275+ PSI or higher?

[thx712517]

Not sure if this belongs in questions or discussions. I'm curious as to why British locomotives (Grouping to BR) seemed to top out at 250 PSI, while American steam of the same time used 275-300 PSI. Playing with a tractive effort calculator it seems that boosting pressure results in a significant increase in tractive effort. 

 

Was it the increase in maintenance? Or more expensive material for the boiler? I was thinking about the Castle/King locomotives, where a Castle at 275 or 300 PSI would seem to outperform a King and reduce the need for different locomotive classes. 

[Titan]

Could be the use of copper fireboxes, were the Americans using steel?

 

Copper is more conductive of heat, which would make them more efficient, but is softer and can have oxidation/corrosion issues.

 

As for power, that is more to do with heating surface. Sure higher pressure might give a higher T.E., but all that might mean is it can slip it's wheels more easily rather than pulling a longer train. At speed it is more about steam production. If the engine is running flat out at 240 psi on a 250 psi boiler, then increasing the pressure alone without increasing the heating surface just means it will run flat out at 240 psi on a 300 psi boiler. 

Edited July 31, 2019 by Titan

[RFS]

Bulleid's pacifics started life with 280 PSI but after a while that was reduced to 250 in order to try and reduce their susceptibility to slipping. 

Edited July 31, 2019 by RFS

[peach james]

Copper boxes.  

 

The advantage of going to higher pressures is mostly shown up in using smaller cylinders, which then can fit outside rather than between the frames.  You can get about 20" between the frames, and up to about 23" outside the frames in the UK loading gauge.  Then you can translate that to wheel diameter, and figure out how many wheels one needs to get the starting Tractive Effort to the rail.  Horsepower isn't what you are worried about, it's the 1/4th of the weight on the drivers for the TE to be able to start the train that matters.

 

By going to a higher pressure (for example, 300 PSI vs 250), it means you can use 2 outside cylinders, and get to the maximum TE that the weight will allow on 4 axles while staying inside the loading gauge.  (this is the classic Lima SuperPower style of loco for North America...)

 

The maximum pressure (and Temperature) are important for Horsepower production, because the laws of thermo demand it- the higher temp difference between inlet and exhaust, the more efficient the engine will be.  So going to 300 PSI and 800f superheat, you can gain a fair amount over being at 250 PSI and even the same S/H.  Steam Density changes as well, which affects flow losses through passages.

 

Note that in order to get enough cylinder area for high HP locos in the UK loading gauge, a 3rd cylinder between the frames is required.  (See DoG as the example, vs the Britannia's)

 

The topic can get deep quickly, I'd suggest reading "The Red Devil" as it has most of the series of choices that have to be made.  Exceeding 250 PSI with a copper box, or 300 PSI with a steel box on a loco type boiler is hard due to the amount of flex required in the boiler and the ensuing difficulty of keeping leakage and broken stays to a minimum.  

 

LMS Fury, and CP 9000 are both examples of attempts to bypass the conventional limits of steam loco's.  The Sentinel-Doble engines made for Columbia are another example, as was the LMS Sentinel-Doble shunting loco.  (or the ultimate last gasp attempt, in the form of Jawn Henry built for the N&W in the US)  Note that none of them were repeat ordered...

 

(the definitive book on Jawn Henry is "Rails Remembered Vol 4", I am not aware of any similar railfan books for the  Schmidt type HP locos.  The Sentinel locos are briefly mentioned in "The Sentinel Vol 2", but again details are lacking.

 

James Powell (Cert 3A, PRO class/3rd Class Power Engineer)

[Ken.W]

Another possible factor, many of the larger UK designs were constrained by axle loading restrictions

A higher pressure boiler would require much thicker boiler plates, and so heavier boiler.

