Optimal number of boiler tubes.

[Steamchick]

Please note: Wok burners are set-up with a jet size for NATURAL GAS, which is as a low pressure, so NOT SUITABLE for Propane or Butane at psi pressures. But if you get a "Natural Gas" regulator, maybe this will reduce LPG suitably so they can be used in a loco on the track?
K2

 

[Nerd1000]

Thanks Martin. Good info!
I always add a "DRYER/SUPERHEATER" in my boilers as the benefits of energy transmitted from burner to cylinder using superheat are just something I new to be worthwhile from being a teenager and seeing the Coal fired boilers in a power station (for 150Mw generator turbines), with superheaters, water pre-heating economisers, condensers, etc. - I worked as a "shutdown" boiler cleaner in the summer hols as a student. Crawled in and outside Babcock's finest coal-dust fired monsters..
On my titchy gas fired boilers the superheater/dryer is at least a steam pipe from the take-off fed through a flue and back - or through the firebox, or a coil in the smoke-box. Depends on physical constraints as much as how much superheat. My 3in vertical Boiler with a coil in the firebox is super-hot! Tarnishes brass and copper pipe easily. at 20psi it really makes a difference to the performance of little engines, typically powering bicycle dynamos.
Many people forget Physics... and think "Pressure" is all you need. But the steam is just an energy transfer medium, so "Enthalpy" is what makes it work (in my head): I.E. Energy as the steam leaves the boiler minus energy as it leaves the exhaust of the engine. I.E. More energy "IN" means more power generated. - But if I am wrong, please teach me, as I was hopeless at Thermodynamics in college!
Must get around to studying your paper again, and having a look at real numbers of Coal fired versus Gas fired, flue sizes, etc.
Incidentally: I have a diagram showing temperatures of radiant Media: but I can't find it! Wire mesh works up to 1200deg.C, but Ceramic only 900deg.C. Higher and they degrade. New foam matrices work up to 1400deg.C - but cost hundreds of £ for a slab!
Cheers!
K2

Regards superheated steam, you also need enough expansion ratio (and a low enough exhaust pressure) to make use of the extra heat.

Coming from the infernal combustion side of things, a typical ICE hasn't got enough expansion ratio to fully utilise the heat in its combustion gases, the exhaust valve opens with several bar of pressure still in the cylinder. Hence the benefit of a turbocharger, which harvests some of the residual energy by further expanding the gases to atmospheric pressure before they leave the exhaust.

Once your gases are at atmospheric pressure an open cycle engine cannot get any more work out of them. Steam engines however can go further by using a condenser to pull a vacuum on the exhaust, effectively allowing more expansion to take place without fighting air pressure.

 

[Steamchick]

Just finished studying the Professor Bill Hall paper again.
Conclusion:
It is "Well worth superheating!"
Steam at 9lbs/hour instead of 15lbs per hour for equivalent power is worth a lot when running model engines with small boilers. Something like a 3in boiler instead of a 4in boiler....?
Interesting though, that he mentions he gets almost no Superheat from his Speedy Locomotive design? - And suggests this is due to the large flue tube CSA that encourages a low speed of flue gas, therefore the steam exiting the superheater (I think) must be "cooling" as it gets to the "cooler" end of the boiler, where flue gases have already lost all their heat. This can be a condition also where the boiler is "relatively very long", so the flue gases are almost at boiler temperature as they get to the smokebox end, thus making for an efficient "steam maker" but a poor superheater.
K2

 

[Steamchick]

Thanks N1000. Probably why I have seen a lot of improvement when running with a copper coil carrying exhaust to the water-feed (cold water) tank. In my boats I fit an external copper tube for condensing the exhaust, or an internal condenser (concentric tube type running most of the length of the hull, cooled by pond water flowing through). The condensate pumped back into the water-feed reserve tank is still warm (I pass it beneath the Butane tank to avoid gas pressure loss.). so I can run continuously, instead of a limited 10mins with the available boiler water. But it spoils the effect of all that exhaust steam coming from the smoke stack! There is a small whiffle of water vapour generated steam from the gas burner emitting from the smoke stack - One observer came looking for the "jos-sticks", or "smoke-maker" to see how I made the "smoke"... He walked away saying I was "cheating by using a steam engine in a model Steam tug... The boat club "preferred clean electrics" on their pond..". Envy perhaps?
K2

 

[Steamchick]

