|
Post by Roger on Nov 16, 2013 18:22:44 GMT
Ah yes, the purpose of the thread... I'd almost forgotten. With Julian's help, I'm homing in on modifications to SPEEDY's boiler that include the following.. 1) 1/4" reduction in the tapered boiler outer wrapper which incidentally will now be 1/8". This is to allow for the lagging which is missing on the plans. 2) Obviously the original tube arrangement is not going to fit into that reduced space, so this has been revised with three larger superheater tubes instead of the original five. There are now 21 tubes instead of the original 26. I make no justification for this other than I rely on the experience of those who I judge to know better than both LBSC and me. It's my risk and it's one I'm happy to take. 3) The top part of the outer firebox has also had 1/8" shaved off all round, also for lagging. 4) The inner firebox and wrapper are subtly different but only to guarantee the clearance for the water jacket is maintained. 5) The inner 'Piston ring' joint between the barrel and the firebox has been discarded in favour of a double flanged throatplate. The final details of that are still to be tied down, but the intention is to make the flange on the outside of the barrel, at least at the bottom to assist water flow in that region. Again, I accept this advice because it sounds reasonable to me. The firehole will be lowered when I've got a handle on where the cab floor is. I'm told that it's far too high up on the model.
I've been poring over the few pictures I've found of the full size locomotive in restoration and have changed the rear buffer to be the full height of the rear of the frame, as it is in the photo. I also note that the plans don't show the correct position of the rear set of the two sets of steps. The plans show them facing rearwards when in fact they face the side. I've corrected that. I'm following the advice of several members in detailing a generous radius on the rear of the bunker. I'm guessing that I'll need to form a section for that and let it into the corner. Advice on this kind of procedure would be welcome, it's not my forte. I've just quickly lashed up a square cornered version of this to check the general dimensions and fit to the boiler. Much more attention to this area is needed. I don't have any pictures of the backhead so I'm completely in the dark as to what that should look like. I'm inclined to attempt something a little closer to scale for that. There's still loads to model and a thousand questions.... watch this space.... All observations, tips and ideas are most welcome. I have no experience of this kind of model so it's a steep learning curve.
|
|
|
Post by Deleted on Nov 16, 2013 20:03:56 GMT
I also note that the plans don't show the correct position of the rear set of the two sets of steps. The plans show them facing rearwards when in fact they face the side. Interestingly, the rear footsteps did face backwards when the 15XXs were built. They were altered to side-facing at a very early date. The same fixing holes were used in the rear beam.
|
|
|
Post by ejparrott on Nov 16, 2013 21:10:40 GMT
Double flanged throatplate, beautiful.
What water gap have you got between the tubes? Also, have you run the length/bore calculation, and a free gas flow area versus fire grate area formula's?
|
|
|
Post by Roger on Nov 16, 2013 23:32:02 GMT
Double flanged throatplate, beautiful. What water gap have you got between the tubes? Also, have you run the length/bore calculation, and a free gas flow area versus fire grate area formula's? The minimum is 2.27mm but there's more room in other places. You can see the layout on the attached model. I've also attached the spreadsheet for the boiler calculations for what they're worth. To be honest, I don't think these rules of thumb are of much value, but then I'm new to this. Unless you can model the gas flow through the boiler with FEM, I don't think there's a hope of predicting what's actually going on with such simple measures as free gas flow area versus fire grate area. If the boiler was twice as long, I don't think such a formula would yield sensible results. It doesn't take into account what happens when the tube size varies. Hundreds of 1/8" bore flues wouldn't give the same result as one of an equivalent size. Maybe I'm wrong, but I'd need to see a stack of proper evidence before I believed these measures were valid. There's a blanket assumption that one superheater tube is equivalent to one flue, but this is completely arbitrary in my view. I've ignored that and used the actual area of the superheater tube in the calculation. I know that there is some restriction, but putting some other arbitrary value on that myself is no better than the assumption I'm critical of. I can see why designers want some guidelines to help them, and it's not surprising that certain formulae become popular. I'd be curious to know where this particular one came from and whether it was just someone's pet idea. Anyway, no hate mail please, I'm just being honest as an outsider who likes to see the evidence for something before accepting it as fact. I'm sure that loads of boilers have been designed using these rules, but that doesn't mean that this is the reason they were successful.
