All the joint areas had been well cleaned before assembly but the assembly was given a final dunk in the pickle bath just to make sure. One of the biggest problems in silver soldering is caused by the surfaces not being spotlessly clean and all joints should be given a final pickle just before the soldering operation. The joints should then be fluxed and soldered as soon as possible as the clean copper will soon oxidise again.
I used Silverflo 24 for this joint which is a high temperature silver solder. This means that all other joints can be made with lower temperature solder with no risk of disturbing this joint. Actually I find Silverflo 24 to be a bit of a pig to use with ordinary propane equipment as it takes a lot of heat to get to the right temperature. It's probably better suited to oxy-acetalyne equipment. There is Silverflo 40 (or the equivalent ) that melts at a much lower temperature than the 24 but still higher than Easyflo so I reckon that could be used instead and would be much easier to use. With care, a complete boiler could be soldered with just easyflo but there is always the risk of disturbing previously soldered joints.
As it happens I had to have two goes at the barrel joint as the solder did not fully penetrate the barrel/wrapper joint all the way around. The overlap is actually quite large at 3/8"and much bigger than the recommended overlap ( 3 to 4 times the material thickness i.e. 3/16" to 1/4" ) so I'm not suprised that the solder did not go all the way through first time. Plenty of solder was used on the throatplate joint to ensure a sound job.
Sunday the 6th saw a trip to the National 2.5" Gauge rally at Sutton Coldfield and another outing for Helen. It was a gorgeous day with a good attendance and most enjoyable (the two pints in the pub just down the road weren't bad either!) There's some photos of the event in the photo section.
Next I needed a firebox. The former plate for the tubeplate and the back plate was again made from 1/8" steel screwed to a bit of plywood. The plate was marked out for the tubes and pilot holes drilled so that the plate could later be used as a jig for drilling the tubeplate. Both plates are made from 16swg copper as this will be plenty strong enough in these sizes. In fact it has been suggested that all the boiler plates need only to be as thick as the barrel etc so long as they are properly stayed and fitted with proper bushes for fittings etc. It may be that LBSC started the trend of using thicker material for the backhead and boiler tubeplate because he originally screwed his fittings straight into the copper without using bushes. Obviously the copper needed to be thicker to take a decent thread. Another advantage of using 16swg rather than 13swg is that the radius of the top corners is quite small and the thinner material will be easier to flange over here.
Firebox plates with their former
An interesting suggestion by Peter Rich in his recent boiler making articles in Model Engineer is to make the actual firehole ring by flanging the firebox rear plate outwards using a suitable former, rather than use a seperate tube soldered in. I can see advantages in doing this but I think it is better suited to larger boilers where there is more material to play with. Also the copper will probably be thinned a little by having to be stretched to form the ring so would be better used with 13swg material. One big advantage I can see is that the opening in the firebox can be made larger than the opening in the backhead as often done in full-size locos. The resulting conical shaped firehole ring would give a much better view of the fire than the normal straight ring. Erik-Jan Stroetinga achieved the same result on his Didcot by turning a conical firehole ring out of solid copper! ( see links page ).
The tubeplate carries eight 3/8" fire tubes and two 9/16" superheater tubes. The original Helen boiler had seven firetubes but by a bit of re-design I managed to fit another one in. There's no point in trying to cram in too many tubes though as the circulation around them will be poor. In any case, most of the heating is done by the firebox surface and the tubes only really act as pre-heaters for the water in the barrel. The holes for the 3/8" tubes were drilled in the drill press by clamping the steel former to the copper tubeplate and then drilling through both at the same time. This increased the thickness that the drill had to cut through and helped reduce the tendency for the drill to cut triangular holes as often happens when drilling thin sheet. The holes were left slightly undersize ( 23/64" ) so that they could be reamed to final size. The holes for the superheater flues were finished to size by boring them in the lathe as the largest drill I have is only 1/2". They were also left undersize to allow for reaming.
Boring the superheater flue holes to size
The tubeplate with all holes drilled
The firehole ring is a length of 1-1/8" dia x 16swg copper tube with a step turned down at each end to fit into the firebox back plate and the backhead. The hole in the firebox back plate was drilled and bored out in the lathe in similar fashion to the tube plate.
Boring the back plate to take the fire hole ring
Back plate and ring
The ring was then fitted through the hole and lightly rivetted over to hold it in place. The joint was then silver soldered on the inside of the back plate with silverflo 24. The outside will be soldered when the back plate is soldered into the wrapper. I noticed that when doing this job on my brothers Rob Roy boiler, the solder did not run through the joint when the ring was soldered in from the outside only. This is probably because the rivetting over of the ring closes up any gap in the joint and prevents the solder penetrating properly.
The wrapper is bent up from 16swg sheet and was a bit tricky to get the right shape. The sharp bends were formed over a piece of bar and the rest by hand using the tubeplate as a former. At this point I left the bottom edge of the wrapper straight and it will be trimmed to final size and shape later. As done with the outer wrapper, the firebox wrapper was first fastened to the plates using 10BA nuts and bolts and then these were replaced one by one with 1/16" copper rivets.
Initial assembly using 10BA bolts
After rivetting the whole assembly was pickled and then soldered with silverflo 24.
Firebox after soldering in end plates
It's now apparent just how big the firebox actually is ( 6-3/8" long ! ). As LBSC said in the original Helen articles - Poor fireman!
