Thermodynamics, DBHP and related technical matters

No, between Carlisle and Plumpton it was 2511 (1822 average), Plumpton - Penrith 2394 (2000 average), Penrith - Shap Summit 2311 (1560 average). Source: A.J. Powell Stanier Pacifics at Work (1986) Ian Allan Ltd, Shepperton ISBN 0 7110 1534 1. This book gives very detailed logs and graphs taken from dynamometer rolls.

 

110 metric ton LMS Coronation making 2282 hp compared to a 95 ton Std7 doing 2000 according to Cox is very close

 

The difference between DBHP and IHP also includes locomotive plus tender air resistance at high speed

 

Yes, I know what they are; they're basically where the point at which the power is measured occurs: i.h.p. = inside the cylinder so no losses;Rail Horse Power = at the wheel rim so takes account of the loco's INTERNAL friction losses; Drawbar Horsepower = at the drawbar so also takes into account the EXTERNAL friction (aerodynamic) and gradient losses, and Equivalent Drawbar Horsepower, which adds the estimated resistance of the loco in moving its own weight against a gradient due to gravity.

 

Excuse my stupidity ..... but not having plumbed the depths of fluid thermodynamics, can it be concluded we're looking at surface loss effects here?

Regarding "all other parameters", are we speaking of a separate set of (unrelated) losses along the steam supply circuit between regulator valve and steam chest and/or losses within the (slide, piston or poppet) valves? Good design - and thermal insulation - should obviously reduce losses in supply pipes to a minimum dictated by unavoidable surface losses (the rebuild of BVR No.9 highlighted this in a way even I could get my head around!), but I'd imagine each flavour of distribution valve has it's own inherent loss levels according to how much surface contact is dictated by physical characteristics.

.... which it did in the years before WWI. The lack of much happening on the Uniflow front in subsequent decades suggests theoretical savings didn't translate into anything on a scale sufficient to interest post grouping CME's. Would that be a fair observation?

I seem to remember seeing one kicking around on "Current and Proposed New Builds". My own penchant for the weird & wonderful, added to a weakness for narrow gauge, favours Bowman Malcolm's vonBorries system 2-4-2t's (which worked) ahead a Webb system compound (where - even accounting for George Whale's well known opinions - the jury is still out).

I suspect any recreated 12in/ft scale Webb System loco would need a lot of serious computer modelling before anyone in their right mind would consider cutting any metal.

 

Wouldn't there also be differences in the area of getting steam in and more importantly out of the cylinders? Or is the piston speed of a steam engine so low that its a minor issue compared to internal combustion?

 

Thanks for correcting me. I guess it must have been some misprint in the book I read.

Knut

 

an I/C engine takes a smallish charge into the cylinder which only becomes useful when it is ignited .you can increase the volume of air in several ways , such as a tuned exhaust which helps to drag in more air before the valves close . other methods of achieving the same objective are turbo & supercharging , and also cooling the intake air .overlapping valve timing can also help the ram effect but can be a pain at low revs - hence VVT which gives a tractable motor at low revs and a rorty one at higher revs . 2T motors have a pressurised crankcase together with a one way gas intake using reed valves or similar.
all these things help to make a bigger bang . head design is still important , because you need to shift gas very quickly

with a steam loco , the charge enters the cylinder intact and the work is done by what enters the cylinder .you don't have an explosion to do the work ,and you cant usefully play with compression ratios . inlet design is still very important , helping to achieve cylinder pressure close to the boiler pressure and avoiding unnecessary loss of pressure while filling the cylinder.

so while I/C and steam engines are both heat engines they require very different solutions to work efficiently

 

I will propose an IOM 2-4-0t or DHR 0-4-0 as tester case.

Space enough for strange cylinders,not to many rev per second and a lot of going sisters to compare with.
Big enough to make extrapolations into mainline locomotives thrustworthy.
Last to run probably.

The Stumpf uniflow system works very good in condensing steam engines but in locomotives exhaust closing /compression beginning comes to early.

A prof Hall did some scientific tests of superheating in model locomotive size cylinders and his theory was that benefit of superheating was mainly that saturated steam condensed on entry and was later evaporated again before exhaust.This is bad and superheated steam does not.It can therefore be argued that if cylinders are kept on boiler temperature at all times no condensation but a somewhat bigger thermal loss from outside of cylinders.If space is no theme some glasswool or vacuum etc can be applied.
It will also answer the Question :was Compounding and superheating dead ends?

 

One thing that has puzzled me is whether a 4 cylinder compound would have worked better with the 45 degree LN crank arrangement, rather than the 180 degree pattern. If I've worked this out correctly, and if you assume early release from the HP exhaust valves, they would coincide better with the LP inlet valves being open.

Obviously balance wouldn't be as good, and the valvegear would have been more complicated, but did anyone ever try?

