular driver-tyre wear) -- and this would best be done with a single throttle in the usual place, downstream of the superheater so there's no separately-fired pressure vessel consternation, probably a typical poppet multiport type.  (This also allows the kind of sliding-pressure firing that allowed NYC to demonstrate 2-8-0 level fuel consumption on Niagaras when doing 2-8-0-level HP output... but that's another story...)

If you are concerned with bypass or compression issues:  (1) use something like a Trofimov valve (and yes, that's the spelling I think you all should use in a modern context at least), and (2) use adequate valve and reservoir capacity to allow Jay-Carter-style modulated compression independent of valve settings and port design. I'd be happy to go into my little 'take' on best approaches for drifting or providing compression 'dynamic' braking on reciprocating locomotives, but that isn't relevant here.

A fun thing to consider (in my somewhat warped estimation) is to coordinate the Wagner throttles directly with Cossart-style drop valves.  (Which lets you use salmon rods to improve longitudinal balancing, etc.). For you diehards that want to keep the horizontal piston valves, Trofimov or otherwise... I believe the Berry Accelerator gear that Richard Leonard has found will also allow the salmon rods to work in the correct phase for longitudinal balance... and it's in period!

That's long enough for this post, for now.  

RME

[Paul Milenkovic]

Pressurized combustion in a boiler?  I think that was tried and had a name -- Velox boiler?  Did it have success in any applications?

[Paul of Covington]

   MidlandMike and selector:

    While the exhaust back-pressure would be higher, the pressure of the air entering the firebox would be proportionally higher, so the flow would be essentially the same.   The goal is to increase the density of the air, which would cause more intense combustion and greater mass to carry the heat and transfer it more efficiently through the firebox and flue walls.   The part I sometimes wonder about is using the exhaust pressure to run the blower, as it seems a little like perpetual motion, but then a jet engine does a similar thing.   The exhaust pressure could be used to run other auxillaries, and other means could be used to help provide the pressure.

[Paul of Covington]

   Juniatha, I'm glad you didn't think my idea was crazy.   I think I started down this path years ago, after reading about the condensing locomotives in South Africa.  (I think.)   It seemed to make more sense to put the blowers at the input rather than at the exhaust, then it grew from there.    The only way I could think of to make it work with coal would be to seal and pressurize the whole coal bunker and stoker, but that would complicate matters if manual intervention would be required (e.g. breaking up clumps).   I wanted to include condensers in my plan for my dream locomotive, but the plans kept getting more complicated with so many heat exchangers and inter-coolers, that I was afraid I was approaching perpetual motion.

    Now I'm intrigued about your power throttles next to the cylinders.   Are you talking about replacing the conventional valves with valves controlled by a combination of, say, throttle-position, piston position, speed and other parameters, similar to the way modern car engines have computers controlling the mixture and ignition timing?    It might provide a more economical use of steam.

   Here is another crazy idea (I think I have too much idle time).   How about something similar to a diesel cycle:  eliminate the whole boiler, have cylinders with massive heads that are heated to a very high temperature and inject water against these heads which would vaporize instantly, providing thrust.   Of course these would almost have to be single-acting pistons, as I think providing a seal for the piston rod at such a high temperature would be difficult.

   It's fun to think up things like this when you don't have to contend with things like reality.

[Juniatha]

>> Why not pressurize the whole firebox and flue system to maybe two or three atmospheres?  <<

Paul ,

there is something in it , for sure , as to intensify heat transfer - and it could also allow for smaller tubes and flues sections in relation to length between tube plates .  

It would ask for some thinking about details with oil-firing but present no major problem in that combination .

I had come to a similar idea in order to solve the problem of rapid abrasion of blower blades with coal fired condensing locomotives ( uhm , steam locomotives , that is - *g* )   All those locomotives featured blowers in replace of ejector type of draughting and invariably mounted to the same position .    My thought was , ok , if you have an exhaust steam turbine driving a blower why not put the blower to the other side , i e upstream combustion , to blow fresh  air through the grate instead of sucking dirty combustion gases out of the smoke box .   This would provide some degree of over-pressure above atmosphere , no ways as much as you mentioned , however it would make a difference in gas to wall heat transfer intensity .

