P2 Locomotive Company and related matters

Mods please move if thread drift is too excessive. I have been following this fascinating conversation (thats why I love NP) and I have had a thought, yea ok! ok! settle down.
It goes like this, does there exist a formula for working out the overall efficiency of a loco, I know there is a lot to be taken into consideration here and I am technically challenged shall we say. But I reckon, ratio-wise, a Terrier @ about 30t all up, hauling 2 Mk1s would have more in hand than a Duchess 177t (?) all up, hauling 12 Mk1s.
Do I need a parachute ?

 

Not remotely meaningfully IMHO, too many variables and too much that can't be measured.

 

I respectfully disagree. It is not one formula, but the outcome of a series of formulae. After all, steam locomotive design was serious business and taught at universities. Been there, done that!
Kind regards
Jos Koopmans

 

Isn't that a different thing though?

I'll cheerfully believe you if you tell me the science exists to do a substantial prediction of the performance of a locomotive if a full evaluation of the design is done which would be a substantial body of work. It would presumably need to fully evaluate valve gear, draughting, steam passages and much else. But isn't that days or weeks work per design?

What I have trouble believing is that you can take a bunch of headline dimensions as published in an enthusiasts book, type them into a simple spreadsheet, and produce something useful that will accurately predict the real performance of very different classes and their subtle design flaws.

After all the presence of 3 15*26 cylinders doesn't tell you much unless you know how capable the rest of the system is at getting steam into them, controlling how it expands and getting the steam out again.

 

Which is why they developed Indicator Testing. You can design many, many things to do a certain function in a certain way, but ultimately you still have to try out the finished product to check that it conforms to all the theories. Performance logs, dynamometer car readings, indicator test results and, ultimately, rolling road tests at Swindon or Rugby are all there to prove the design.

Today, only the first of these - the least accurate - is available, but I suggest that in these days of electronics, Bluetooth, etc., it might be possible to indicate an engine from the support coach? I can't imagine the Health & Safety Executive allowing engineers to ride on the front running plate, as in days of yore.

 

Modern marine diesels use strain gauges or Piezo crystal pickups coupled with a flywheel pickup of either optical or proximity type to get indicator diagrams - no more doing drawcards by hand - and it's very reliable.

 

Agreed - you can have all the formulae in the world and make top trumps comparisons between cylinder size, boiler pressure, heating surface and all the rest of it, but in days past it was still the fact that some locomotives were superb and some were duds. Sometimes you can look at things after the event and understand "that will never work", but clearly doing so before construction is much harder: I don't believe any designer deliberately set out to design a poor engine, but plenty were built, including by people ostensibly in the highest regard as designers **cough** Drummond **cough**.

The P2 is indeed a case in point. On paper, the numbers look like a world beater, but the history shows they weren't out and out successes. With the computing power available today and modern analytical models, it is possible that the faults of the originals might be overcome to unleash the full potential of the design - I suspect many of us hope so. But those tools certainly weren't available in the 1930s, even if the basic theory existed.

Tom

 

Help! I am not getting any text in the pink 'quote' boxes, just the box and pictures and highlighted/coloured words, if included
Edit, now everyone's name has disappeared from their avatar box, all the other info is there how bizarre

 

They are machines. Folklore and past performance aside, there is nothing about these machines that mathematics and engineering cannot very accurately represent.

However, given the complex nature of these machines, very small flaws or errors in design could render what on paper looks to be a decent engine, pretty useless. Particular variables can have huge knock on effects to the overall design. Nailing down which ones in any particular design would indeed require significant time and resource.

The analytical methods available today are worlds away from what Gresley/Stanier and co had at their disposal. Computational Fluid Dynamics and Finite Element Analysis being two cases in point that would have given the designers huge scope to iron out any shortcomings.

If the P2's were such a terrible design, why is it that the spent the majority of their life on the line for which they were built...?

 

To be fair, I can't recall anyone saying they were a terrible design, only that, like all things, they had their problems.

 

They were north of Edinburgh so the LNER probably forgot about them. The small design/construction errors that can ruin a sound design were what hampered Duke of Gloucester and we may well find that it is the same story with the new Clan.

