Boilers & Accidents

This leads to my favourite possibly apocryphal tale of Bulleid speaking to a crew at Waterloo on one of the then-new West Countries, and asking the fireman what he thought.

"You'd have thought the bloody fool who designed them would have had the sense to fit dampers," came the fireman's verdict (Col Rogers suggests he presumably did not recognise the CME).

"Dampers! I do not fit my engines with devices to prevent steaming."

 

It's deeper at the back to accommodate the combustion chamber, which protrudes into the barrel. There are nice wide water legs around there to promote good circulation. Worth mentioning that LNER Pacifics have straight conical tapers, so they also have some taper on the bottom and also have combustion chambers. I believe the LMS went more for heavily shaped throatplates to accommodate some form of combustion chamber, but I've not got drawings of one of their boilers to hand. Narrow firebox boilers don't tend to have full-on combustion chambers, so the need for taper underneath rather goes away.

 

It was mentioned in Churchwards 1908 boiler water circulation paper that the taper boiler was originally invented in Canada, the purpose being to reduce the surging of water when the locomotive hits a snowdrift. Clearly this would only work if the taper was at the top.

 

The origin of the taper boiler may be true but the reality is that having a taper on top would have no effect on this. The top part of the barrel is filled with steam. The main reason designers opted for a taper boiler is weight saving, not reducing the impact of snow drifts.

 

Because it would limit the number of tubes in a boiler of given diameter (and therefore weight) ...

Imagine the front tube plate. It is a circle of a certain diameter, and that diameter fundamentally limits the number of tubes that can fit (dependent on diameter of each tube and spacing between them to leave adequate strength in the remaining metal). The diameter of the front of the barrel needs to be only just big enough to surround that tube bank.

Once you have set that limit, you require the same arrangement of tubes at the firebox end taking up the same amount of space; however, the firebox has to be wider because you require space around the inner firebox for the water legs down each side of the firebox (typically about 3" at the foundation ring, and wider as you go up the firebox). So the barrel either has to be wider at the firebox than the smokebox (which leads to a tapered boiler); or else the tube bank doesn't fully exploit the space available in the front tube plate (which is the case on a conventional parallel boiler); or you end up with a strange “step” around the firebox to take account of the firebox being fundamentally larger than the barrel (as was seen on many Victorian boilers, right up to the early Churchward-designed Belpaire fireboxes on Dean locos).

Fundamentally, the boiler barrel needs to be as wide as the outer firebox at the back, but not much wider than the inner firebox at the front. That either means a taper barrel, or a boiler that has too much volume, i.e. weight. As @Steve says, the benefit of a taper boiler is primarily weight saving, at the expense of more complicated construction.

Here's a couple of photos to illustrate:

Taper boiler (on 80151) - note how the tubes go out to the edges of the front tube plate:



Parallel boiler (on Baxter) - see how much space is wasted, i.e. not as many tubes as potentially could fit in the available area, particularly down the sides.



Tom

 

Wow! We're rapidly heading into M.I.C territory here .....

 

PS - meant to add: the above is a somewhat crude approximation. There is actually a huge amount of subtlety in boiler design, and the shapes of various components. A Belpaire firebox, for example, is an enormously subtle shape: for example, wider at the front than the back, in the Churchward iteration considerably higher at the front than the back; the sides curve in a subtle S-bend; the water space gets wider from bottom to top; the backhead and front tube plate may both slope. The tapers on the barrel, as discussed above, may be even along the length, or a mix of a tapered ring and a parallel ring; the taper may be a pure cone (i.e. equal on top and bottom) or an oblique cone (flat on one side and tapered on the other, as on a Bulleid). Then all of that subtlety may be accurately followed by the boiler cladding, or disguised: the apparent flat sides of a Midland 4F firebox cladding aren't the same shape as the firebox underneath. All of those subtleties affect weight, weight distribution, water circulation, gas flow through tubes, ratio of how much heat goes into evaporation and how much into superheating (if fitted - the more you superheat steam, the less heat there is available to boil the water in the first place, so there is a trade off between superheating nd steam production for a given size of grate). So lots of subtleties and lots of compromises, and that is before you then get to workshop considerations such as "will this new boiler require new flanging blocks - if so, is the design tweak worth the expense?.

I can't recommend AF Cook's book, "Raising Steam on the LMS", enough for a discussion of a lot of the above, through the lens of one company.

Tom

 

I'm not sure that's necessarily quite the case. I'm looking at a GWR Std 7 boiler drawing at the moment, and the arrangement of the tubes at the firebox end is subtly different from the arrangement at the smokebox end.

 

So Manning Wardle ‘SirBerkeley’ is not the only loco to have more holes in the smokebox than the firebox tube plates

 

I don't understand your reply. Water is incompressible, and when the train hits the snowdrift, the volume of water which can surge forward will be limited to the volume above the water line (when on level track).
This will be lower for a locomotive with a taper on top. I agree it has nothing to do with Bulleids though.

 

Now I can't understand you statement that the water which can surge forward will be limited to the volume above the water line. Surely, there isn't any water above the water line?

 

OK, by the water line, I am referring to the level of the water immediately before the train hit the snowdrift.
When this occurs, the locomotive slows down sharply, but the water will attempt to keep going, and "pile up" into the space which was above the original water line.
When the water eventually stops it will create a water-hammer effect onto the front tube plate.
With a boiler with a top taper this space will be smaller than for a parallel boiler or a boiler with a bottom taper. In my opinion it would be better to slow down the water (as far as possible) at the same speed of the locomotive, rather than the surge of water hitting the front tube plate a little later and with significantly greater force.

 

I hate to engage in these learned engineering discussions but I've never heard of "a water hammer effect onto the front tube plate". Surely a water surge such as you describe is likely to have more serious consequences to the exposed firebox crown sheet than it is to the front tube plate. But I'm happy for someone to tell me I've got it all wrong.

Peter

 

I think the idea is that a reduced volume at the front means less water can go there (when you hit the snowdrift), thus leaving more at the back to keep the crown sheet covered. But presumably Churchward's reason for adopting taper boilers was to use less steel and save weight.

 

Exactly. I doubt that it has anything to do with worry about stressing the front tube plate.

Peter

 

Water hammer is a known effect in full pipes, but I don’t think you can get it in a boiler which is part full and the contents are free to slosh around.

My own thought about crashing into a snowdrift would be along the lines of the mechanical shock propagated through the boiler, and whether that might cause for example broken stays or leaking tubes over time.

Tom

 

You beat me to it, Tom!

 

Thread drift or snow drift?

 

It’s all interesting stuff, though.

 

Does the effect on the tube plate vary when Snowploughs have different sets of buffers?

My Plough is fitted with sprung buffers but the NRM's plough at Shildon is Dumb.