With coal firing - that's only where the problem of abrasion of blower blades was severe - that disposition would present one immediate problem : how to tighten ashpan and firebox feeding to prevent escaping of air and - more importantly - flames .   Clearly , any detachable feed line , any type of fire doors inside cab were ruled out .  While a stoker feed line with archimedes screw filled with coal could probably provide enough resistance for loss of gas flow to become negligible in view of boiler performance , escaping flames and hot gases would still present a very real risk of igniting coal before it had reached firebox .   Again , this problem could turn out rather theoretical as long as feed was kept up at pretty progressive working rates - yet it was another thing when stoker should be substantially slowed or stopped while engine still goes on working hard , say to finish an up-grade pull , just going over the top .  Some sort of positively closing flaps at the firebox end of feed would have to be provided and they'd have to be working automatically and absolutely reliable , too .   Plus , for safety a sprinkler system would have to keep coal wet in continuous spray of water , independant of working rate of stoker feed - much the way as blower has to be 'on' at all times with oil fired boilers to prevent blow back hazards by starting burners or by irregular combustion .   The loss of water would not really be prohibitive , not would it present any loss of heat worth mentioning if fed by a small electric pump taking water from tender directly ( my idea for any concept of steam loco tending towards more sophisticated demands of auxiliaries would be to have an enlarged turbo generator provide enough electric energy to power electric motors for diverse smaller auxiliary devices )   On the other hand you'd have to forget about thinking a condensing steam locomotive wouldn't have to take water at all - mind  water lossed just by cylinders having to be drained by opening cocks when starting cold engine , boiler blow down valve and other sources .   Quantity of water refilling could be down by some 90 % yet never become nil this way - questionable if it needed to with fully effective feed water treatment .  

But , hey :  my putting power throttles into live steam feed line next to cylinders raised one major problem which only came to me later on - and it provided something to engross thinking about until I got at least some flash that  promised a working solution - although still incomplete .

What was it ?

Curious regards

Juniatha

[selector]

A blower/turbo in an internal combution engine works because valves open and close  which permit the charged or supercharged air from escaping as the piston compresses it further.  There is no conceivable mechanism which would allow the same thing to happen in a firebox.   You must allow the hot gases to flow readily through the flue pipes in a conventional steam locomotive, but in order to supercharge the firebox, you'd either have to close the flues in the rear flue-sheet or at the far end, at the smoke box.  Otherwise, everything just goes faster, and more of it, if you blow into the firebox to raise pressure.

Improved flue design and firing might make a difference, and even the use of improved blowers in the smokebox, although the latter were used routinely right from the start.  Except that they drew firebox gases through the flues, and didn't force it through them.

Crandell

[MidlandMike]

Paul of Covington

   Here is an idea which has been kicking around in my head for some years, and maybe someone can expand on it or tell me why it wouldn't work:  Why not pressurize the whole firebox and flue system to maybe two or three atmospheres?    This would increase the intensity of the combustion, and the higher density of the combustion gases would cause a more efficient transfer of heat through the firebox walls and flues.   The pressure at the smokebox would be used drive a turbo to compress the air for combustion, providing a sort of feedback similar to that provided by the exhaust steam in conventional locomotives.   Of course, this would be a bit tricky to do on a coal-burner, so it would almost have to be done on an oil-burner.

It would seem that to hold back-pressure in the firebox, you would have to severely restrict the exhaust outlet.  This would also increase the partial pressure of the exhaust gasses, which I can't imagine would help with efficient combustion.

[Paul of Covington]

   Here is an idea which has been kicking around in my head for some years, and maybe someone can expand on it or tell me why it wouldn't work:  Why not pressurize the whole firebox and flue system to maybe two or three atmospheres?    This would increase the intensity of the combustion, and the higher density of the combustion gases would cause a more efficient transfer of heat through the firebox walls and flues.   The pressure at the smokebox would be used drive a turbo to compress the air for combustion, providing a sort of feedback similar to that provided by the exhaust steam in conventional locomotives.   Of course, this would be a bit tricky to do on a coal-burner, so it would almost have to be done on an oil-burner.