The firebox is fitted with the usual 'Z' type of girder roof stays made from 16swg sheet but I am also adding an extra angle piece at the bottom of each stay to give a bit more support to the firebox crown. There seems to be several schools of thought on the effectiveness of girder stays and Martin Evans designed several boilers where the stays were not connected to the outer firebox wrapper at all ( Simplex, Nigel Minor, etc ) and these seem to be perfectly satisfactory in service. The theory is that the stays merely strengthen the top of the firebox to stop it sagging and the firebox itself is supported through the foundation ring and the side stays. A lot of full size traction engine boilers and similar were designed like this. I think the main problem with full height girder stays is that they tend to restrict the water circulation over the firebox crown. However, it is suggested that the full height stays increase the heating surface of the firebox and I think I will need all the heat I can get!
The stays have four 5/8" diameter holes in them to help circulation and I punched these out using a chassis punch as this was easier than drilling. The stays were then attached to the top of the firebox with a few 1/16" rivets and then silver soldered. This time I used 440 from Cup Alloys which is the equivalent to silverflo 40. This melts at about 100 degrees C lower than the silverflo 24 and is much easier to use. The melting point is still higher than easyflo so will not melt when the final soldering is done.
Crown stays silver soldered in place
I was going to tackle the tubes next but when I looked through my stock of copper tubing I realised that I didn't have any 9/16" diameter for the superheater flues, only 5/8". I think I must have made a mistake when I ordered the boiler materials so these will have to wait. Live Steam Services are only 15 minutes drive away so I'll get some from them. So I decided to make the smokebox tubeplate instead. I had a former to hand that I had turned from a steel blank for the Rob Roy boiler which has the same size barrel. The barrel for the A1 boiler is also 3-1/4" diameter so it will do for that as well!
The tubeplate was formed from 3/32" thick copper as that is the thickness that the former was made for and after flanging the plate the outside diameter was turned down to a slightly loose fit in the barrel. This was down by holding the steel former in the chuck, putting the tubeplate on the former and clamping it against the former with a disc of wood pressed against the tubeplate by a revolving centre. Provided that light cuts are taken this method will give sufficient grip for the turning operation. The ragged edge of the tubeplate flange was also trimmed up at the same time.
With regard to the fits for silver soldered items, in the past authors such as Martin Evans always specified tight fits for tubes in tubeplates etc but it is now accepted that there should be a small gap to allow the solder to penetrate the joint by capillary action ( 0.01mm to 0.25mm depending on the solder). This gives the maximum strength of joint. It was always recommended that the tubes should be a tight fit in the tubeplate and then you were told to file shallow nicks in the sides of the tubeplate holes to allow the solder to penetrate to the other side. What often happened was that you finished up with nice fillets either side of the tubeplate but no solder between the tube and the tubeplate inside the hole itself leading to a weak joint.
Turning the tubeplate to fit the barrel
Next the holes for the tubes were drilled using the firebox tubeplate former as a jig and the holes for the superheater tubes bored out in the lathe as for the firebox tubeplate.
Drilling the tubeplate
It occurred to me as I was doing this that the bottom tube is going to finish up very near the bottom of the smokebox and may well get blocked by ash building up. We'll have to wait and see!
I had a trip to Live Steam Services in Little Eaton this morning and picked up a length of 9/16" diameter copper tube for the two superheater tubes so the next job was the tubes. They were cut roughly to length and then the ends trimmed up in the lathe to bring the tubes to the right length. A light skim was then taken off one end of each to form a slight step to stop the tubes falling through the holes in the firebox tubeplate. The holes in the tubeplate were then reamed out until the ends of the tubes were an easy fit in the holes, again to leave a slight gap for the solder. The holes in the smokebox tubeplate were reamed out to full diameter to take the other ends of the tubes. The ends of the tubes were cleaned up with carborundum paper and then dumped in the pickle bath for 10 minutes for a final clean. In the meantime I made some rings of 1.5mm Easyflo solder to go over each ring and sit on the firebox tubeplate. This is much easier than trying to feed solder into the joints from outside.
After pickling, the tubes were fitted back into the tubeplate, the rings of solder slipped onto the tubes, and the smokebox tubeplate fitted to hold the tubes in alignment. Normal proceedure is to then go ahead and solder the tubes in relying on the fit of the smokebox tubeplate to hold the tubes rigidly. However, because of the slight clearances I had allowed in the joints, the assembly was far from rigid and the tubes moved around very easily so I took a leaf out of my brothers book. For this operation on his Rob Roy he'd made up a simple jig using a piece of steel clamped to the side of the firebox to which the smokebox tubeplate was also clamped. This made the tube assembly extremely rigid with no chance of anything moving during soldering.
Using this idea, I clamped a length of steel angle to one of the firebox crown stays, clamped a block of steel to that, and finally clamped the smokebox tubeplate to the block of steel. I then adjusted the position of the smokebox tubeplate until the tubes were all vertical and not twisted.
Simple jig for holding the tube assembly rigid during soldering
Plenty of flux was then slopped onto the tubes and tubeplate and the whole lot heated until the solder melted and ran into the joints. Inspection after pickling showed that on one tube the solder had not quite ran all the way around the inside of the tubeplate so the job was refluxed and reheated, concentrating on that one tube. Pickling again showed that the joint was now perfect. Better safe than sorry!
Tubes straight after soldering ( note improvised brazing hearth!)
After a good soak in the pickle bath and a clean, it's beginning to look like a boiler! Next job is the bushes for all the fittings etc.
Previous Page Next Page