|
|
|
Post by Roger on Nov 16, 2013 23:33:37 GMT
I also note that the plans don't show the correct position of the rear set of the two sets of steps. The plans show them facing rearwards when in fact they face the side. Interestingly, the rear footsteps did face backwards when the 15XXs were built. They were altered to side-facing at a very early date. The same fixing holes were used in the rear beam. Well, you learn something every day! Still, I may as well make it how the only surviving one is, it's my only reference to the real thing.
|
|
jma1009
Elder Statesman
Posts: 5,901
|
Post by jma1009 on Nov 17, 2013 1:32:56 GMT
it may be of interest to know what design parameters myself and roger have used in arriving at an improved SPEEDY boiler. the starting point has been keith wilson's BULLDOG boiler and the boiler of one of my own award winning locos. the BULLDOG boiler is the GWR standard No.2 boiler - the 15XX No. 10 boiler(SPEEDY design) has the same 'size' boiler so far as firebox and barrel are concerned in outside dimensions although the firebox and barrel are less in length. the same GWR flanging blocks were employed on each. careful consideration of the LBSC design showed it was overscale in diameter and firebox width over the belpaire firebox and shape of the belpaire firebox. hence the cleading and cladding problems. i considered the original superheater flue arrangement to be totally OTT and adding nothing to the loco, 3 x 1" superheater flues being quite adequate. roger may even incorporate radiant superheaters in due course, with considerable advantage. we have tweaked the boiler dimensions to get something pretty close to scale outwardly, and roger's tube arrangement is in my view far superior to that of LBSC. the free gas flow percentage of grate area is still very generous, and water circulation and space much improved upon the cramped LBSC layout. roger will fit proper bushes throughout updating the original design which doesnt come up to the current specs, and add additional stays on the backhead to inner firebox. the proper double flanged throatplate is something ive always been keen on as aids water circulation and provides a very sound boiler construction. we have tweaked a few other details and therefore roger's boiler should have a bit more up it's sleeve than the original and be an easier boiler to build and an easier boiler to make look like fullsize.
this is the sort of approach to model engineering of which i heartily approve - practical, sound, carefully considered and worked out, more efficient (therefore a better steamer), and closer to fullsize outline.
after many years playing about with these things (and driving/firing same) one can almost tell at a glance what will make a successful boiler - unfortunately IMHO the original SPEEDY boiler doesnt fall into this category. it has numerous design deficiencies and dubious areas of design, plus prevents the builder building a boiler that is anything like the fullsize boiler plus fittings in the proper place!
i just pull my hair out when looking at the original design of boiler!
cheers, julian
|
|
|
Post by suctionhose on Nov 17, 2013 4:38:24 GMT
I can see why designers want some guidelines to help them, and it's not surprising that certain formulae become popular. I'd be curious to know where this particular one came from and whether it was just someone's pet idea. Striplar, I can't comment on the origin of the rule of thumbs or how they equate to engineering formulae. I think I picked it up from K N Harris or Martin Evans during early years of live steaming with my Dad. We had purchased a loco that turned out to suffer from too much grate area for tube area & tubes too small for their length. We also removed blown superheater elements in another engine and ran saturated a until we'd gathered the know how to put them back. The first thing is that satisfactory grate, boiler and smokebox layout is the key to success. Having said that the working envelope is wide. To stray from roughly generic FS proportions (I don't mean for tubes) ie freelancing some weird and wonderful machine that hasn't been seen before, often gets you into trouble. Generally speaking, fullsize prototypes had good reason to be the way they are and reproducing them in miniature lands you in good stead for weight distribution, steaming capacity, etc, etc. So to follow the 15XX for Speedy will take care of a lot of variables without over analyzing them. FS steam loco's differ from models obviously because you can't scale nature. In FS grates are designed for best combustion (fuel air mixture), draughting is designed for freedom of exhaust as well is not lifting the fire while providing adequate steam production. Boilers are designed with fuel economy, steam production and superheat in mind. A diverse set of requirements to find a good compromise for. In models we tend to arrange grates so the ash can be cleared. Typically there are much larger air spaces and slower air velocities than