    

[Firelock76]

WOW!  This is SOME thread we got goin' on here!  Excuse me while I put my eyeballs back in the sockets and reach for the Seagrams to ballast myself a bit. 

[JamesP]

Paul Milenkovic

You or some others could always build it in "zoo gauge" (12" narrow gauge or whatever is popular for amusement parks and zoos featuring steam railroad lines traversing the ground).

While I realize that the reference to 12" gauge was just an example of a park train size (most are 24" gauge these days), I still had to smile at the thought of Juniatha's 2-10-10+10-2 built as a 12" gauge locomotive... that would be a monster!  Aside from the time and money it would take to build it, I can only think of a single 12" gauge railroad in the US where it would be able to run.  Most 12" tracks (mine included) are populated by relatively puny 4-4-0's, with tight curves to match.  The normal hobby gauges (7+ inches) might be a better choice - there are a lot more of those tracks available - and quite a few articulateds in the 7.5" gauge realm, including at least one Erie Triplex:

http://www.youtube.com/watch?v=7JemEIAnww8

Nonetheless, I would heartily encourage anyone in who is interested in increasing the efficiency of steam locomotives to get involved in live steam.  It would certainly be a more economical way to experiment as opposed to full size steam, and it would be very interesting to see the various ideas put forth in this thread actually operating, albeit in a smaller size.

I thoroughly enjoy these types of threads, even if I'm not an active participant.  Always fun to read Juniatha's posts!  Thanks to all for the interesting reading...

 - James 

www.nfrailroad.com

 

[Juniatha]

>> what sort of (flexible?) steam pipe connections  <<

Principally same variety as in Mallets ..

|  = J =  |

[MidlandMike]

On the Garratts, what sort of (flexible?) steam pipe connections did they have at the articulation points?

[Juniatha]

Hi Paul

 

Model building :

Thanx for the proposition , Paul , yet quite certainly I'll never have time to realize a running steam powered model - nice idea though and I would supply drawings and additional help to someone who would indulge in building one .

 

Pictures of Garratts with live steam pipes along side of frame

(4-8-2) - (2-8-4) BGB (Beyer Garratt Belpaire) - see pipe emerging from smokebox as usual and going along bridge frame towards rear cylinder

http://www.members.optusnet.com.au/~harburg/locos/DeGarratt-17-aw.jpg

(4-6-2) - (2-6-4) - see live steam pipe monted above bridge frame beam

http://users.powernet.co.uk/hamilton/

(4-8-2) - (2-8-4) - see live steam pipe running along boiler side with exhaust steam pipe returning along outside of bridge frame beam

http://www.flickr.com/photos/sierpinskia/5916460832/

(4-8-2) - (2-8-4) - see live steam pipe running along outside of bridge frame beam ;

http://www.scotlandforvisitors.co.uk/garrett.jpg

(4-6-2) - (2-6-4) - Narrow diameter live steam pipe with at best 'T-shirt' type of insulation running along outside of bridge frame beam .   Clearly , the PLM Algerigue 231-132 series passenger Garratt with Cossart poppet valve gear was a locomotive intended for permanently warm Southern Mediterranean / North African climate without severities of Continental European winters .

http://enginemanwook.files.wordpress.com/2010/09/plm-algeria-garratt-231-132bt1-societe-franco-belge-raismes-france-2697-1936-g-hamilton-collection-0909102.jpg

(4-8-2) - (2-8-4) drawings - see live steam pipe specifically running along both sides of bridge frame beams while exhaust pipe is located centrally and returning even below ashpan - arrangement which looks a bit tongue-in-cheek

http://www.johnnyspages.com/rail_dittys_files/garratt_diagram.gif

On most Garratts live steam pipe neither looks of ample diameter nor seems all too well insulated .   Live steam pipe placed along outsides of bridge frames doesn't look too elegant as a design nor too thoughtful in the event of flank collision ;  since a pipe is not a component normally plagued by wear or failure it should better have been run in protected if hidden space behind frame beams .  Because of considerable distance between superheater header and rear cylinders , substantial losses in steam pressure and temperature should be expected , especially since most of these Garratt designs definitely lacked attention to inner streamlining not to mention Chapelon level of scientific care taking for free steam flow .   What helped saving temperature losses for sure were hot sunny days in Africa - although I haven't heard of reduction of performance in the dead of a cold stary night ...*g*