 

Some routes were an even greater problem, the LNER had great difficulties with getting the B17 design to meet the GE section loading restrictions for example

[62613]

The other problem might be the axle loading on the driving wheels. You will be aware that, if you increase the boiler pressure, you would have to increase the thickness of the boiler shell plates, for the same boiler diameter;  the boiler diameter used on most if not all, of the most powerful locomotives was over 5 feet, which would almost certainly mean the next plate thickness up, if you increased the pressure by as much as 50p.s.i. The track used on most main lines in the UK seems to have meant a self-imposed limit of 22 tons axle load; and several classes reached that limit.

 

Besides British engineers were a pretty conservative lot when it came to new ideas; look how much time it took for pressures, boiler diameters and so on, to reach the levels they did between the wars.

 

Lastly, was there any need for locomotives that powerful in the UK? This topic has been discussed to death in the Imaginary Locomotives thread.

I've just seen that KenW has been saying more or less the same thing while I was typing!

 

Edited July 31, 2019 by 62613

[thx712517]

I loved going to the 5AT website and reading all the mini articles on various locomotive systems and potential improvements. 

 

It's not worth a new thread, but why would Collett/Hawksworth/Stanier/Riddles/etc reject roller bearings? I think Riddles did a few Standards with roller bearings and didn't find them worth it, while they were adopted more widely in the US. 

[jim.snowdon]

4 hours ago, 62613 said:

The other problem might be the axle loading on the driving wheels. You will be aware that, if you increase the boiler pressure, you would have to increase the thickness of the boiler shell plates, for the same boiler diameter;  the boiler diameter used on most if not all, of the most powerful locomotives was over 5 feet, which would almost certainly mean the next plate thickness up, if you increased the pressure by as much as 50p.s.i. The track used on most main lines in the UK seems to have meant a self-imposed limit of 22 tons axle load; and several classes reached that limit.

Or, make use of higher tensile strength steels, as was done in a number of instances, both in the UK and elsewhere, although in some cases, not without problems. I can recall one of the last classes of US steam locomotives, I think, the NYC 4-8-4s, suffering problems with cracking in the boiler plates.

 

Jim

[jim.snowdon]

4 hours ago, thx712517 said:

I loved going to the 5AT website and reading all the mini articles on various locomotive systems and potential improvements. 

 

It's not worth a new thread, but why would Collett/Hawksworth/Stanier/Riddles/etc reject roller bearings? I think Riddles did a few Standards with roller bearings and didn't find them worth it, while they were adopted more widely in the US. 

The case for their use in British locomotives would seem to have been marginal, but for many of the later US locomotives, the axle loads and piston thrusts were a lot higher.

 

Jim

[Pandora]

I think I read Bulleid was thinking of 350 psi, and compromised maximum boiler pressure because of maintenance costs of the higher pressure,  google LMSR  6399 Fury, the experimental LMSR 4-6-0 with three steam circuits, 1400 - 1800 psi,  900 psi for the cylinders and a 250 psi circuit 

Edited July 31, 2019 by Pandora

[thx712517]

Fury puzzles me to no end. They go through all of that complexity and trouble when I would assume a 275-300 PSI conventional boiler would have been far less complex. It was within the state of the art at the time, and I would also assume the weight of a 275 PSI boiler would have been similar or less than Fury's crazy system. Same with LNER's Hush-Hush 10000. Why go to the effort of a Yarrow water tube boiler when you could just crank out a stronger A4 boiler and squeeze it to 300 PSI? 

 

But it's remarkably easy to armchair decisions made a century ago, especially when viewing it through the narrow goal of improving mechanical performance. US railroads certainly accepted the maintenance of higher pressure boilers but it seems their British counterparts placed a higher emphasis on maintenance cost than Americans did. 

 

It's like wondering why Churchward didn't make use of much higher superheat temperatures. I could say a modified Castle operating at 275-300 PSI with high superheat (and roller bearings throughout) would negate the need for a King and its associated non-standard components, but at the end of the day the past is the past and cannot be changed. 

