How to heat a large boiler for soldering
|Bob Worsley||25/10/2020 10:36:08|
|59 forum posts|
Another 'why do we do it this way' posts to antagonise and irritate the members of this forum.
Looking at something like a 2" scale traction engine, or probably a 5" loco.
These boilers are large and heavy, and require enormous amounts of heat to reliably silver solder, so need to heat the whole thing to 700C. The whole thing ends up at 700C radiating that temperature.
The problem seems to be that the gas flame is used to heat the boiler, but once the flame has impinged on the boiler it is then redundant. This redundant flame is still at 2000C so is going on to heat up everything around other than the boiler. Talking to people and stories of setting fire to the shed roof, or anything else, are not rare.
What is needed is a heat source that is in the copper. What I suggest is resistance heating using a normal arc welder.
The advantages are that there are no flames, and the whole boiler can be closely insulated to keep the heat in the metal.
Will it work? Copper has a specific heat capacity of 0.38J/g K. So to raise a 1kg boiler by 600C takes 0.38x1000gx600C = 228kJ. 1J = 1W/s so this is 3.8kW for 60s, or one minute.
A real boiler would be 20kg and 700C = 5.32MJ or 5kW for 18 min. This of course assumes no heat loss.
You would need a decent sized, or more than one, welder and because it is slow then need to run them at their 100% rating. Looking through Machine Mart the concept of continuous duty rating seems to have gone. You would also need to use one of the old transformer based welders, no silly electronics saying to turn off.
How to use them? On the barrel to firebox ring joint would need a clamp around the barrel for one lead and two clamps, one either side of the wrapper for the other lead. The current flow will then be all through the ring joint and down the throatplate. Boiler will need moving to do the top and bottom, but need to do that with gas anyway. Just also need to shift the insulating cover first.
The power leads will get hot, 700C, so not your average rubber coating, needs to be glass cloth or something similar. Because the welder is only 30V or so there is no electrical hazard, so bare wire might be ok.
It is also feasible to use some high power resistance wire or resistors under the boiler as background heat. This could be used with a gas torch as well.
Drawbacks? Will be a couple of large cables, taking 200A at 5kW, so quite stiff. Will need some heavy duty clamps for connecting, with AC looking at about 10-15A sq mm connecting area. Clamps possibly need to be connecting to a sheet rather than around the boiler with thermal expansion.
An alternative is induction heating, would be even better but a 5kW induction heater is probably rather expensive. The old transformer based welders are two a penny at sales.
The normal welder used would be AC, so to reduce losses the power cables need to be tightly twisted together, also minimum length. If DC then these worried don't apply. Also modify the welder to fit a number of old computer fans to cool the transformer.
6452 forum posts
Only that resistance welding copper is challenging because it's the second best electrical conductor in the world! A boiler will have lower electrical resistance than most cables and all small welder transformers, so most of the heat will be be lost before it reaches the boiler. Even replacing the cables and rewiring the transformer with Silver won't help much, though super-conductor technology might do the trick.
Should work on a steel boiler though.
|Clive Brown 1||25/10/2020 11:40:01|
|537 forum posts|
Dunno if an electrical method of heating copper boilers would work, but I suspect it would be quite complex and probably require a hefty electrical supply circuit. OTOH, I can siver solder a boiler with relatively low cost gas equipment which is useful for all sorts of other tasks and is easy to move and store. Might be wasteful of heat for the very big jobs, but I'll put up with that.
|Pete White||25/10/2020 12:12:28|
|121 forum posts|
The wire would be a problem, certainly where it leaves the transfomer core to the leads? My oil filled Oxrord say at 150 amps 50 volts will run all day, but the heat output is on the end of the electrode. ?
5567 forum posts
"so need to heat the whole thing to 700C. "
Well no, actually. You do a section at a time.
Edited By Bazyle on 25/10/2020 12:33:42
|Martin Kyte||25/10/2020 12:42:40|
2124 forum posts
Induction heating maybe?
Personally I don't see the point.
|Sam Longley 1||25/10/2020 13:14:33|
|800 forum posts|
I am going to build an open coal forge for other jobs. Could one lay the copper over the heat source on bars or a cast iron grill, or even in the forge to maintain the necessary temperature? One might build the fire wall up above that to retain lots of the heat
Is that a possibility?
Edited By Sam Longley 1 on 25/10/2020 13:15:40
|Clive Brown 1||25/10/2020 13:21:48|
|537 forum posts|
In the days of LBSC and the fearsome "Five Pint Blowlamp" the usual insulation material for soldering a boiler seemed to have been coke. Presumably this would burn and add considerably to the heat input to the copper.
|J Hancock||25/10/2020 14:28:11|
|479 forum posts|
Sounds more like something Nikola Tesla used to do , I would 'phone your local area electricity board so they can arrange for a bit more to be put on the bars before you switch on.
No, please , don't do it.
|Andy Gray 3||25/10/2020 15:35:04|
|44 forum posts||
Work out the resistance of your potential boiler combo, and then use the P=I^2 x R formula to figure out how many amps you'd really need to get 5kW dissipation.
It's likely to be 2 orders of magnitude higher than you've guessed
Making connections for this sort of thing is also non-trivial. (I've worked with similar technology - One needs to ensure that the current is delivered uniformly, as resistivity tends to increase with temperature, so any non-uniformity results in localised hot-spots which progress very quickly to a puddle of molten metal and a gap where it used to be.)
|Andrew Johnston||25/10/2020 15:47:44|
5726 forum posts
Copper has an appalling resistivity temperature coefficient, about 0.39% per degree C.