It has been suggested the Garratt - possibly developed into a Mallet-Garratt , or as I'd call it : a Mallgarret - could have helped US railroads steam traction with heavy loads on ramps .   However , as I had commented in an earlier posting , these suggestions didn't take account for accompanying problems with design of suitable boilers to produce necessary quantities of steam - a problem which already had become virulent in the largest of x-8-8-x simple Mallets and a key reason for their less than dramatic specific power outputs (or higher total mass per ihp or kW ) in relation to large single frame locomotives such as 2-10-4 and 4-8-4 .

Also , each further engine unit added meant no small multiplication of mechanical complexity of a locomotive design , mind the locomotive had to be made flexible enough to pass through superelevation ramps leading into curves , through curves / S-curved switchwork , over humps and sinks in rail alignment without complaint and had to be ready to apply power of all engine units to the fullest in any position and at any point of the line - beyond two engine units related engineering problems tended to really score .

Regards

 

Juniatha

p.s.:

I have edited my previous posting on double truck steam motor and turbine locos and inserted extra information on the experimentals - see page 2 this thread , scroll down 80 % .

[Paul Milenkovic]

Juniatha

You know , at that time I was not all too much concerned about how such an assembly could be put together in actual construction practice and what it would mean to extract one unit for an overhaul in case of wear or breakage .   The effects of time did not loom too large in my thoughts back then - yet it was not without charme and I think it wasn't a bad idea in principle - although it clearly would have demanded more in-depth attention to details , mostly about manufacturing , design and shaping of parts and ease of mounting / dismantling .

You or some others could always build it in "zoo gauge" (12" narrow gauge or whatever is popular for amusement parks and zoos featuring steam railroad lines traversing the ground).

If some airplane hobbyists "from Texas" (folks with gobs of money to spend on eccentric and expensive hobbies seem to hail "from Texas") could perform aerial flyovers of Oshkosh, Wisconsin U.S.A. in a MiG 21 (before the fall of the Wall and before the ouster of Gorbachev, no less), there are people who would have the money, skills, and opportunity to build Juniatha's Triplex Dream Locomotive (Traum Lokomotiv?).

[Juniatha]

Hi Crandell

At this point I have to admit , in my 'walk on the wild side' type of private (pre-)engineering me too I had put up a concept of a Triplex , and just because of this very point - supplies vanishing & and tender adhesion mass alleviating - I had carried it just one step further , figuring if you can have two axles coupled in front of the main driver you could as well have the same on the back side of it - makes for a 10 coupled unit .  

You'd think I configured it to be a 2-8-8+10-2 ?  

Naw-naw-naw , sir - not with my way of thinking !   Ten coupled once introduced , I immediately thought "Why not have it on all the three engine units ?"   Mind my previous mentioning preference for identical sets throughout . So I ended up with a 2-10-10+10-2 - all simple expansion .   Sure enough there was that vanishing adhesion mass problem again .   

Ah , but not with young June :  "Let's fit individual throttles , one for each engine unit , three in a bunch to be grabbed together as the throttle levers in a BUFF yet individually adjustable to make the most of adhesion ."   The throttles were to have on spot hydraulic power operation so moving the levers didn't take but light effort as you effectively just opened the power valves of hydraulic lines .   The idea was , you would start opening the combined levers as one yet just to an extent sure to be supported by adhesion and then as the engine began to move you'd open up further on the second and first engine units , leaving the tender unit as set or just cautiously opening up depending on degree of supplies spent .   Since the locomotive was purely intended as a ramp pusher , tender capacity wasn't all that large anyways while no attempt at low empty tender mass was to be made .  

Arrangement of live and exhaust steam piping was a point I was particularly content with :  it featured a central live steam pipe with short ball type articulated sections with pipes branching out to cylinders ( it was a slightly more complex system than my words describe it here )   Exhaust lines from cylinders would combine in reverse to form a steam jacket around the central live steam pipe thus insulating it against the outside .  Each of the short articulated sections also was to contain the throttle to those very cylinders .  

