How did engineers choose how high the boiler pressure should be?

[Matthew T]

I remember when Chris Eden-Green's YouTube channel was very young and his videos were free, and learning that the GWR 5700s has higher boiler pressure than the LMS Jinties, and were more powerful because of it.
This kind of felt like you wanted the boiler pressure to be as high as possible, which obviously sounds wrong. What were the drawbacks of high-pressure firetube boilers, and vice versa?

 

[randyrippley]

The boiler pressure - within reason - would presumably have been as high as possible given the physical safety limits of the welding/riveting/metal bending techniques used, and taking into account the heating capacity of the type of coal used.
I see you're in the USA where wood was commonly used as fuel. That simply hasn't got the heating capacity to raise the kind of boiler pressures seen in Europe, or in an oil burner.

The evolution of steam locomotives is really just a history of increasing boiler size and boiler pressure, along with larger fireboxes to fuel them.
The number of wheels increased to carry the larger boilers, while improved valve and crank gear helped deliver the increased power. But essentially the bigger and higher pressure the better - while staying within safety limits

 

[6Gman]

The boiler pressure - within reason - would presumably have been as high as possible given the physical safety limits of the welding/riveting/metal bending techniques used, and taking into account the heating capacity of the type of coal used.
I see you're in the USA where wood was commonly used as fuel. That simply hasn't got the heating capacity to raise the kind of boiler pressures seen in Europe, or in an oil burner.

The evolution of steam locomotives is really just a history of increasing boiler size and boiler pressure, along with larger fireboxes to fuel them.
The number of wheels increased to carry the larger boilers, while improved valve and crank gear helped deliver the increased power. But essentially the bigger and higher pressure the better - while staying within safety limits

You would also need the capacity to create that degree of pressure i.e. you can combine sufficient heat and water to create that level of boiler pressure. I'm no engineer but presumably there's no point having a nominal boiler pressure of (say) 400lbs psi if the firebox/ boiler set-up can only generate 220lbs psi ?

 

[randyrippley]

You would also need the capacity to create that degree of pressure i.e. you can combine sufficient heat and water to create that level of boiler pressure. I'm no engineer but presumably there's no point having a nominal boiler pressure of (say) 400lbs psi if the firebox/ boiler set-up can only generate 220lbs psi ?

Agreed, everything has to be a tradeoff in terms of what you can fit within the available size/weight limits for the available technology. A lower pressure boiler is going to be lighter as it need not be as strong, or maybe for the same mass you could make it bigger and so have more capacity at a lower pressure: that's going to be the question when you're looking at power vs distance/endurance for a given mass of locomotive.
I was trying to be deliberately simplistic when I posted

 

[DerekC]

I am not a steam locomotive engineer, but (as in a power station) the higher the pressure the higher the power that can be transmitted in the steam and the higher the theoretical efficiency. However you rapidly get beyond the capability of fire tube boilers in a simple locomotive. Very high pressure compounds were tried - the LMS "Fury" had a complex three stage boiler driving a two stage compound, operated at 900/250psi at the cylinders - but they always seem to have been too complicated to build and maintain to justify any saving in fuel cost.

 

[coppercapped]

The tubular sections of a boiler are naturally resistant to distortion when subjected to a high internal pressure — which is why all high pressure gas cylinders and the like are cylindrical. However the the large flat(ish) panels making up the sides and top of the firebox tend to flex as the steam pressure builds up. The internal walls of the firebox are linked by stays to the outer wall to hold them in place - but the outer walls are also not cylindrical and tend to flex outwards under pressure.

Materials were not so good in days gone by as they can be now and boiler pressures - and therefore water temperatures - crept up as steels, riveting and welding techniques improved. However there were still a large number of holes and joints to be made in the boiler - both ends of each fire tube, steam pipes, packing for regulator and valves, water feed pipes, washout holes, safety valves and whistles, steam feeds for injectors and ejectors and so on and these had to stay leak free even though the boiler was thermally cycled.

Designers were always chasing higher pressures and temperatures as the 'engine' part of the steam locomotive could be made more efficient, steam consumption was reduced for the same power. Nevertheless experience showed that 250psi to 280psi (17 to 20 bar) was the upper limit for reliable boiler operation at the end of the steam era.

 

[Bevan Price]

I think that the incessant vibration on a moving locomotive - caused in part by the jointed track then present on most railways - probably created stresses that - coupled with high steam pressures - helped to cause metal failures at any weaknesses in the boiler.
Static or marine boilers would be less prone to vibration effects.

 

[Taunton]

The benefits of higher pressure are eventually outweighed by the downsides.

A further benefit is, to achieve the same power, you can use smaller cylinders. As in a number of designs they had reached a limit of cylinder diameter within the loading gauge (outside) or space available between the frames (inside), this can be useful. Having more than two cylinders is generally just because you can't get the power needed from two.

