5" gauge Britannia (a builder's diary)

[simplyloco]

Jun 1, 2020 8:49:01 GMT springcrocus said:

I've looked around for a casting for the exhaust injector but there only seems to be one on offer, a white-metal casting for about £38 but it looks a bit squashed to me. So I've started to sculpt one out of some old brass offcuts.

SNIP
Sorry, bit of a long one, that post.

Steve

Very realistic Steve. Mind you, I would probably have filed it out from the solid...
John

[springcrocus]

Jun 1, 2020 9:12:39 GMT simplyloco said:

Jun 1, 2020 8:49:01 GMT springcrocus said:

I've looked around for a casting for the exhaust injector but there only seems to be one on offer, a white-metal casting for about £38 but it looks a bit squashed to me. So I've started to sculpt one out of some old brass offcuts.

SNIP
Sorry, bit of a long one, that post.

Steve

Very realistic Steve. Mind you, I would probably have filed it out from the solid...
John

Probably quicker, too! However, for me a file is a tool of last resort.

Regards, Steve

[barlowworks]

It looks pretty spot on to me. I hope these 2 shots are useful, let me know if you need any other detail shots, I have quite a few.

Mike

IMG_0934 by Michael Cawthorne, on Flickr

116_9029 by Michael Cawthorne, on Flickr

[92220]

Hi Steve.

If you are referring to the Hewson white metal casting of the exhaust injector, the casting is actually fairly close to drawing. I bought one because, at the time, I didn't have a BR drawing. Now I do. Having said that, your injector will be your own work(!!) and it looks superb! I am watching this with great interest as, although I have the Hewson casting, I am still aiming to machine up my own in brass or bronze. Well done. It is great to see someone interested enough in the appearance of their loco to make a scale injector body rather than just fit 2 similar commercial injectors.

Bob.

P.S. Like you, I machine everything, and regard using a file a 'last resort'!!

[springcrocus]

Bob, you are far more meticulous than I am, your version will be much closer to the prototype.

The platform for the exhaust injector was made from 16 swg brass and shaped to suit the angle of the ashpan. A pair of holes were drilled, the idea being to use a pair of 8BA screws with spacers. Also in the picture is a piece of brass, about 3/16" square, that screws to the back of the assembly to simulate the live steam inlet.



It's a good job I managed to get the two flanges in line, drilling the two 8BA fixing holes in the bottom of the barrel was easier than expected.



A slim triangle of brass was made to form the web in the top of the injector ...



... and the platform had it's cutaway made for the overflow pipe. The small piece of brass angle and the flat offcut bolted at each end are the mounting feet. The little brass rectangle is the overflow flange and the turned piece of 1/4" hex to the right is the exhaust steam valve nut on top of the injector.



This is a view from the back showing how the two parts are screwed on. None of this will be seen once all is assembled.



And this is the upper web soldered into place, a bit scruffy but needs filing down somewhat. I haven't made the small front web, that's a detail too far.



With all the screws in place, and ready to fit to the ashpan, it became obvious that the platform was never going to hold the injector in the correct position. The designer had made a complete pig's ear of placement, as seen when comparing the drawing to the photo.



The injector sits below the ashpan, not above the baseline as drawn, so a new bracket had to be made. This entailed a bracket folded in the opposite direction with fixing screws above the injector. This is the new platform with an acute-angled bend.



And this is the injector temporarily mounted on the ashpan. It needs tweaking slightly but is pretty-much in the correct position. The pipes, once made, will be a light push fit into the recesses except for the exhaust steam pipe which will have it's own flange and be bolted on.



It will now be painted and, hopefully, look a lot better when the paint softens the lines. Oh, and I've forgotten to fit the overflow pipe, too.

Steve

[springcrocus]

There we go, it's amazing what one can hide under a coat of paint.

Couple of holes in the large flange to fill before the topcoat.

Steve

[springcrocus]

There are two injector steam valves on Britannia and I have made a pair from some Pumphouse bronze offcuts that I had. There will be a pair in here somewhere although I have another piece in the scrap bin and will make an extra pair at the same time.