As alluded to by Andy those pesky electrons will follow the path of least resistance whatever one does. So one would need to be very careful with the positioning of the clamps to ensure that the area that needs heating is in fact the area that does get heated.
|Ed Dinning 1||25/10/2020 21:48:23|
|27 forum posts|
Hi Folks, on a similar vein we needed to heat treat Ti at about 350C with a gas oven . As these were large pieces it was a large expensive oven and the Ti needed to be sheilded so that TiN would no form on the surface. This was a time consuming process, so we came up with the idea of passing a current through the Ti sheet; as Ti is quite resistive we needed about 3 to 5 volts at 300A.
An old high current mains transformer had its secondary rewound with 1.5 * .25 " Cu bar and connected to the Ti sheet by welding clamps. Heating under ceramic insulation took less than 10 mins with negligible TiN formation and aslo reduced the energy bill
5567 forum posts
Correct. One of the sets of 'words and music' by LBSC describes cutting a circular hole in the hearth to put the boiler barrel through and surround with burning coke. Remember to use coke not coal or any kind of compressed powder boiler nut or charcoal.
|Bob Worsley||27/10/2020 09:50:20|
|59 forum posts|
Not too enthusiastic response, seems no one has ever tried it other than Ed, thanks.
Yes, there will be lots of amps needed, but was just thinking of kW input to see what the figures looked like. Change welding transformer to spot welding transformer. Or just strip the secondary off and replace with just a couple of turns or so. The important thing is the kVA rating of the laminations to carry the power continually. There is a low power requirement, easily managed by a domestic 100A supply, the shower takes far more. It is this low power requirement compared to a gas torch that is the attraction.
With the boiler almost totally shrouded with insulation and with the excellent heat conduction of copper I would expect the whole boiler to be at a similar temperature, possibly only 50C from one end to the other. This solves many of the temperature coefficient problems with connecting the power, but it will need multiple points with 2000A.
I have lots of high power resistors, possibly use them as a hot bed to indirectly heat the boiler but getting back to the gas torch problem. Has anyone used these gas radiant heaters as typically sold by Machine Mart to act as a hot bed? There is a 10kW one, couple of those with thermal shrouding might work.
What did surprise me when I looked in RS was the absence of any high temperature cables, insulation and similar, about 350C maximum, far too low. The idea of using the ceramic cases from fuses has been duly filched from another forum.
Charcoal was recommended as a bedding material in the fire as being clean burning?
Also one of the boiler books said dropping steel clamps in the pickle is fine. Remember being warned never to do that, meant the pickle needed to be changed?
|Nick Clarke 3||27/10/2020 11:10:09|
937 forum posts
Heat is heat whether it comes from electricity or gas.
The largest Sievert burner which I have read several times in mags and books is necessary for a 5" boiler, sometimes together with a second torch, is 86kW.
The voltage round here is closer to 230V - it hasn't been 240V for years.
At 230V the electrical current needed to provide this power is about 375A which is way beyond the capacity of house wiring. Your electrical shower at 8kW max is less than 35A
IMHO not practical I'm afraid.
|Andrew Johnston||27/10/2020 11:16:08|
5726 forum posts
That's interesting, my plug in mains monitor currently (!) says the voltage is 243.8VAC.
|Bob Worsley||27/10/2020 11:17:41|
|59 forum posts|
You are making the mistake of equating the heat from the gas torch to what is needed to heat the boiler. My calculations above, assuming zero losses, suggest a vastly lower heat requirement. The problem with gas is that 99%, or so, of the heat is basically wasted by the flame only being used to heat the boiler for a second or so. It is then blown around the brazing hearth heating everything else. This is why oxy acetylene is so useful, the flame is only 2kW or so, but it is used where it is needed.
Just been reading about induction heating, and this does seem a far more sensible route. Can buy 2kW heaters for £2k, and the heat generated is within the metal, not blowing around your workshop.
Anyone any practical experience of induction heating? People like Sealey seem to sell self contained heaters for taking oxygen sensors out of exhaust pipes.
|Robert Atkinson 2||27/10/2020 11:54:16|
820 forum posts
As Andy and Andrew have said, the problem is coppers low electrical resistance. In practical terms this means the leads and the connections to the boiler need to huge. If the lead resistance is the same as the boiler resistance , ie same thicknes and length of copper (or thicker and longer) the same amount of het will be produced in them as the boiler. Resistive heating is OK for resistive materials like steel and and titanium, not copper.
|Nick Clarke 3||27/10/2020 12:08:39|
937 forum posts
UK Mains Voltage since 2010 has been 230V +/-10% ie 207V to 253V so your supply, while within tolerance, is towards the higher side. Prior to 2010 it was still 230V but with different tolerances.
The reason is that our 230V supply can then be cross connected to both 220V and 240V systems across Europe. See BS 7697
19123 forum posts
Just worked it out for my 2" Fowler boiler, to get the same area as the CSA of the barrel the leads would need to be 1100mm2, that's 39mm dia if solid, best go see what you can find laying about beside the local railway.
Or you could buy a lot of gas for the cost of the copper leads.
Edited By JasonB on 27/10/2020 12:13:41
Please login to post a reply.
Want the latest issue of Model Engineer or Model Engineers' Workshop? Use our magazine locator links to find your nearest stockist!
You can contact us by phone, mail or email about the magazines including becoming a contributor, submitting reader's letters or making queries about articles. You can also get in touch about this website, advertising or other general issues.
Click THIS LINK for full contact details.
For subscription issues please see THIS LINK.