Rosebud grates

Very interesting series of articles on this topic. Here's my twopennorth

1. The flow resistance of a boiler is made up of several components, outside world to ashpan, through the ashpan through the fire and through the tubes. Doubling the resistance through the grate will not therefore double the overall resistance, and so will increase the velocity through the holes.

2. The tapered inlet to the holes will reduce the entry loss, hence reduce the pressure loss. For a given flow rate this will reduce the air velocity. The traditional shape would be ‘well rounded’. I can imagine that some clever hyperbolic curve is best of all, but could anyone be bothered.

3. Full size ‘modern’ British locos typically had grates with quite low air inlet area. ESCox quotes ‘cast iron sections having air slots cast in, each 11.375” long by 2.25” wide……...giving 37% open space” for the standard class 7. According to the ultimate steam page, fires in full size locos were ‘a few inches thick’ If one patch is half thickness, and the grate itself has low resistance, then a lot more air will flow through the thin bit, which will burn faster, and get even thinner, and so on until there is a hole in the fire. If the grate has more resistance, this effect will be much reduced. Red Devil had smaller holes (need to check), but was designed to be a gas producer fire, only admitting about half the air through the fire, the rest as top air.

4. Those with an enquiring mind might like to try diverting some exhaust steam to the ashpan, and running with the firehole door slightly open. This should give a lower fire temperature, as the steam/coal reaction is endothermic, producing hydrogen and carbon monoxide. This should give less clinker, but more flame in the firebox. Probably needs a brick arch to allow time for the gasses to burn. Whether this will work in such small fireboxes is a matter for experiment, but it certainly works in 15” gauge (Kirklees Light Railway), you can see the blue tinge to the fire.

I read 'Thought's on Rosebud Grates' with breakfast coffee. Good article well written apart from one small problem : this ignoramus doesn't know what a Rosebud Grate is! Seems I have the same problem as Martin and Peter: on page 27 Leslie writes "Martin's response was exactly the same as Peter's; 'what's that?' His dad explained..."

I see from the photo and text that the grate has tapered holes in a plate rather than fire-bars. Is that all there is to it? If so the detail about tapers, hole size, spacing and air-flow makes more sense to me.

The article and Duncan's comments have widened my ideas about fire boxes. Getting a locomotive to steam is clearly much more involved than setting fire to any old rubbish in a tin box. I imagine it's difficult to get the very best performance out of a fire; too big or too small is bad, and presumably the calorific value of the fuel, impurities, air velocity and volume at input and back pressure on the exhaust can all interact in complicated ways.

Dave

The late Bill Hall reckoned that modelling the heat flows in a nuclear reactor was child's play compared with those in a loco firebox. Results from full size cannot be easily scaled to our sizes without a good deal of care, for a start the grate load (lbs/sq.ft/hr) is a lot lower in our case, but the coal we throw in is in much smaller lumps.

How about fixing the problem by replacing a model locomotive's coal fire with a Dampferzeuger in the tender? Not only no issues with pressure and super-heat, but I'd pay good money to watch a Model Engineer sitting on a mixture of Hydrogen Peroxide and Potassium Permanganate. And if IMLEC is ever to get on Reality Television there will have to be a lot more explosions.

well in principle you could arrange a feedback loop between the reaction chamber pressure and the HTP pump to give a steady steam pressure, but as the breakdown of HTP is catalysed by both copper and iron I'm not sure what you'd make it all from. When you've built it let me know and I'll arrange to be at least 2 counties away!

Your biggest problem might be explaining to MI5 why you want large quantities of HTP, which is a favourite ingredient for IEDs.

this technology was used in the 1950s UK rocket research program, all of whose launches were successful, beat that NASA! Of course as usual we gave up when it became too expensive, and so wasted the money which could have been spent on something else.

There was competition between the ballistic missile people and the submarine people for access to limited supplies of HTP. A number of accidents demonstrated that HTP was NOT a reliable fuel for submarines, so they went for nuclear instead...

Neil

Duncan

Reference the loss coefficient for the entry to a system.

Plain hole eg flat face k = 0.5

Lowest k is using a plain radius of 1/6 the inlet diameter. K = 0.02. Which when applied to the velocity head loss reduces it to next to nothing. Hence the reason a bell mouth intake is used on engine intakes and the entry to a venturi.

Reducing the radius increases k but it never exceeds 0.5 (r = 0).

Using a taper gives good results, it is easy to get down to a k of around 0.1/0.2. Whilst this is an order of magnitude higher than the bell mouth, html the difference in the total system loss would probably not be measurable.