You know , at that time I was not all too much concerned about how such an assembly could be put together in actual construction practice and what it would mean to extract one unit for an overhaul in case of wear or breakage .   The effects of time did not loom too large in my thoughts back then - yet it was not without charme and I think it wasn't a bad idea in principle - although it clearly would have demanded more in-depth attention to details , mostly about manufacturing , design and shaping of parts and ease of mounting / dismantling .

Maybe I'll put up that old side view picture of back then , if you'd care to see it ?

Regards

Juniatha

 

[Paul Milenkovic]

Answer me this.

The Garratt seems to be a good design for getting good steaming and a lot of powered wheels.  There were even plans (suggestions?  pipe dreams?)  for "double Garratts" -- essentially a Mallet's double sets of engines and drivers on each of the front and back ends fo the Garratt boiler.

A Garratt may have been able to pull this off -- the big-fat-short boiler with enormous firebox and ashpan suggests that the Garratt's steam-raising ability is without peer.

But . . . even on the "normal" Garratt, how did they get live steam and exhaust steam to and from the two bogies, without a lot of losses from the flexible connections, without thermal and pressure-drop losses in the long pipes involved?  Where did those steam pipes even go?  Even on a Mallet or simple-expansion articulated you can see those pipe runs to the front engine and driver set, but you don't even see any large-bore steam pipes on the pictures of Garratts I have seen.

[selector]

I can't say I'm not enjoying this.

Apart from the burner problems with the Triplex, I also have understood that as the tender-***-engine supplied water to the boiler, it tended to get lighter, and that affected adhesion.  So, while the burner couldn't really heat the water to provide the steam needed, when it was drawn down by the injectors, the water volume lightened in the tender/engine and made for a lightweight set of drivers.

Okay, so boosters are not ideal, and we have agreed we want unsprung and reciprocating masses minimized?  Yes?  Then we need a drive mechanism that rotates around an axle that is geared or fluid-coupled to the drive axles.  Or a jet engine.  Or electric.  Cylinders with large reciprocating and valved masses, and connecting rods, are out.

Could we have a burner array that runs along the belly of a boiler, thus doing without the problems of a crownsheet, stays, and so on?   We would still need a pretty stiff blower to help the draught up the vertical flues, and we would need a collector system for hot gases to get them to the flue exit, or stack, probably centered on the boiler.  The flue pipes would be 'starred' in cross section, with fins running along their outsides parallel to the main axis of the flues for maximum heat transfer.  Same for the collector.  This system, if it can be built and controlled effectively, would supply heat to the boiler at about twice the rate of a conventional firebox arrangement.  To help wick heat away from the bottom of the boiler, to minimize heat gradient stresses, the plating or inner liners of the boiler would need to be designed and built of the correct materials, and probably also have ridging or fins to increase heat transfer.

Some of you probably are thinking to yourselves that it's a good thing some people don't design steam locomotives.

Crandell

[Juniatha]

Hi Craig

The Heilmann locomotive - yes I had read about it in an early number of the LokMagazin ( ger for loco magazine )  of a collection lot of well over 100 numbers I had bought in an antiquarian bookshop in Berlin in 1992 .   I remember having to urge the storekeeper to put them aside for me since as a teenager of sixteen I sure didn't have the money right with me and he was so sceptical if I would ever come back and buy them as I said .  

Edited paragraph with extra information :

The loco concept put up by Jacques Heilmann was remarkably advanced with a frame supporting boiler , cross mounted two cylinder compound engine plus generator unit , cab and supplies , resting on two eight wheel power trucks , all axles driven by electric motors .   As with later diesel-electrics the locomotive showed a great potential in starting and hill climbing , the article described it as quite successful and two more powerful locomotives were built with an early 'Jules Verne' or 'Nautilus' type of streamlining while on the prototype the engine unit had been mounted pretty much on open deck with the front shaped in quite an earnest looking wooden wedge .  