 

Ihp allows for some interesting comparisons without necessarily thrashing locos. Holcroft graphed the mep figures for Schools at 25% cut off and a range of speeds. From that he concluded that indicated tractive effort had an inverse linear relation to speed, and that peak indicated power at 25% would be at 75mph.

He then went on to note that peak Ihp for the King Arthurs was very similar, and yet the Schools were renowned for being able to perform feats beyond the Arthurs in the right circumstances. He attributed this to the lower weight of the Schools (they wouldn't be adhesion limited at 75mph) and the lower rolling resistance on curves of the 4-4-0 chassis. This may have some relevance to the P2/Duchess question.

I still suspect that on a wet day west of Salisbury crews would have been glad of the Arthurs extra adhesion.

 

Design engineers stand on the shoulders of past generations who gathered data which the clever ones converted into formulae. One should realise that there are hundreds of years without computers in which former generations made their calculations by hand, used a slide rule or a primitive mechanical calculator and logarithm tables for difficult roots and powers. So a proper design proces took week/months of careful consideration and calculation. As an example, I recently made a weight calculation of the King class locomotive based on german formulae of 1948. My weight was 789 kilo less than the 90 tons but needed a 1440 kilo addition due to the difference between imperial and metric tons so it was only 2.5% off in the end. That is how these things worked in the past.
As for the simple spreadsheet, anyone just taking the trouble to dig into the steam locomotive design process, there exist books you know,
is able to (re)calculate as much locomotives as he likes, the foundation for this is laid in the past.
Kind regards
Jos Koopmans

 

Correcting the thread swerve with a dose of opposite lock..... splashers!



Foxy

 

With all the will in the world, I have to challenge this. The P2s weren't a terrible design (clearly capable machines with the potential for very high performance in terms of sheer power and length of train against timekeeping) but it is clear they were flawed mechanically in several very specific areas and Gresley had no interest in developing them further after the first two became standard with the latter four.

The overheating axle boxes, the several instances of broken crank axles and the reputed track spreading all forgivable in isolation, but when put altogether, and coupled with poor maintenance regime at Cowlairs and poor utilisation (resulting in heavy coal and water consumption by comparison with the Pacific classes) you can see that anyone looking from the outside in, would wonder what was going on.

Thompson had the report from ES Cox which pointed the finger at the P2s directly. If you think Thompson didn't have any ammunition in order to rebuild the P2s, I suggest you go to the Thompson thread where I posted a copy of the ES Cox report I typed up last year. Bear in mind that ES Cox on his later books then described Thompson as conducting a Machiavellian campaign against all things Gresley, and you have to wonder if that was entirely fair given the language and tone of ES Cox's report and how much he took aim at the P2s, and consequently the conjugated valve gear and middle big ends' failings.

All of these issues, incidentally, the P2 Trust has worked on and the finished 2007 is likely to prove that the class - had it been developed in the way it could have been, were it not for the war and circumstance - had the potential to do some incredible work for the LNER.

 

I've just looked at the Cox/stanier report again. It makes only passing reference to the P2s themselves, when referring to big end failures. So your statement that it points the finger at the P2s directly is something of an exaggeration.

 

I'm not sure how this isn't pointing the finger:

But then I'm looking at this from the point of view of trying to remain objective about the report.

 

It's a matter of emphasis. The report was about the conjugated valve gear, as you well know, and the P2s were referred to in passing because they had that gear and were suffering the same problems as the pacifics and V2s.

I'm truly sorry, but I cannot accept that you are objective about that report - and if you simply want to make snarky remarks, don't bother

 

The report does point out other flaws, including but not limited to the middle big end.

That's up to you in terms of my objectiveness and you are free to feel whatever you like about me. However the 2-8-2 classes are mentioned directly: it's there in black and white. I'm afraid I didn't write the report, ES Cox did...

 

Yes they are, but the complaint was about the gear and bearings and hence the pacifics, V2s and P2s which had them, not "and how much he took aim at the P2s, and consequently the conjugated valve gear and middle big ends' failings" to quote you above. As I said, it's a matter of emphasis. The report was not about the P2s

For those who actually want to reread the report the link is
https://www.national-preservation.c...lysis-2012-present.35938/page-33#post-1081993