FS. To then regulate the fuel air mixture we open up the blast nozzle until adequate but not excessive steam is produced, combustion is Ok and the hot sparks out the chimney are bearable. All this dovetails into less exhaust back pressure, better performance and a set of compromises that work well in miniature although the rationale behind them is somewhat different to FS. To labour the point further, to generate sufficient smokebox vacuum to operate with a 6% air gap grate in a model would cause excessive exhaust back pressure and provide no means for the ash to drop out. It simply wouldn't work for long. Given that, the primary issue is providing the means to burn coal continuously in a tiny firebox. A close second is to supply enough air for combustion and to produce sufficient heat to continuously generate steam against demand. That is, generating smokebox vacuum without undue exhaust steam back pressure. Smokebox proportions commonly followed in models are rooted in fullsize practice because the blast pipe / chimney is an ejector, a venturi and physics is physics. The third thing is, having developed suitable smokebox vacuum, the volume air flow through the fire has to pass through the boiler but not so fast that the heat doesn't transfer. Wherever the l/d squared = between 60 & 80 for tube diameter come from it is very reliable in practice. I tend to aim toward the low end so my tubes are larger rather than smaller to gain more of them. Reason being that they are less susceptible to sooting up, they can take more sooting before the diameter becomes restrictive. I've gone as low as 50 (traction engine - to promote natural draught). I not ever found insufficient heating surface to be a problem nor efficiency compromised. The model has to first burn coal successfully before all else. The engine we brought that had all the problems had a figure above 100. The initial observations were that it wouldn't steam because the fire was too dull even when working hard. i.e. not enough air volume because the tubes restricted the air flow to the smokebox. The grate area was disproportionately large (it was a freelance of unconventional appearance) and this exacerbated the problem. The solution was to optimize the smokebox vacuum - measured with manometer while adjusting proportions - reducing grate area by 30% and keeping those long small tubes clean. The engine had no tolerance for sooting up but did end up steaming reliably overall. The opposite happened when we removed superheater elements and found all the gas rushed down the flues burned the smokebox door and not heating the water. After restricting them down with a ferrule about the same bore as a tube then the boiler become successful again. Your flue dias are going to be determined by what fits in the tube nest and what size tube you need to supply steam the the cylinders. The moral of the story is that there is a wide range of successful compromise possible provided things are roughly centre of the envelope. Tweaking the smokebox, changing a blast nozzle dia by 1/64 up or down will provide the fine tuning necessary. Not a concise answer to your question but solid experience intended to reassure you. Incidently, Yesterday, between noon and 4pm, the loco pictured earlier ran 14 kms double heading a 7 car train of children / adults without a moments bother. Here we are with a half a load on near the top of the 1 in 57.
|
|
|
Post by Roger on Nov 17, 2013 8:48:56 GMT
That's actually a great answer Suctionhose. It explains the juggling act of proportions and what each element of the design is expected to do. It's a bit like the valve gear issue, where changing one dimension affects all of the others. Heaven knows how many failed experiments the pioneers must have tried before coming with the proportions we see today. I'm not going to stray into unknown territory with the design because I simply have no experience myself. I can only rely on people who have built and seen enough to guide me and I'm in their hands. In the end, the risk is entirely mine and it feels like a small one. Julian has posted the logical reasoning behind his advice and it sounds very sensible to me. It will certainly be easier to make in terms of the smokebox arrangement and will avoid having tubes that are too close to the flanges. The real gain is in the lagging and scale appearance though.
That's an awful lot of work your locomotive did yesterday!
|
|
|
Post by ejparrott on Nov 17, 2013 9:49:00 GMT
2.27mm sounds good to me. Looking at your tube layout, I'd suggest just pulling the top row of tubes down slightly, just get them a little bit away from the edge of the plate. It'll make life easier for you to drill if you're not trying to drill in to the flange when you break through, and it'll also help a little to stop the solder wandering off round the corner - I've had that problem in the past!
Do us a favour, just spend a minute putting length and bore through the Keiller proportions, lets see what you get.
