P2 Locomotive Company and related matters

Hopefully someone with the knowledge to answer will chip in.

 

There is a lot in there about horse power, but with a locomotive you need to specify where it was produced, in the cylinders, at the wheels or at the drawbar. The first will be highest with a drop to rail h.p., and a further drop to d.b.h.p. If it was measured at the drawbar, it's probably in line with the Pacifics, but if in the cylinders it would be a bit weak.

I don't think you're right in saying that an A3 could not produce 2000 h.p. In normal service and just to maintain the timetables, the Stanier Pacifics were putting out 2100 d.b.h.p., and I don't think that an A3 couldn't make similar, even if only occasionally and for fairly short distances.

 

Indeed so (and I'm sure you realise my post wasn't to be read seriously), but taking the 9F in particular, a couple of things jump straight out. The most obvious wasn't an issue during steam days, but would the flangeless centre drivers, employed by Riddles to mitigate the long fixed wheelbase, actually be suited to higher speed running (even had the ban on flangeless wheels not occured)? Then comes the wholesale redesign of bearing surfaces (which certainly wouldn't increase route availability), the reason BR concluded their unsuitability for sustained high speed running.

If we accept Mr Riddles knew his business, we need to recall the 9Fs were designed for brute strength, over outright speed,"Fast freight" being a decidedly relative term. As 92203 proved in starting a record 2,178 ton train, back in 1982, the design evidently isn't lacking in the "sheer grunt" department and as was demonstrated by HNG's P1 class, a generation earlier, being able to shift humongous trains doesn't by itself make actually operating them any more practical a proposition.

As a 'what if', I'd be more inclined to wonder whether Bulleid wasn't on to something with the total adhesion approach. Even Mr.Kevan Ayling's 5in gauge edition of "that" loco, encumbered by neither offset boiler nor sleeve valves*, gives a fine impression of the step-change in performance over traditional layouts. That little beast can shift numbers on sit-astride carriages you'd more normally associate with a substantial 71/4in gauge loco. On one particularly memorable occasion it also demonstrated quite clearly the limitations of ME braking systems never designed with anything like those sort of trailing loads in mind!

I was surprised to see mention of Mr Webb's layout mentioned a few days ago, but do wonder about the traditional UK disdain for compounding. Both the Smith/Deeley 4-4-0s and Bowman Malcolm's (3'0" gauge 2cyl VonBorries) 2-4-2Ts operated successfully into the 1950s. I'd love to understand the finer points of the steam requirements of compounds v simple expansion. Even accepting the improved economy gained with well designed long travel piston valves matched or exceeded the performance of compounds of that era, I've never really understood why development here came to an abrupt end.

*I was fortunate to see this incredible loco under construction. Kevan Ayling sensibly opted to rework the undergubbins with 2 Walschaert's activated piston valved cylinders per bogie, coincindentally derived, much altered, from a 5in gauge 9F design. Not a million miles removed from the approach adopted by OVSB in Ireland.

 

He patented conjugated gear that only could benefit three cylinder simples with drive to second set of driver.
He patented a double swing link pony trukc

 

A 9f high speed version with 3 cylinders would have had lateral space for Krauss Helmholtz/Zara truck up front and then 120 kmh was scheduled in france for Chapelon 141P.
Nearly 20000 German ,Polish ,Check and russian 2-10-0s had it as well.

 

But that is not the same as a) patenting all three cylinder drives nor b) all pony trucks.

 

Bulleid could have made a much better job and cheaper as well by doing a Hymek like Heisler variant.
A boiler instead of Diesel engine and a V2 compound engine instead of torque converter reduction gearbox.

 

But the track on railways does give. Its not totally fixed and does deflect somewhat given the forces it encounters.
I hope we are also remembering that axles do have axial float within the frames/boxes. Its not 8 drives couples rigid in one fixed plane. There are degrees of freedom.

Except that it isn't and I did not say that. Nor would I. A P2 isn't articulated for a start. It will apply different forces, but if the argument being pushed is that 8 coupled locos spread the track....which incidentally is not my argument. I think that one is a red herring.

Primarily because there is a civil engineering department who have a large say in curvatures that loco designers must adhere to.


Its almost as if some want to pin the failures on extraneous circumstances or secondary contributions whilst ignoring the primary cause as identified at the time of the failure. Its basically the Tornado gibberish all over again. Ignoring primary causes in favor of spurious and vacuous arguments devoid of evidence or reason.
Was it not also the case that the keyway as manufactured on the original P2 axles was too deep/not conformant to drawing?