[Pandora]

The Castle was the benchmark for efficiency , witness the LMS claim  for their fresh new  4-6-0 runs with the dynamomter car showing figures which  matched (note did not exceed) the thermodynamic efficiency of the Castle.  Gresley was sceptical and after a suitable wait  requested the loan of the LMS dynamometer car, LNER  back to back trials of the LNER and LMS cars showed serious design flaws in the mechanism of LMS dynamometer car drawbar giving exaggeration of the actual figures to the embarrassment of the LMS. The GWR rested on past glories and after 1948, Doncaster Drawing Office staff  headed off to Swindon to deliver a few lessons

Edited August 1, 2019 by Pandora

[The Johnster]

I think the usual answer to the question of why British railways avoided higher pressures is fairly simple; they did not want to take on the increased maintenance cost and the down time that it represented.  Put another way, suppose you wanted to replace 31 Kings with higher pressure Castles, as suggested above; you'd need to build perhaps 35 or 36 new locos to cover the work as you'd have to reckon on a higher percentage of your fleet not being available for traffic at any given time, but still have to run the timetable.  

 

After that, the problems start to mount up and the advantages of higher pressure are further eroded.  Sticking with the above example, your HP Castles would probably weigh as much as the Kings and be just as route restricted because of the heavier boilers; the only advantage you've actually achieved is 2½ inches increase in driving wheel diameter, which might give you a few mph at the top end of the loco's capacity on the flat, but is actually just dead weight when you're trying to lift a train over the South Devon banks.  Hawksworth's Counties were originally pressed to 280psi like Bulleid's pacifics, and dropped to 250 later for the same reasons, which meant that all 3 classes carried heavier boilers than they needed to...

 

As has been said, British loco practice was very 'small c' conservative and there was a palpable reluctance to indulge in any change.  Even the Kylechap exhaust, a proven record breaker, was not widely adopted, though it is fair to say that draughting in general was being closely investigated at Swindon in the 1950s.  Castles and Kings were 'souped up' with double chimneys and better superheaters, but the advantages of this (which destroyed the frames of the Kings) were never cascaded  to the 2 cylinder locos; altered chimney profiles improved the Manors and the Ivatt 2MT moguls, though.

 

British main line locos had reached a state by the end of the 1930s in which they were capable of running the timetables with the required loads.  You generally didn't need to pull more than 500 tons of passenger trains at much more than about 70mph, or 800 tons of part fitted goods at 35mph; most loops and refuge sidings could hold trains of 60 short wheelbase wagons, no more, so there was little point in extending loads beyond that; the couplings were not up to it anyway.

 

The 'bottom line' is headed by a £ symbol; there was no point in improving a loco, irrespective of how much engineering sense it made, if costs rose beyond the fiduciary benefits.  Effort after effort was made to improve the efficiency of steam locos post war; Bulleid's oilbath valve gear, Franco-Crosti boilers, feedwater heaters, thermic syphons, roller bearings, Caprotti valve gear, even one or two Giesl oblong ejectors.  Like the compounds of 40 years earlier, nothing much ever really came of them, and the last steam locos built were pretty conventional.  The very last, the last batches of 9Fs, had double chimneys but no more; George Stephenson would have recognised them immediately.

[The Johnster]

1 minute ago, The Johnster said:

I think the usual answer to the question of why British railways avoided higher pressures is fairly simple; they did not want to take on the increased maintenance cost and the down time that it represented.  Put another way, suppose you wanted to replace 31 Kings with higher pressure Castles, as suggested above; you'd need to build perhaps 35 or 36 new locos to cover the work as you'd have to reckon on a higher percentage of your fleet not being available for traffic at any given time, but still have to run the timetable.  

 

After that, the problems start to mount up and the advantages of higher pressure are further eroded.  Sticking with the above example, your HP Castles would probably weigh as much as the Kings and be just as route restricted because of the heavier boilers; the only advantage you've actually achieved is 2½ inches increase in driving wheel diameter, which might give you a few mph at the top end of the loco's capacity on the flat, but is actually just dead weight when you're trying to lift a train over the South Devon banks.  Hawksworth's Counties were originally pressed to 280psi like Bulleid's pacifics, and dropped to 250 later for the same reasons, which meant that all 3 classes carried heavier boilers than they needed to...