The first test locomotive of 1892 had Lentz boiler of 180 psi , 1000 A / 400 V generator , 44 kW electric motors , all electrical equipment came from BBC ;  tests were run on the St.Germain-Ouest to St.Germain Grande Ceinture railway of the Ouest ( Western Railway Co.)

The two locomotives built for the same railway company by Cail & Co , Demain , in 1896 as # 8001 and 8002 intended for passenger service on the Paris - Trouville and Paris - Rouen mainlines were 60.7 feet between bumpers, had a boiler with a Belpaire firebox , nominally steaming 30000 lbs/h at 200 psi , feeding a logitudinally mounted vertically standing six cylinder Willans steam motor with external excentric shaft for driving piston valves and wooden shrouding for insulation ;  electric equipment again was provided by BBC , eight traction motors of 92 kW each were suspension mounted in truck frames and developed a total of 736 kW ( 991 hp ) , service speed 75 mph .

In the end the unconventional locomotives were deemed to expensive and operation was considered somewhat complex - probably much of it due to the greatly novel nature of the locomotives - and it appears railway officials didn't really know what to make of it .

In 1910 the North British  (NBL ) built a steam turbine-electric experimental , called the Reid-Ramsey for their originators ;  the loco was 60 feet between bumpers had an unusual wheel arrangement in two eight wheel trucks in which only the axles to one side of the truck were powered by electric traction motors making it a (2Bo) (2Bo) wheel arrangement - an arrangement that would be hard to define in Whyte classification .   The idler axles were however not arranged in guiding bogie to each power truck , all four of them were mounted in common truck frame .  

The general concept of the locomotive looked logically sound and included a condensor which was to run in front in order to provide enough air stream for cooling .   The experiment was brought to a stop with the outbreak of WW-I .   In 1924 the same builder went for another try with the Reid-McLeod steam turbine condensor locomotive , below a frame 66.9 ft between bumper apparently using the trucks left from the earlier experimental , although with electric equipment thrown out and mounting them in symetric positions , thus now resulting in a (2B) (B2) arrangement .  This time each bogie was powered by a steam turbine mounted directly to its frame , one having the HP unit and the other the LP unit - the arrangement of steam flow thus resembling that of a compound Mallet locomotive ( if that was a good idea I'd doubt seriously - yet nontheless it was a brave attempt )   external shape was quite harmonious and pleasing , even more car-like than the earlier R-R experimental , with cab mounted pretty much centrally and condensor enlarged and improved .   The Reid-McLeod didn't escape the ever-repeated aberrand mounting of forward and reverse turbine on one common shaft and thus suffered from ventilator action losses by the reverse turbine forced to idle inversedly .   While power output was up from 590 kW to 734 kW ( 794 to 988 hp ) the R-McL experimental didn't fare decidedly better than the earlier one and to me it looked a bit of a step back in its concept , with bogie mounted turbines by default being a feature of rather dim outlook :  being shock exposed and mounted in a quite inacessible position , demanding flexible steam pipes , etcetera .

Only much much later , in the mid 1990s , for afternoons on ends spent in enclosure to the bewilderment of her mother , one American in Berlin privately made another attempt at designing a double truck full adhesion Do-Do steam loco concept ( in quadruple truck BoBo-BoBo flexicoil suspension version , actually ) this time with a full treatment of everything of promising potential to improve thermal efficiency :  oil fired water tube boiler , high pressure / high superheat steam , turbine-electric drive plus condensor .  Well , I tell you I really dug into it , bugging dad about it until he raised arms admitting he was at his wits end , too , and why wouldn't I just see a movie or go to a concert with friends like any other girl .   What I can say from it :  a high pressure / high super heat water tube boiler steam turbine-electric condensor locomotive is a formidable excercise in engineering indeed and while it can be done , the sheer complexity and total mass of it will always stand against it .   In the end I threw out all steam equipment , replaced it with a gas turbine ( incl beefed up electrical equipment possible with mass saved by getting rid of boiler and condensor ) and - wow ! - there was the power I was looking for !

Regards

Juniatha

[cutting]

This might shock you – it did me! – but the French tested a steam-electric locomotive in 1897.  No, I haven't transposed the digits – that's 1897!!  Try googling "Heilmann locomotive" ...

Craig