|
|
jma1009
Elder Statesman
Posts: 5,901
|
Post by jma1009 on Nov 17, 2013 11:22:23 GMT
i was very interested to see the pic of the smokebox end of the 7.25"g ROMULUS boiler in the ads section modeleng.proboards.com/attachment/download/3155ive never driven a ROMULUS but if the boiler is built to drawings it doesnt appear to have many tubes and as is often the case on smaller 7.25"g locos (for reasons that ive never understood) no superheater flues. cheers, julian
|
|
|
Post by Roger on Nov 17, 2013 12:04:09 GMT
2.27mm sounds good to me. Looking at your tube layout, I'd suggest just pulling the top row of tubes down slightly, just get them a little bit away from the edge of the plate. It'll make life easier for you to drill if you're not trying to drill in to the flange when you break through, and it'll also help a little to stop the solder wandering off round the corner - I've had that problem in the past! Do us a favour, just spend a minute putting length and bore through the Keiller proportions, lets see what you get. Ok, I'll double check that top flange clearance. I actually dropped that and the firebox crown my a millimeter, so I may just change that back. I've put the revised figures for Speedy into the spreadsheet that was kindly sent to me by John Baguley but I ought to ask him if he's happy for me to attach that here before I do that. The key figures that come out are as follows though... The Tube area as a % of grate are comes out at 14.79%, down from 19.1% Ee = 0.193 which is unchanged Eb = 63.396 up from 49.084 Eo = 12.264 up from 9.489 Kt = 89.119 up from 89.115 Total tube area 2.53 down from 3.26 Hopefully that means something to you, I'm afraid it doesn't to me beyond the obvious reduction in surface area
|
|
|
Post by Roger on Nov 17, 2013 12:06:33 GMT
i was very interested to see the pic of the smokebox end of the 7.25"g ROMULUS boiler in the ads section modeleng.proboards.com/attachment/download/3155ive never driven a ROMULUS but if the boiler is built to drawings it doesnt appear to have many tubes and as is often the case on smaller 7.25"g locos (for reasons that ive never understood) no superheater flues. cheers, julian That's a real surprise. From my position of complete ignorance, I would have thought that the larger the locomotive, the closer it would be to full size practice ie many more tubes and definitely superheaters. This design looks like it goes in the opposite direction, with less of those things than you find in a 5" loco. Can anyone explain why that model's arrangement makes any sense?
|
|
|
Post by andyhigham on Nov 17, 2013 12:47:47 GMT
Many of the larger 7 1/4 locos are designed for commercial passenger hauling, as such will spend hours in steam. Fewer large diameter tubes will allow longer runnig time without stopping for cleaning
|
|
uuu
Elder Statesman
your message here...
Posts: 2,808
|
Post by uuu on Nov 17, 2013 13:26:31 GMT
I suspect many narrow guage models miss out the superheater to stay true to prototype - even if they're not really based on any particular engine - true to the generic NG engine. This thread has noted Speedy's deficiencies as a model of a 15xx. If you were building a Kerr Stuart Wren, for example, would you fit a superheater (I know Julian would say yes, so would I, having got a bit damp driving one)?
|
|
jma1009
Elder Statesman
Posts: 5,901
|
Post by jma1009 on Nov 17, 2013 15:18:44 GMT
hi wilf, i did suggest to Reg that he fit superheaters but as you know he didnt and has told me many times he regretted this decision! i dont really want to re-open the non-superheat -v- superheat debate though! cheers, julian
|
|
|
Post by Deleted on Nov 17, 2013 16:52:36 GMT
While we're nit-picking, is it calliper, or caliper? Interesting point. I've checked two dictionaries (Collins and Longmans) and they both agree that either spelling is valid. 'Calliper' seems to be more common in the UK, 'caliper' more so in the U.S. Anyway back on topic, here are some pics (not very good ones) of the backhead on 1501:
|
|
|
Post by Roger on Nov 17, 2013 17:11:40 GMT
After all this, would it not make more sense to alter the Bulldog boiler design? A fair point, but any modifications are going to have to pass the inspector so I'm not sure if that would help. Again, without any experience of boiler inspectors and their ways, I have no way of knowing what they will say. If they need something changed then I'll just change it. Time will tell, at the moment it's just a 3D model and I'm enjoying the learning experience.
|
|
uuu
Elder Statesman
your message here...
Posts: 2,808
|
Post by uuu on Nov 17, 2013 17:23:33 GMT
I will take quite a while to make a 3D computer model if the backhead, I think. Very atmospheric pics.
|
|
bhk
Part of the e-furniture
Posts: 458
|
Post by bhk on Nov 17, 2013 17:45:34 GMT
i was very interested to see the pic of the smokebox end of the 7.25"g ROMULUS boiler in the ads section modeleng.proboards.com/attachment/download/3155ive never driven a ROMULUS but if the boiler is built to drawings it doesnt appear to have many tubes and as is often the case on smaller 7.25"g locos (for reasons that ive never understood) no superheater flues. cheers, julian there are two versions of the romulus boiler, one with 17 smaller tubes and the other with 12 larger tubes, from speaking with a commercial boiler maker recently the larger tube version is much more popular.
|
|
|
Post by Roger on Nov 18, 2013 21:44:50 GMT
Oooooh! My first glimpse of the cab! I'll squirrel those away for safe keeping, thanks superseven.
|
|