Rather than waste this material, I set up each operation with brass bar first, then followed with the bronze. The first thing to make was a tool to form the neck radius of the valve and for this, I chose to use the back end of a 1/2" diameter end mill. The drawing is wrong (again) where it specifies 1/2" radius, it should be 1/4" radius (1/2" diameter).



As can be seen, the tool gets mounted upside down in the toolpost holder, the angle being enough to give a bit of clearance without losing too much of the form.



The holder is set high in the toolpost and because I want to finish at 1/4" diameter, I set the tool just a bit less than 1/8" above the centreline of the work.



This is another shot showing the tool in action. It's also a good example of why setting tools exactly on the centreline of the work is a nonsense. I've never used a tool-height setting gauge in my life, always just eyeballed the height and moved it if it wasn't cutting quite right.



So here is the fully-formed curve of the valve neck ready to be threaded.



To get the outer form of the globe I used a HSS tool already in the toolbox. The diameter is nearer 7/16" than 1/2" but it's close enough for this. The 1/4" x 40 tpi thread has already been cut with a button die. A bit of emery cloth helped to blend things together.



Before they were parted off at an inch long, a centre drill with a ground-back pilot was used to create the front form for the nipple. Here are the first two, one with a nut and nipple attached.



The next job was to create the recess for the branch of the valve and they were held in the milling vice as shown, it not being worth setting up a fancy holding solution to get around the differing diameters problem. Although it can't be seen, there is a nut screwed to the front to protect the thread from crushing.



To turn the opposite end, they were screwed into a tapped mandrel and the procedures repeated. To release them from the mandrel, a piece of emery cloth was wrapped round the valve and pliers used to grip. The recess for the branch came in handy here.



The last job done today was to part off 1/4" lengths of 1/4" diameter brass and silver-solder them into the branch recesses. A turn of 0.5mm silver solder wire was wrapped around each branch and the parts held with some sacrificial clips. The threads were coated in correction fluid for protection. This picture was taken immediately after soldering.



And here is one alongside the drawing. These will go in the pickle overnight and I will machine the rest of the valve tomorrow.



Steve

[mr swarf]

That's a great way to form the radius, now I know what to do with that jar of broken end mills.

Paul

[springcrocus]

Once the globe valves were removed from the pickle and washed, the next job was to machine the spindle branch. To ensure this was running true, I used the drill chuck to align and load them to the independant 4-jaw chuck. It might have been smarter to use 5/16" diameter material and skim it down but it's what it is. I also have a backstop in place to control the depth.



With the branch axially aligned to the lathe centreline, they were then clocked for minimal runout.



It was important to be able to control the depth of the drill and not penetrate through the base of the valve so I set up my spare digital caliper in the contraption I cobbled up a few years ago. I made sure that the 3/32" pilot drill didn't exceed 5/8" deep, leaving about 1/16" wall thickness at the bottom.



This was followed with a 3.3mm drill that I resharpened as a counterbore, then tapped 5/32" x 40 tpi although I could have used M4 if I had thought about it in time. The 1/4" x 40 tpi outside thread was cut last.

Setting the parts to drill the angled steam ports was achieved by holding them, tilted over, in a small machine vice using the counterbore drill to help set it true in the other plane. Not easy to describe, nor to photograph, unfortunately



This vice was then held in the main milling vice and the hole drilled using a PCB drill. Alignment was done purely by eye; as long as the drill meets the drilled spindle hole then all is well. The opposite end was done by turning the vice over.



The draughtsman doesn't show a spindle, he had obviously lost interest by this time. Design your own seems to be the order of the day and I have made mine from 5/32" diameter stainless steel. I have designed my spindle to act onto a 1/8" diameter ball bearing and to be captive in the valve. Length is also an unknown until the pipework is in place and my best guess is about 3.1/4" max. The front of the spindle was reduced to 0.115" diameter by 3/16" long, then threaded 5/32" x 40 tpi.