 

The only comment I feel qualified to make regarding uniflow systems is that AFAIK, no installation trialled by any railway resulted in the development of any class of loco being built on the principle.

Superheating, on the other hand, produced enough by way of operational savings to interest just about every mainline company. I'm not familiar with Prof. Hall or his work, so don't have sufficient knowledge to comment one way or the other, but the manner in which events actually unfolded suggests experience trumped theory on superheating. The (doubtless simplistic) view is that it was the remarkable performances of Marsh's I3 'atlantic tank' locos on through workings from the LBSCR was what finally convinced the LNWR to adopt superheating for express passenger classes and that that just about every noteworthy CME from Churchward onwards seems to have been convinced by the merits of superheating, so I'd have to call it a 'case closed'.

Compounding always seems less cut and dried. It's a matter of record that only two (edit .... three .... as @MellishR kindly pointed out) UK compound loco classes served beyond WWII and these were of not only different gauges, but in the case of the NG S/S1/S2 classes, different compounding systems too. Although there's a widely held view that steam can be expanded just as efficiently by well designed valve events as by compounding, the UK's notoriously tight loading gauge seems to have been an equally significant limiting factor. I've always considered conditions in marine applications are too different to draw any meaningful comparisons, but note that condensing seems a damned sight easier to arrange when space isn't at such a premium. The only example of a successful railway application I can think of offhand were the SAR class 25 locos, where the lack of watering points across the Karoo made condensing (and the associated maintenance costs) worthwhile. Even then, it took a lot of effort to make the system work reliably and resulted in condensing tenders of comparable length to the locos themselves ..... and heavy with it.

btw .... Unless you're the sort of person who enjoys the feel of a sharpened pitchfork where no pitchfork should ever go, I'd think very carefully indeed before suggesting doing anything to the IMR's beloved Beyer Peacocks - even in theory!

 

I assume you mean the LMS and GNRI 4-4-0 compounds. Those were both 3-cylinder classes. What was the difference in the compounding systems?

 

The link to Bill Hall:

https://www.advanced-steam.org/ufaqs/prof-bill-hall-papers-and-locomotive-simulation-software/

Much of the efficiency gain due to superheat was the neccesaryty of using better piston valves.
Railways used untreated water and superheated locomotives used sometimes 20% less water and this cuts boiler maintenance heavyly.
The coal saving was not so high.
Thank You for advice concerning IOM.I find the DHR locomotives much better engineered anyway.

 

OK .... confession time .... somehow, I clean forgot about the GNRI V Class (built to pretty much the same arrangement as the Derby compounds) when I wrote that - sorry. T'others I was thinking of were the vonBorries 2-cyl NG locos, which outlasted all 5'-3" gauge vonB compounds by decades .... and were a very different kettle of fish.

 

hmmmm
I think the Lanky compound 4-6-0 was around until just after the war . it was a one-off which was a shame as it seems to have been very good .

I don't think the loading gauge argument holds much water . the GWR imports seemed to fit ok , and the Stanier compound Castle that Collett objected to must have fitted .

the NER 3 cyl. loco(were there only one or two?) were very good and would have been multiplied if Smiths executors had been more reasonable in their demands .

without saying too much about Webbs locos' he came very close to producing something exceptional. the Ionics did some marvellous work and were reliable .

The 5 Smith/Johnson locos were also excellent . what a shame the Midland felt the need to simplify them for common usage . they were far better than the standards . they could have been better still if Johnson had not contrived to reduce the cylinder pressure .

the point of the foregoing is that apart from Webb nobody made a really determined attempt to make the system work . I am sure we would have had several exceptional classes had they done so .
how good the Fowler compound pacifics would have been I cant say , but if the 2-6-4 tank is anything to go by they could have given the LNER pacifics summat to chew on .

how good would it be for a group to build an oil burning compound Duchess with all gen that we now have . no doubt mr Kloopmans would like to have a go at the draughting .

 

I don't know. AIUI the De Glehns and the compound Castle proposal both had the high pressure cylinders outside, which meant that the low pressure cylinders were limited in size by the amount of room between the frames.

 

Perhaps not on sections of the GW open to 'Kings'... but their range was dictated by size as much as axle load. Didn't Stanier comment on compounding and loading gauge considerations with reference to his 4-cyl LMS pacifics? On the LNER, overall width considerations seem to have played a significant part in development of Gresley's 'Hush-Hush' 4-6-4 (4-6-2-2 .... or whatever it was!), with it's (inside) HP cylinders being lined to reduce volume and post nationalisation, the restricted British loading gauge certainly led to Riddles' class 8 having three (simple expansion) cylinders in lieu of his preferred layout.

 

This thread seems as good a place as any to link to a video just posted yesterday about the return to operation of the Red Devil:

 

just WOW!