 

As has been said, British loco practice was very 'small c' conservative and there was a palpable reluctance to indulge in any change.  Even the Kylechap exhaust, a proven record breaker for the price of a few bits of pipe and steel sheet, was not widely adopted, though it is fair to say that draughting in general was being closely investigated at Swindon in the 1950s.  Castles and Kings were 'souped up' with double chimneys and better superheaters, but the advantages of this (which destroyed the frames of the Kings) were never cascaded  to the 2 cylinder locos; altered chimney profiles improved the Manors and the Ivatt 2MT moguls, though.

 

British main line locos had reached a state by the end of the 1930s in which they were capable of running the timetables with the required loads.  You generally didn't need to pull more than 500 tons of passenger trains at much more than about 70mph, or 800 tons of part fitted goods at 35mph; most loops and refuge sidings could hold trains of 60 short wheelbase wagons, no more, so there was little point in extending loads beyond that; the couplings were not up to it anyway.

 

The 'bottom line' is headed by a £ symbol; there was no point in improving a loco, irrespective of how much engineering sense it made, if costs rose beyond the proven fiduciary benefits.  Effort after effort was made to improve the efficiency of British steam locos post war; Bulleid's oilbath valve gear, Franco-Crosti boilers, feedwater heaters, thermic syphons, roller bearings, Caprotti valve gear, even one or two Giesl oblong ejectors.  Like the compounds of 40 years earlier, nothing much ever really came of them, and the last steam locos built were pretty conventional.  The very last, the last batches of 9Fs, had double chimneys but no more; George Stephenson would have recognised them immediately.

 

[Pandora]

We tend to have a blinkered line of sight in our examination of British steam,  USA manufacturing was so far in advance of us,  such as steam loco frames cast as a single bed complete with cylinders,  steam locos with bhp per cylinder 3 times greater,  the duplex steam   locos which by legend in  normal  passenger service exceeded  the record set by Mallard.  The USA dropped steam quckly as they had  home produced oil supplies on hand for dieselisation.

The industrial might of the USA arose In WW2 as Britain passed the theoretical knowledge of the nuclear bomb across to the USA in the estimation  that Britain did not have the industrial might for manufacture of the weapon

[jim.snowdon]

5 hours ago, Pandora said:

We tend to have a blinkered line of sight in our examination of British steam,  USA manufacturing was so far in advance of us,  such as steam loco frames cast as a single bed complete with cylinders,  steam locos with bhp per cylinder 3 times greater,  the duplex steam   locos which by legend in  normal  passenger service exceeded  the record set by Mallard.  The USA dropped steam quckly as they had  home produced oil supplies on hand for dieselisation.

The industrial might of the USA arose In WW2 as Britain passed the theoretical knowledge of the nuclear bomb across to the USA in the estimation  that Britain did not have the industrial might for manufacture of the weapon

I wouldn't be so certain of that. The Americans had been into bar frames almost from the beginning, and learned the hard way that they are prone to working loose if fabricated as bolted assemblies. The cast steel locomotive frame (or bed) came about through necessity, especially as locomotives got larger, which itself is a pretty good spur to industry investing in the very large machine tools and other facilities needed to handle them. The fact that most railroads bought their steam locomotives from a small number of manufacturers (Alco, Lima, Baldwin in particular) made it easier for them to invest in the specialised equipment required. British Railways intended their larger standard locomotives to have cast steel bar frames, until it emerged that none of the works could have handled them, and there wasn't a business case for the investment required, with the result that British practice remained firmly wedded to the plate frame.

US locomotives had far larger cylinders than British locomotives because they could, and one very large steam locomotive is operationally and economically preferable to two smaller locomotives running in tandem. The prime examples of that are the various large simple Mallet locomotives developed by the likes of Union Pacific and others. It was the forces that such very large cylinders could develop that led to the Pennsylvania's Duplex locomotives, the primary object of which was to get the power from the boiler to the wheels without either bending the rods or shearing the crankpins. They worked, but they weren't overly successful, and not adopted by any other US railroad.