The main part of the spindle was reduced to 1/8" diameter by holding on the front turned portion and supporting the far end in a hollow live centre. The middle section grew a few thou, as expected, and was abraded out with emery cloth. I left about 3/32" length of thread.



The final picture shows the component parts of one valve and the other assembled. Once the final overall length is known, a square will be milled on the end for a handwheel. I also need to make a pair of gland nuts to hold the spindle captive, the one shown is a standard pipe nut with oversize hole.



This type of valve is directional, steam flowing right-to-left as drawn, and I have embossed arrows on the valves to ensure correct orientation.

Steve

[springcrocus]

I'm scratching around for jobs to do now. I needed five dummy gauges and made these from brass bar. They were each made in a single operation, drilling and boring the minor diameter through for either two or three parts per size, then boring the other inside diameter for a depth of 5/32". The o/d's were skimmed and the smaller diameter turned behind with the parting tool, prior to parting off.



The brackets for the gauges have also been made from 20 swg brass offcuts, hacksaw and files being the main tools of the day. This one carries the main pressure gauge with the dummy steam heat gauge above. I will make inserts for these.



The two brackets for the injector stop valves have been cut from 14 swg brass and the support for the manifold spindle also made. Every photo I've seen shows the shape that I have made, not what the draughtsman drew.



And I've also done a bit of painting on the tender. Not my favourite pastime, 28 tinlets of Humbrol Brunswick Green.

Steve

The paint - not really, my little joke with Julian.

Steve

[springcrocus]

I have made the former for the backhead cladding, or false backhead, in a similar fashion to the other firebox formers, using an offcut of 40mm solid oak kitchen worktop. This backplate is not flat, there being a fifteen degree change of angle about in line with the lower water level bushes and a return change near the bottom. After cutting out the basic shape with a jigsaw, I mounted the former on the mill table and set up to flycut the first angle.



The nails are to prevent the wedges wandering downhill due to the slope. I have set the plate at ten degrees for this cut as it is the shortest distance and will leave the greatest thickness of material. The point in the chuck is being used to roughly true the former to the table. The flycutter to be used is a home-made affair with a boring bar in it.



This is the first section finished. Each depth of cut was 100 thou, four passes being needed to clean the top section. The small boring bar has one of the very sharp, polished carbide tips loaded.



I needed the top to be square to the newly-created face and this was milled with a long-series 5/8" dia end mill, finishing at 2.15/16" from the bend line. A quarter-inch allowance had been left on the height specifically for this.



Then the work was set up at five degrees in the other direction to get the planned fifteen degree face. This one was more of a challenge to set up because the clamps have to be between the two bearing points of the former. To stop the workpiece rolling away, a cheap, rubbish square was clamped down to act as a fence where it overhung the edge of the table.



The final setup was to create the 3/4" wide section at the bottom and for this the former was clamped flat to the table and the tool height set to just kiss the scribed line.



The final shaping will be done tomorrow and, meantime, I had a really good cleanup around the mill. I remember the tannin staining after the last time I milled oak and didn't want a repeat of that. This is also one of those times when the table power feed really comes into it's own; hand-feeding all those long cuts would have been no fun.

Steve

[springcrocus]

The backhead former has now been shaped and had the radii filed on. I tried to use the linisher but my belts aren't coarse enough and I was scorching the wood.



The former was laid onto a sheet of 20 swg brass and the basic shape marked out on the brass with 3/4" allowance on each side for the returns. A fifteen degree fold was put into the workpiece and this was then clamped to the former and flanging began.



I annealed the material about four times and the next photo was taken after the second annealing. Because of the fairly tight bends at the top, the material was creasing quite considerably.



After the fourth annealing, it became obvious that I would not be able to eliminate the creases. Unlike the copper for the boiler where the material was able to fatten up and flow, the thinness of the brass meant it was easier for it to deform rather than flow. Therefore, a pair of triangular sections were removed at each corner and the material beaten over in separate sections. These were then silver-soldered together prior to further melding.



Once the final shape had been achieved, the flanges were trimmed back to about 5/8". Then the backhead was offered up and the height set with packing. The bar across the top is to check that the top is about level with the bush for the steam manifold.