 

I fear that Hermod has yet to realise just how restricted our UK load Gauge is that gives little scope for incorporating a Krauss-Helmholtz or Zara truck.

 

There are a pair of NG15's at Dinas (F&WHR) which use the system, now re-wheeled No.134 is well into restoration, No.133's hasn't yet started. AFAIK, this class are the only locos in the UK with this feature. There are some good photos of the bogie on the restoration website* (https://www.ng15-134.co.uk/). AIUI, the usual reason given for the K-H system being seen in the UK, aside from a marked lack of need for many locos of that size, was the restricted loading gauge, specifically clearances over outside cylinders, though I suspect an element of 'not invented here' might exist too.

* for latest news, always check for recent updates, link at top right, as 'working party notes' is more often up to date than 'latest news'!

 

We know quite well how much well made two step compounds improve steam consumption in Post Chapelon machinery.
He tried,but not desperate,to even work between high pressure and low pressure cylinders in four cylinder compounds.
In three cylinder compounds be they Webb or Sauvage it is a litle more tricky as we want as even a torque as possible.
I have not seen it mentioned anywhere what is ultimately most efficient as nobody ever cared but the ideal receiver steam conditions are not the same as in a four cylinder compounds.
TheWebb system was the thing for UK loading gauge and it is my feeling that it is more frugal ,but very little ,than the LMS system.

 

I have realised that two twenty inch outside cylinders 6 feet 8 apart will not allow plus minus 1 inch lateral movement of first driver ,but three 17 inch is OK for the full KH

 

That's a huge stretch. If you could justify to me not rebuilding or scrapping locomotives that by 1943:

• Had annual availability of less than 50%
• Had annual mileages less than 30,000
• Had suffered crank axle failures - serious enough to see several of the class out of action at the time
• Had suffered recurrent hot axleboxes and other related frame issues
• Had suffered recurrent middle big end failures

...then I'd love to hear it, because if you took away the names Gresley and Thompson, and asked an audience of railway enthusiasts whether "locomotive Z" should be kept in service as is, rebuilt, or scrapped based on the above, I think you'll find the latter suddenly becomes more palatable to those not emotionally involved in the history of the thing.

An objective viewpoint would recognise the Gresley P2s as poor locomotives, pre and during the second world war, bettered by other Gresley designs (A4s, V2s) and then rebuilt into something more reliable (which we can easily evidence - the Thompson A2/2), and then outright replaced by locomotives that were by far superior (the Peppercorn A2s).

During the war and post war the work the P2s did and would have done was covered by Gresley, Thompson and Peppercorn Pacifics. If anything, the failure of the P2s in terms of mechanical reliability vindicated whoever's decision it was to switch to Pacifics on that route.

I am not convinced that this is true. I think this is a romanticised viewpoint that does not have a basis in fact.

Gresley had moved onto other projects after the building of no.2006: the P2s were six non standard locomotives running on a route that whilst important, was overshadowed by the other LNER projects going on, like the Silver Jubilee, the Coronation, an electrification project and even the V4s later on. Gresley clearly wasn't interested: the issues of the P2s were not new by 1941 and Gresley's death, and by 1942/43 when Thane of Fife was rebuilt, their availability and mileages were significantly lower than other Gresley designs doing better mileage and with higher availability too.

Arguably Thompson's decision to rebuild them is vindicated by the evidence we have available to us: and anyone who seriously tries to argue that he rebuilt them on grounds other than their poor availability and mileages is not being objective at all.

 

The thing that Thomson did wrong from a politico-enthusiast point of view was to ' rebuild ' them. if one had been plinthed ( a ridiculous and unaffordable luxury at the time) and sent the others to scrap ( ie component recycling) and then produced his own design with different names there would not be half of this 'gresley trasher' nonsense. The fact that for reasons of accountancy /cost/ other war time restrictions it was cheaper to pose this work as a 'rebuild' than a new ' mixed traffic' class gets us where we are now ...

 

I think there is something in what you say; however, just to pick up specifically on your last sentence:

"The fact that for reasons of accountancy /cost/ other war time restrictions it was cheaper to pose this work as a 'rebuild' than a new ' mixed traffic' class gets us where we are now" (my emphasis)

In the case of the P2, I don't think there was any "posing" about a rebuild going on: they really were a rebuild, with a significant amount of the originals ending up in the new locomotives. Maybe that accounts for the slightly ungainly appearance round the front, but they were no accountancy job such as early King Arthurs nominally rebuilt from Drummond 4-6-0s, but containing precious little of the original beyond the number plate. The Thompson rebuilds were genuinely a rebuild, and that is apparent not only dimensionally, but also in the price - where all six were done for less cost than one (comparable in size and era) brand new Merchant Navy pacific. Had they just been posing as rebuilds but were really entirely new locos, they would have been of the order of £15 - £20k each, rather than about £2,400.