US steam went relatively quickly, not because the US had indigenous oil supplies but because the diesel electric was very competitively sold. Not every railroad fell for the saleman's charm - some, like the N&W, worked out that with good design and effective usage, steam locomotives could be utilised as economically as the new diesels at the time. That would, as time would tell, not always be the case, and as trains have increased even further in size and weight, so have the diesel locomotives and the technology used to get the engine's power onto the rails.

The US had for greater industrial might than Britain long before the war, if for no other reason than that it was a far larger country, with greater natural and labour resources. The war was won on the back of US industrial might.

 

Jim

[kevinlms]

6 hours ago, Pandora said:

We tend to have a blinkered line of sight in our examination of British steam,  USA manufacturing was so far in advance of us,  such as steam loco frames cast as a single bed complete with cylinders,  steam locos with bhp per cylinder 3 times greater,  the duplex steam   locos which by legend in  normal  passenger service exceeded  the record set by Mallard.  The USA dropped steam quckly as they had  home produced oil supplies on hand for dieselisation.

The industrial might of the USA arose In WW2 as Britain passed the theoretical knowledge of the nuclear bomb across to the USA in the estimation  that Britain did not have the industrial might for manufacture of the weapon

The Americans needed much more powerful locomotives, because their distances and load to be hauled was far greater. Their biggest locos weren't normally double headed, because they realised that wasn't the best solution for their long distance freights. Instead they pushed the limits for steam.

 

The coming of diesels turned everything around and it now made sense to multiple head with less powerful individual units.

[jim.snowdon]

9 minutes ago, kevinlms said:

The coming of diesels turned everything around and it now made sense to multiple head with less powerful individual units.

Although for a while, until the coming of the road switcher type locomotives (the "hood" units) the railroads tended to treat their diesels as single entities, even though they might be composed of four locomotives semi-permanently coupled, as in A-B-B-A combinations.

 

Jim 

[brack]

7 hours ago, jim.snowdon said:

British Railways intended their larger standard locomotives to have cast steel bar frames, until it emerged that none of the works could have handled them

Several of our export builders (eg. Beyer, Peacock and NBL) used cast steel frames, but imported them from the US. General Steel Castings (joint owned by alco and baldwin) were the usual supplier. I believe GSC supplied much of the other us builders too, so perhaps the ability to supply large one piece cast frames was much less widespread than simply "not available in britain" but pretty much down to one supplier having developed the ability to produce such things and the world having to beat a path to granite city for them. The ability to create patterns, pour (multiple ladles simulataneously), cool and machine an 80 foot frame is pretty specialised.

[62613]

On 01/08/2019 at 12:35, thx712517 said:

It's like wondering why Churchward didn't make use of much higher superheat temperatures. I could say a modified Castle operating at 275-300 PSI with high superheat (and roller bearings throughout) would negate the need for a King and its associated non-standard components, but at the end of the day the past is the past and cannot be changed. 

You're aware that most of the energy in steam is imparted to it when it changes its state from water to water vapour, i.e the latent heat of vaporisation? Heating the water from ambient temperature to the boiling point for the pressure required adds some energy; superheating adds some more, but not as much as is added when the water is changed to "dry" vapour. You must also be aware that if you expand the steam in a cylinder, it gives back some of the energy as work; not much, I know but some. The effect can be, if the steam is not superheated, but containing water droplets (saturated) If enough energy is taken out, some more of the steam might condense into water in the cylinder. By superheating the steam, enough energy  can be lost without the condensation problem, and it  was found that once you'd done this, there was little point in adding more energy; that is, it was a waste of coal.

Would higher-degree superheat also involve a larger superheater, with the attendant weight penalty?