Yeah, done that on my 3.1/2"G Tom Rolt. (Ian Screeton on the KLR was very helpful about it!) It was always intended as a testbed to ascertain whether it would work in a small firebox. Unfortunately, it's still too early to tell if it works or not, as I'm still having great difficulty with other aspects of the loco. I need it to work conventionally at first, and then try it on GPCS. (Having said that, I have tried it briefly when running on the "home trainer", and it didn't appear to have any significant adverse effects, but it was far too quick a test to be significant.)

I fitted hollow stays to admit secondary air above the firebox; it's got a "brick" arch made from titanium, and the exhaust steam feed to the ashpan is controlled from the cab, as well as having an additional live steam feed as well. Making the boiler superheated, whilst maintaining an external regulator like big brother was interesting; I ended up making it a screw-down valve but with a 5-start thread to allow it to pass enough steam.

Instrumentation and measurement of the changes is likely to be an interesting exercise. A Lempor exhaust is also part of the experimentation; I've made several chimney chokes and even more four-nozzle blastpipes with differing hole sizes to try out. So when I've got the blasted steam-operated draincocks to work, I might get somewhere. (Then if I'm happy with the results, I can decide how I'm going to make the boiler for my 9F when I get around to that.)

Anyway, when I eventually get some results one way or another, I'll write it up for ME, but it won't be this year.

Sorry to resurrect this thread, but I'm eagerly awaiting Nigel's update.

By all means experiment, especially if you intend to enter you locomotive in the efficiency competitions; but I can't help thinking we are all in danger of over-thinking things here.

There is an optimum ratio of hole to metal areas, whether a Rosebud-pattern or conventional grate; but I remain to be convinced by some of the discussions I have seen about them, especially the supposed effect of tapering the holes in the former.

As with a bar-grate, the taper's main value is simply in mitigating clinker choking the grate.

Holes that converge upwards towards the fire will increase the air's velocity, and reduce its pressure, through them (think of what happens in the injector), but it's pressure you need to lift the fire-bed as claimed apparently a Good Thing. Holes that diverge upwards will just choke with ash, of course. It would probably need a rather thick grate too, for the hole to be long enough to give any real gas-flow effect.

There are just too many variables at play to be sure of what modification really does what to improve combustion.

One of them is the nature of the fire itself. Whilst Duncan rightly points out we use much smaller lumps of coal than on a full-size locomotive, they are often over-scale; and I have known some drivers of miniature locos aver any sized will do if it fits through the door. The recommended coal size for full-size locos was actually about the size of a man's fist - and that for a firebox the size of a small bathroom. So maybe a bit larger than a broad bean would be about right for 5" and 7-1/4" g locos; and similarly-sized traction-engines, depending to some extent on fire depth.

Rosebud grates do seem to have a good name, provided the ratio of holes to metal is right for the boiler, and are simple to make although could not be readily of rocking pattern and may be harder to clear in operation. I may make one for my steam-wagon's boiler, with its circular grate presently of bar type. I doubt though they are any more exemplars of Advanced Thermodynamics and Combustion Chemistry in Practice, than a conventional grid.

Thanks for bringing a thread up that I wouldn't have read otherwise.

Recently I have had suggestions stating that I should fit a rosebud grate to my build of 4472, I wouldn't as it's not prototypical and thus would negate the very reason why I am building this model. Having read the posts in this thread, particularly Nigel's last post I feel more positive about my position on this.

Although I am hearing more about rosebud grates these days most loco's that I see have the traditional bar grate, in fact, I can only recall seeing one rosebud grate in the flesh. Also it's a valid point about the rosebud being more difficult to clear of clinker which I would think to be of more relevance for a wide body firebox that may not have the ability to easily remove the grate.

Interesting discussion...

Pete

Sorry to have kept you in suspenders, Duncan! I've remained silent on my experimentation with Cudyll Bach (Tom Rolt) because after struggling for quite some time I came to the disappointing conclusion that the arrangement I have just doesn't work in small sizes - or it's just too finnicky and sensitive to varying steam demands to control. In the end I abandoned any attempt to run it with GPCS and just ran it conventionally. One of the big reasons for giving up on trials with it is that we no longer have our track at Eggborough so there's no real possibility of popping up to the track when I feel like it for further experimentation. Running on the turbo trainer doesn't really help a lot.

I'm also very much involved with a later build - Ashey, an Isle of Wight 0-4-4T in 5"G which is now nearing completion, and then it's going to be a 5"G Invicta, which will owe absolutely nothing to LBSC! Going back to Ashey, it's taken me about three hours just now to fasten two nuts inside the smokebox to secure the handrail knobs... you get days like that. Ashey has a Rosebud grate and it seems to make steam OK on the turbo trainer. Hopefully I've now sorted out the interference fit of the piston and cylinder at steam temperatures so next time I try it, it might actually move...

It's like Citizen Kane - if they all knew what Rosebud is, it would spoil the movie/article.

Neil

Yes, just tapered holes instead of firebars.