Tom

 

It wasn't "posed" though. Most of the locomotives were retained completely unchanged, including two sets of the three sets of valve gear which were simply mounted further back onto the new cylinders and crank axles. Getting a new Pacific locomotive for £2400 was an absolute steal in 1943, that much is certain - getting one that was as good as Thane of Fife was?

Thompson and his department absolutely nailed the brief of retaining most of the original parts and materials, and simplifying the design to such an extent that building a new Pacific was possible.

I am in danger of turning this into an off-shoot of the Thompson thread, so I'll leave it at that - if you want to do the P2 versus A2/2 debate, it's best done on the Thompson thread now frankly.

 

German 2-10-0s were, by a wide margin, the world's most numerous types of 3-cylinder engines.

https://en.wikipedia.org/wiki/DRG_Class_44
https://en.wikipedia.org/wiki/Prussian_G_12

If British Rail had built a Class 44 equivalent, it would have been a 10F. But it would have been a tight fit to get a P2 (or A1 or MN or Duchess) boiler onto a 2-10-0 chassis within the British loading gauge.

 

The P2s were built because the LNER believed that the traffic on the Aberdeen road needed more power and adhesion weight than was available from any existing locos. It is somewhat surprising that the first one, after rebuilding as a Pacific, with the same power but less adhesion weight, turned out to be adequate for the traffic after all. If the slightly smaller A4s were adequate as well, that does suggest that the original decision was a mistake; whether because the train loads weren't as heavy as expected or for any other reason.

 

I'm not sure accountancy really came into it. I've studied this for the GWR, but I doubt the LNER was much different - there was, AIUI, legislation controlling this. There were basically three pools of money available for locomotive building. Capital, Renewals and Revenue. Capital was for expansion of the fleet. The GWR stopped expanding their fleet in the late 1920s, I doubt the cash strapped LNER was much different. Its probably a fair assumption that Thompson wouldn't have been allowed to build the new locomotives and keep the P2s! Revenue was of course money taken from income, and the main use in this context was keeping the fleet running. Post the great depression the GWR revenue wasn't great, and again I doubt the LNER was any better. That leaves the renewals fund. This was money that was required to be put aside for the specific purpose of renewing - literally making new again - life expired locomotives. Renewal covered a very broad range of options on the GWR. Scrapping 3 Dean Goods and building 2 Halls was done under renewal. So was (surprisingly perhaps) sending an absorbed Welsh 0-6-2 off to Stephensons and instructing them to make it as good as new whilst reusing any bits that were good enough to be reused. Upgrading a Star (or the Great Bear) into a Castle was done both from revenue in the 20s and renewal budgets in the 1930s. And, more obviously, scrapping a pre group pannier tank and replacing with a new 57xx was renewal. The precise nature of the work being done really doesn't seem to have come into it, except that they were not allowed to pretend that a major mid life heavy general overhaul with new cylinders etc was renewal, but they were allowed to take a locomotive requiring the same work and upgrade it with a larger boiler or smaller wheels at the same time, and call that renewal.

So what of the P2s-> Pacifics? I bet this was done on the renewal fund, 'using such parts as may be suitable' and that building new locomotives, not replacements, on capital simply wouldn't have been an alternative. As @S.A.C. Martin points out, there were a lot of secondhand parts used, and they got new locomotives at a bargain price. The only alternative as I see it (but I could easily be wrong, I am not professionally trained in finance) to remedy these locomotives would have been redesign of major components to resolve their problems and replacing the components out of revenue. , but this assumes that the LNER drawing office was going to be capable of a major redesign that would be guaranteed to resolve the flaws, and they weren't intrinsic. I submit it needs to be borne in mind that a great deal of work had already been done on their flaws, and self evidently it hadn't been successful. The Pacific rebuild feels like the safer option. The LNER knew they could build satisfactory pacifics.

 

Or else, in the middle of a war that had seen a complete turning upside down of the way the railway system was used, the traffic had changed. There must be a fairly narrow window between traffic too heavy for a pacific but within reach of a 2-8-2, and traffic so heavy that you need to run more trains. It might just be that the goalposts had moved.