On condensing steam inside the cylinder... I do notice when running a 2-stage compound engine that the boiler needs 15~18psi to run the engine when cold, but when the LP cylinder gets hot enough the engine suddenly speeds-up so the burner can be turned down a bit and the engine will run successfully at 10~12psi. Not just the oil getting hot and reducing viscosity, as it is a significant "step-up" in speed and the exhaust has less "condensation droplets" - or so it appears - just a plume of water vapour. I rune with a copper tube oil separator on the exhaust, which fills with condensate while the LP cylinder is warming through, but after that starts to develop power, I notice there is no longer much condensate accumulating in the oil separator. Perhaps this is acting as an exhaust condenser slightly, by reducing the exhaust (back-) pressure seen by the LP cylinder? But maybe not....?
K2

 

[user 52078]

I promised to have a noodle about with Rayger's boiler design. Well here is the result of my noodling.

The original query was whether fewer and larger tubes would affect the steaming rate. I ran a few scenarios through my spreadsheet program and here are some tentative conclusions. Columns 1 - 3 of the spreadsheet for 55 x 7/16", 37 x 1/2" and 31 x 9/16" tubes respectively show a progressive reduction in evaporation rate. All the above cases allow for a 3 mm ligament between tubes.

Notice that the steam generated is 0.0012 m3/s and slightly less for the simpler boilers, but deducting the effect of condensation in the engine according to Hall's test results on a 5" gauge cylinder reduces the useable volume to around 0.00044 m3/s. That would do the job, just about.


I then tried a couple of superheater options:

I first tried 4No. x 1" superheater flues with a 3/8" diameter superheater with a single spear - i.e. "out and back" as far as the firebox tubeplate. The flues could be arranged on a diameter of the tube nest, which would make the superheater headers simple. The number of tubes would reduce to 33 x 7/16". I have only done a very rough layout, and it would probably be feasible to include a couple more tubes with careful design. This option would give somewhat less evaporation (8.6 lb steam/lb coal, as compared to 9.1 lb steam/lb coal for the 55 tube option). However, the steam is superheated to 187 C and the steam volume increases (because of the higher temperature and hence lower density) to 1.28 m3/s with this option, (compared to 1.2 m3/s for the 55 tube option). The useable steam volume allowing for condensation losses increases to 0.00072 m3/s.

I also investigated whether even fewer but larger tubes could be used with the 4 superheaters as above. I found that around 18 tubes of 1/2" diameter could be fitted into the tubeplate in addition to the 4 superheaters. That option gives a further reduction in lb of steam per lb of water, but an increase in superheat temperature to 201 C. Gross steam volume barely changes from the previous option, but net of condensation losses, the useable volume is 0.0008 m3/s.

I have attached a small spreadsheet summarising the options I looked at. If you would like a copy of the spreadsheet, please p.m. me.


Just a comment on Steamchicks thoughts on condensation. It has nothing to do with available enthalpy drop. That stays fairly constant regardless of superheat (a look at an enthalpy / entropy diagram shows how that happens). What does change is how the engine uses the steam. With saturated steam, the incoming steam hits cool ports and cylinder walls and so some steam is condensed to water. As the steam expands, after cutoff, the reducing pressure causes the condensed water to evaporate as it's temperature is now above boiling point at reduced pressure. The boiling process comes too late in the cycle to give useable power, but it does cool the cylinder walls and ports as it passes to exhaust - just ready to cool the incoming steam and the whole process starts again. In effect, some steam is bypassing the expansion by being water for the important bit and hence doing no work.

By using superheated steam, it needs more heat taking out to cause condensation so reducing the quantity lost. In addition, the cylinder walls are made hotter so that condensation is even less likely. The link to Bill Hall's work I gave explains it better than I can.