But boilers simplistically need to be thicker (thus heavier) and all their myriad components are under greater stress, which means more maintenance work. British designs from multiple designers seem to have topped out at 250psi. Both the GWR County and the SR Merchant Navy went for 280psi in the 1940s, and both later retreated to 250psi, which is just a simple case of resetting the safety valves. Hawksworth on the GWR was trying for something close to a Castle power without needing more than two cylinders, with all their additional weight and mechanical complexity. In the 1948 loco trials the Merchant Navy, still at 280psi, was noted for excessive coal consumption, the opposite of what might be expected.

Tuplin, no mean mechanical engineer, in a number of his books was very dismissive of high boiler pressure and gives one plenty to read about the subject.

 

[Matthew T]

The benefits of higher pressure are eventually outweighed by the downsides.

A further benefit is, to achieve the same power, you can use smaller cylinders. As in a number of designs they had reached a limit of cylinder diameter within the loading gauge (outside) or space available between the frames (inside), this can be useful. Having more than two cylinders is generally just because you can't get the power needed from two.

But boilers simplistically need to be thicker (thus heavier) and all their myriad components are under greater stress, which means more maintenance work. British designs from multiple designers seem to have topped out at 250psi. Both the GWR County and the SR Merchant Navy went for 280psi in the 1940s, and both later retreated to 250psi, which is just a simple case of resetting the safety valves. Hawksworth on the GWR was trying for something close to a Castle power without needing more than two cylinders, with all their additional weight and mechanical complexity. In the 1948 loco trials the Merchant Navy, still at 280psi, was noted for excessive coal consumption, the opposite of what might be expected.

Tuplin, no mean mechanical engineer, in a number of his books was very dismissive of high boiler pressure and gives one plenty to read about the subject.

So a branch line might not be able to tolerate a heavy boiler? I had assumed it was the general size of the locomotive but not necessarily the construction, but this makes good sense!

 

[edwin_m]

So a branch line might not be able to tolerate a heavy boiler? I had assumed it was the general size of the locomotive but not necessarily the construction, but this makes good sense!

The higher pressure boiler would probably have been more efficient in its use of water, so the extra weight of steel might have been counteracted by less water in the boiler itself and in tanks waiting to be used.

 

[Spartacus]

The higher pressure boiler would probably have been more efficient in its use of water, so the extra weight of steel might have been counteracted by less water in the boiler itself and in tanks waiting to be used.

With a higher pressure boiler you're likely to need to boil more of it though.

It's worth remembering that the boiler pressure is only part of the equation for power though, what you do with the cylinders makes a great deal of difference, a greater bore with a lower pressure can produce the same, or more power than a lesser bore with a higher pressure.

Going back to the example at the top, shunting and trip locos tended to be built along the lines of being simple and cheap in built, maintenance and operation. If they were too high you might build no new engines at all, and just use old ones. A lower pressure boiler and associated components is simply cheaper, and as you wanted something reliable unless something better was needed you might as well continue to build an old design. A lot tended to have very long production lives, in essence due to being a update of a 1899 design the jinties were in production of 32 years, the even lower pressure J72s were in production for a remarkable 53 years!

By comparison too you've got designs like the Hunslet Austerity which got nearly 23,870 lbs from only 180 psi, while Peckett pushed their OQ to 200 psi, the same as a 5700, and an eye-watering 29,527 psi!

 

[Taunton]

With a higher pressure boiler you're likely to need to boil more of it though.

No, less. Because each unit of water converted to steam will contain more energy, so for the same work you need less of it.

You do need to boil it for longer though. Theoretically it ought to need the same amount of fuel (less water boiled for longer) but that depends on the efficiency of the boiler, for which we need a mechanical engineer to comment (hello grandfather can you inspire me here).

Using less water is not normally an issue for railways, except in desert areas (not the UK today, looking out of the window just now …). It was however for steamships, which needed to be economical with their fresh water supply, leading to the complexities of condensers and compounding, triple or even quadruple expansion, obviously starting at a high pressure

 

[Lucan]

Apart from the need for thicker steel and stronger joints in a higher pressure boiler, many of the auxiliary items such as stop valves and glands were designed for, and had become standardised for, a pressure of around 250psi. If you went to a higher pressure you needed to take a quantum step up in the design, cost and maintenance of these items. Later coal and oil fired power stations worked at much higher pressures (some even up to 3000 psi) but they needed some very fancy seals on glands etc.

Also because most steam locos were single expansion, it was not mechanically possible to take advantage of a higher pressure - you would be discharging the exhaust at a rather high pressure which would be very wasteful. In a compound engine the exhaust from the primary high pressure cylinder would go on to a second, larger but lower pressure, cylinder to extract more work from it. Steam marine engines were often triple expansion, but they (like power stations) had condensers cooled by sea water (or river water in the case of inland power stations) so the final exhaust was a partial vacuum to suck as much work out of the steam as possible. Non-compound and non-condensing steam railway locos were very inefficient thermodynamically.