[Pete the Elaner]

On 01/08/2019 at 12:35, thx712517 said:

Fury puzzles me to no end. They go through all of that complexity and trouble when I would assume a 275-300 PSI conventional boiler would have been far less complex. It was within the state of the art at the time, and I would also assume the weight of a 275 PSI boiler would have been similar or less than Fury's crazy system. Same with LNER's Hush-Hush 10000. Why go to the effort of a Yarrow water tube boiler when you could just crank out a stronger A4 boiler and squeeze it to 300 PSI? 

 

Calculations at the time showed that a much high pressure gave better efficiency.

Experience with Fury proved this was true, but the increased engineering cost was considered disproportionately high.

[thx712517]

2 hours ago, 62613 said:

You're aware that most of the energy in steam is imparted to it when it changes its state from water to water vapour, i.e the latent heat of vaporisation? Heating the water from ambient temperature to the boiling point for the pressure required adds some energy; superheating adds some more, but not as much as is added when the water is changed to "dry" vapour. You must also be aware that if you expand the steam in a cylinder, it gives back some of the energy as work; not much, I know but some. The effect can be, if the steam is not superheated, but containing water droplets (saturated) If enough energy is taken out, some more of the steam might condense into water in the cylinder. By superheating the steam, enough energy  can be lost without the condensation problem, and it  was found that once you'd done this, there was little point in adding more energy; that is, it was a waste of coal.

Would higher-degree superheat also involve a larger superheater, with the attendant weight penalty?

My understanding is when Stanier went to LMS, his initial designs followed Swindon protocol and included the lower degree of superheat that was the norm. Those locomotives performed poorly in LMS hands, but adopting higher degree superheat resolved that issue. There was also an issue in the design of the King front end and draughting that Stanier had to deal with when he essentially copied that design over to LMS. The restrictions were removed and if I recall he sent word back to Swindon so the Kings could be similarly rectified but nothing came of it. 

 

I was merely entertaining the idea of GWR adopting the higher degree of superheat that the LMS did as a way to further improve their power.

 

As to the boiler, what increase in weight would you think result from increasing pressure from 225 to 250, or 250 to 275? I would think at the time the Kings were being designed it would be within Collett's abilities to draw a replacement Castle boiler of higher pressure that could then be fitted to the entire Castle stud. Didn't he do essentially that in converting the Star to the Castle?

[LMS2968]

13 minutes ago, thx712517 said:

My understanding is when Stanier went to LMS, his initial designs followed Swindon protocol and included the lower degree of superheat that was the norm. Those locomotives performed poorly in LMS hands, but adopting higher degree superheat resolved that issue.

Not entirely. The new 5XPs, later Jubilees, certainly had problems but these went further than the degree of superheat and revolved around heating surfaces, boiler ratios and blastpipe / chimney issues. His 2-6-0s and the Black Fives worked well even with the low degree superheat, but it depended too much on things being ideal, for instance, that the boiler pressure was very close to maximum. LMS enginemen didn't work like that and would allow the pressure to vary quite a bit; nor was the coal universally of good quality, and certainly not to the standard the GWR provided.

 

The degree of superheat was raised to address these issues, but the 2-6-0s, many Black Fives, 8Fs and 5XPs carried on with 21 element superheaters, which cannot be considered as more than moderate.

[thx712517]

Ah! I misread the nature of the situation then. It's been a while since I last read my Stanier book. So in essence, Stanier and LMS were operating in conditions that Hawksworth later encountered on the GWR postwar as far as poor coal and suboptimal operations? That would explain Hawksworth's adoption of higher superheat then? 

 

 

[jim.snowdon]

5 minutes ago, LMS2968 said:

The degree of superheat was raised to address these issues, but the 2-6-0s, many Black Fives, 8Fs and 5XPs carried on with 21 element superheaters, which cannot be considered as more than moderate.

But rather better than 14-element. The usefulness of superheating, and the degree thereof, depends to an extent on the normal duties of the locomotive, in particular the typical distances between stops.

 

Jim