 

[Steamchick]

Thanks Tenor, I understand better now than before... My earlier appreciation of what was happening was a crude simplification - so, as is often the case - simply "wrong". I was likely comparing "Bananas with Apples"? I.E. a boiler (e.g. one I have) with direct steam, versus when there is superheat added - collecting more heat from a steam pipe taken through a flue to the firebox and out. When running an engine (without changing the cut-off) it runs better with superheat - as you describe. But probably (in my head?) I was comparing 2 scenarios, with the same "steam mass flow" but one containing more energy than the other. Whereas, you keep the input energy constant, and "tune" the boiler (and engine valve cut-off) to maximise the extracted heat as a useful cycle, instead of an inefficient cycle, due to the internal engine pressure/condensation cycle. - Very clever.
Someone said "A man with a little knowledge can be dangerous". - That's me, to a tee!
K2

 

[Steamchick]

Hi Tenor.
Perhaps your Excel model can take a 10kW gas burner for comparison, with just a 5in diameter flat circle of radiant heat (a simplification) radiating heat at 900deg.C? - If necessary (to model it like a coal fire) it could be considered to be a thin radiant fire at 3/16" thick. - with an average temperature around 800deg.C, which is more like real ceramic burners (Not over heated ones!), so only the hottest parts reach 900deg.C? (This is the "weakness" of a ceramic burner compared to Coal firing!). The remainder of the heat is carried away as hot (fully burnt) exhaust gas leaving the burner around 900deg.C. I suggest Propane is the fuel gas for this model? A "loss" with a simple ceramic burner is that the radiant heat that shines "straight through" the flue tubes is lost to the lid above the steam generator. Perhaps in reality a double coil superheater capturing the radiant heat could be utilised? (maybe I should do a drawing as my ideas develop?).
Using a "Wire" matrix the matrix would be at 1150deg.C and could be considered as a dome, say 5in diameter at the base in this (Rayger's/Earnest Winter's) boiler, with a dome raised maybe 1 1/2in above the base line? Again a total of 10kW "gas combustion power"...
Don't bother if very time consuming, I am curious - like the cat - that's all.
K2

 

[Steamchick]

Hi again Tenor. In your spreadsheet summary, you mention "Gate Loading" 0.0275. - Should that read "Grate loading"- for the coal fire? - I really don't know what this means... - is it a ratio of air-gap to total grate area? - I have missed it somewhere along the thread.
I am not sure about the "Unburnt fuel lost" either. Listed as 10 - BAR? - I expected "percent", for losses of small fuel dropping through to the grate, smoke and non-burnt fuel going up the flues and out of the chimney, etc.?
Also Dryness factor? = 98Pa? - is that Pascals? - I expected "percent"?

Here is a large ceramic burner I made for an unforced 9" diameter boiler, running on Propane was ~4.4kW. May seem "Odd" that I fitted a rectangular ceramic in a 9in circle, but this was only about 15% less power, and the cost was 2 ceramic plates instead of 4 cut (to fill the sides) with a lot of wastage. The Customer did not reply to say how well it performed... so maybe he wasn't happy?

Also a development wire wool burner ~3kW (~3in dia).

And a "wire-wool" Propane burner, around 8kW,

but needs to be completely remade as I now have more (better) wire wool, and the inner "can" needs restricting (smaller holes) to prevent flashback. I was making this to understand the design parameters "a bit better". Combustion "happens" within the fibre matrix, outside of an inner gas distributing "can". A version half this height could power Rayger's boiler, but maybe limited to 4 or 5kW - with my "technology" level.... Above that would need a Commercial (Bekaert?) burner. - Which costs less than my materials as they bulk-buy the wire matrix, etc..
Cheers,
K2

 

[user 52078]

Hi again Tenor. In your spreadsheet summary, you mention "Gate Loading" 0.0275. - Should that read "Grate loading"- for the coal fire? - I really don't know what this means... - is it a ratio of air-gap to total grate area? - I have missed it somewhere along the thread.
I am not sure about the "Unburnt fuel lost" either. Listed as 10 - BAR? - I expected "percent", for losses of small fuel dropping through to the grate, smoke and non-burnt fuel going up the flues and out of the chimney, etc.?
Also Dryness factor? = 98Pa? - is that Pascals? - I expected "percent"?

Here is a large ceramic burner I made for an unforced 9" diameter boiler, running on Propane was ~4.4kW. May seem "Odd" that I fitted a rectangular ceramic in a 9in circle, but this was only about 15% less power, and the cost was 2 ceramic plates instead of 4 cut (to fill the sides) with a lot of wastage. The Customer did not reply to say how well it performed... so maybe he wasn't happy?
View attachment 151307
Also a development wire wool burner ~3kW (~3in dia).
View attachment 151305
And a "wire-wool" Propane burner, around 8kW,
View attachment 151306
but needs to be completely remade as I now have more (better) wire wool, and the inner "can" needs restricting (smaller holes) to prevent flashback. I was making this to understand the design parameters "a bit better". Combustion "happens" within the fibre matrix, outside of an inner gas distributing "can". A version half this height could power Rayger's boiler, but maybe limited to 4 or 5kW - with my "technology" level.... Above that would need a Commercial (Bekaert?) burner. - Which costs less than my materials as they bulk-buy the wire matrix, etc..
Cheers,
K2

Thanks Steamchick,
I have truly screwed the dimensions which were copied and pasted over in a hurry. And yes Gate should read grate! I'll correct and re post.
Your radiating gas burners are certainly impressive. Without running numbers, they ought to be a better retrofit to a coal burner than the non radiating marty or wok units.
Martin

 

[Steamchick]

Hi tenor. Hope you are right about the burners...
I didn't mean to criticise your spreadsheet. The numbers look meaningful, but I didn't understand the units presented. I understand how they could have been "mis-copied". I do that often enough myself!
I usually make the mistakes that confuse others, not the other way around... Glad to know it's just a human thing, and I'm not getting too old....
K2

 

[Steamchick]

Another radiant burner for a 40mm dia Firetube (4in dia marine boiler in the making). A simple blowlamp fired into a tube of wire mesh...

It changes the flame and hot zone when inside the fire tube - which exhausts via 6 smaller flue tubes... as -

K2

 

[user 52078]

Right folks, found the problem with the summary spreadsheet. I should have copied and pasted SPECIAL, not just pasted. It picked up all manner of other garbage for the units from elsewhere. So attached herewith is a corrected version (plus my real human spellign mistakes sorted out).

Grate loading is the mass of fuel put on unit area in unit time. It is a measure of how hard the fire is working. I have chosen quite a conservative value - about half what a 5" gauge loco might see but stationary (or marine) boilers usually work a lot less hard than locomotive boilers.

You certainly seem to be the go to guy for gas burners, Steamchick! I will try and have a noodle around to quantify the value in radiant burners. I have some ideas for at least a roughcut way.

Nerd1000,
You are right that a steam engine struggles for sufficient expansion raio. About the best you can do in a conventional slide or piston valve engine is a bit over 2.5 times (based on 20% cut off). You can push it a little further with poppett valves and cam valve gear. After that, it is down to compounding or triples and more! That said, it does not change the fact that superheating is usually a very good idea, even with condensing where S/H can help prevent condensation in the later stages as the working fluid has heat taken out to produce work, and hence cools down. Just don't overcook the superheating on a condensing duty or the condenser has to work extra hard.

Enough rambling for one night, it has been a long day,
Martin

 

[Steamchick]

Hi Tenor,
Thanks. That makes a lot of sense now that I understand the units!
Amazing how a simple table summarising the results of the various integrated calculations can tell a story...
As your model is built around a "loco-style" boiler, it includes for Superheaters in flue tubes. Although I have bench boilers with the outlet steam pipe fed down the main flue tube and back up to the top, I also have configurations where the main steam pipe is fed down the main flue tube, then a length (sometimes coiled) in the fire-box, before exiting (VERY HOT! - Cooks cotton lagging!) to the engine. I also have boilers with a coil (More dryer than superheater?) in the smoke box, after the boiler flue tubes, but before the Chimney.
Does your "Excel model" allow for such variations of "superheater" position?
In the case of Rayger's current boiler design, it would be relatively easy to have a superheater coil (2 or 3 turns?) inside the smokebox above the flue tubes, before the chimney. The exit temp from firetubes is predicted at 208C. So the superheat could be approaching that temperature, perhaps? Maybe raising the 146C steam temp to 175C - or something?
Considering the total length of Outlet pipe, if coiled in the smoke box and then out and across "whatever space" to the engine, I presume the max length of this pipe - From Steam take-off to engine valve chest - needs to follow similar rules to the flue tubes, I.E. Not more than "80 (bore) diameters" of the pipe? (or whatever?). - I fear I have too much length for the bore diameter of some of my boiler to engine connections.
Thanks,
K2

 

[Steamchick]

Raygers: If I were to suggest a Burner for Propane firing....

1. Option 1. A 4 3/4in Ceramic circle burner. (Dia 120mm.). Gas jetted for about 4.5kW? Considered as a radiant plate 1mm thick at 900C max temp, the rest of the heat as Hot burnt gas, and just the right amount of air... (No secondary air supplied for post-burner combustion). - Cost of ceramics £15~£30?
   
   
   
   
2. Option 2: An hexagonal Ceramic burner, assembled such that the top ceramic hexagon is 45mm across the flats, and side ceramic surfaces (8 off) are about 52mm x 45mm wide (combustion surface). This could perhaps take gas for ~5~5.5kW? - Cost of ceramics £20~£40? (Maybe I should make one?)
3. Option 3: A drum shaped burner, 65mm high (2 1/2") x 90mm dia. (3 1/2"). This made of wire mesh would have a higher gas power of maybe 5kW~6kW? Considered as an area of radiant heat equal to the surface area of the burner, x 4mm thick, with the remaining heat dissipated as hot exhaust from full combustion of the gas. - I guess a commercial burner (like this one from Bekaert?) - could cost ~£75??? - But I really don't know?

Hope this tickles the brain? - Prompts ideas...?
K2

 

[Steamchick]

Forgot to mention... For "commercial" burners, you would need to design and make a gas-jet/air-intake/Venturi and mixing tube, and a pressure-levelling plenum to go beneath the burner itself. (I can draw a design, based on a 4in design I have made).
I just wonder what Tenor's calculations make of these options in terms of useful Steam production? - Perhaps the coal fire is simply "more powerful"?
K2

 

[Raygers]

"And a "wire-wool" Propane burner, around 8kW,"

As soon as I saw this burner of yours I thought this is the one. If I stay with the sized firebox door I am limited as to the size of burner. However, if I wanted to run a burner I could remove the firebox plate and come into the chamber from beneath.
A simple stand to elevate the burner to the correct height should work quite well.

 

[Raygers]

There's a lot of information from both Steamchick and Tenor and Charles Lamont, some of it beyond my ken, I have to read it several times to let it sink in.
Thank you all for taking the time to impart some of your knowledge and experience to me, I really appreciate it.

 

[Steamchick]

Raygers, if you build the boiler for Coal, with a firebox hole and door, it is great fun (for many) to run a "proper fire". The firebox hole is also very convenient for lighting a Gas burner.
Looking at your existing design, the "wire-matrix" burner could occupy the space in the firebox with 10~20mm clearance at the sides, and perhaps 30~50mmclearance above the top of the burner to the Top firebox/flue-tube plate. - This space for residual combustion to take place, and allowing space for gases to expand a bit so the burner isn't choked. But that is my "unproven" idea...

Thanks for your kind comment about our offerings... Tenor and Charles have helped me a lot in understanding "boilers" with their expertise!
K2

 

[user 52078]

A few points arising. I interrogated my computer model for Raygers boiler. The coal fire is actually emitting 8.5 Kw, as I have set it up. The air ratio is about 20% excess air, so in that respect quite comparable to a properly set up gas burner. The estimated firebed temperature is 1220 C. I am sure Steamchick can do something with that data, but I am pretty sure gas firing would be feasible if we are all denied access to coal.

I am always a bit hesitant about smokebox mounted superheaters. It can be done, but needs quite a lot of tubing - not just 2 or 3 coils. I attach a photo looking up at the smokebox superheaters in my steam lorry and there are 3 or 4 layers of these coils - around 30 feet of pipe in there to get a calculated steam temperature of 300 C. OK, it's a big scale model, but I hope it makes the point about how much tubing you will need to get real superheat.

Steamchick is right that my computer model is primarily for locomotive type boilers, but I have written a vertical boiler option for my own lorry. That fairly rough code is not really stable enough for public consumption though. I have not yet coded up the option of firebox mounted superheaters, but I am pretty sure it could be done.

Steamchick questioned whether there was any limit on pipe length between boiler and engine. No, not really it is a different consideration to a boiler firetube. In a firetube, after 80 to 100 diameters of length, most of the useable heat from the flue gas has been transferred into the boiler, so more length is really just a waste of material. Provided you have good insulation, the steam supply pipe could be a lot longer. In large process plants you would not be bothered by a couple of miles of steam supply pipe - but very thick insulation and condensate drains at regular intervals.

Raygers, You are welcome to my ramblings. Your OP was a question I set out to answer around 10 years ago now. I was very dissatisfied with what others had published about the thermodynamic design of model steam boilers, so set out to do a proper job based on industry accepted heat transfer formulae. If there is anything that needs further explanation, I am happy to reply either on here or as a PM. Just for curiosity, what are you proposing to to run with the boiler?

Martin