Then it was just a case of marking out the various cut-outs for the boiler bushes, the firehole and the platform support blocks. Here is where I reached by close of play tonight.



More tomorrow.

Steve

[mbrown]

Flanging thin brass sheet is a pig of a job - you seem to have overcome the difficulties well. But why not use copper, or gilding metal for that matter, which flow rather more easily?

Malcolm

[springcrocus]

Malcolm: Brass is what I have in stock. I don't have any thin copper sheet and I've never used gilding metal but it seems quite expensive.

Continuing with the backhead, I marked the position of the gauge glasses by blue-ing the bush faces and offering the plate up. After drilling the four holes at 1/4", I checked the position of them and eased them to the correct positions with a rat-tail file. Then a step drill was used to open them up to the nearest clearance size.



Next, I made up some spacers and washers to hold the backhead in the correct place using the blanking plugs I'd kept from the pressure test. I was then able to mark out the firehole from inside the firebox. This was set up on the mill to remove the waste.



Who needs CNC? CHC will do. Careful hand control enabled me to cut the waste out freehand using a 1/8" slot drill. The slot on the side is where my blower pipe comes out and should be almost invisible when the cab is built.



As can be seen, the workpiece is clamped down onto a piece of timber and everything done by eye. Quicker than chain-drilling is one is careful.



The backhead was replaced on the loco and the firehole dressed with files. Attention then turned to the firedoors and rails. The rails are made from 3/16" square brass but, not having any, I cut down some 1/2" x 1/4" material from the scrap box. This is a very good example of the stresses in drawn materials.



These were straightened up and milled to size, followed by milling the door channel and drilling the mounting holes. For the lower rail, a pair of bosses from 3/16" plate were silver-soldered on and the holes for the bushes marked out.



A pair of bosses with 3/16" dia spigots were also made up and these were soldered into 3/16" dia drilled holes. This provides a little mechanical strength as well, although I expect it's not needed. Ive also cut the material for the firehole doors from 16 swg brass sheet but I'm making these an 1/8" shorter than specified on the drawing because the oversize gauge glass assemblies would otherwise make it a bit cramped -looking.



Getting there.

Steve

[springcrocus]

The firehole doors are more than just flat plates, they have some sort of baffle on the front. A piece of 16 swg brass was bent around a 5/8" diameter bar and two firedoor fronts cut from it. The sides were marked out by tracing from the drawing, then cut and linished to shape.



As noted on the drawing, a spacer was used to help hold things together during silver-soldering. I made a pair of 3/16" dia rods drilled and tapped 8BA each end, then drilled 8BA clearance holes in the sides, a pair together each time. The small drilling fixture was particularly good for this.



The rods held the side nice and square and also acted as a clamping point. 1mm silver solder was laid inside against the corners and heat applied from below. Between the screws of the clamp the second one can be seen waiting to be soldered.



The bottom was milled to size next; I removed the screws but left the spacer in place for rigidity. I scribed a line at 1.9/16" and set this, by eye, level with the vice jaw. Exact size is unimportant.



To solder the two parts of the door together, I put the screws back, just in case the second heat-up softened the earlier work, and made a horseshoe of silver solder to rest around the baffle. Heat was applied from below until the solder melted and flowed into the joint. One hand for the torch and one for the camera here.



The next picture was taken a minute or so later.



Last operation on the doors was to add the bosses for the operating levers. I made up a pair of 5/16" dia brass bosses that went right through to the door plate and step-milled holes to suit. This makes the boss stand vertical to the back face and saves having to make an angled face on the bosses. They can be soldered on the inside, too.



Linkages next.

Steve

[92220]

Very nice job Steve, those fire doors!!

Bob.

[jon38r80]

Were these some sort of drafting vent on the full size loco?
Very neatly executed anyway as usual. For someone who professes not to like silver soldering (I don’t either) they are very neat .

[springcrocus]

Jun 24, 2020 20:25:37 GMT jon38r80 said:

Were these some sort of drafting vent on the full size loco?
Very neatly executed anyway as usual. For someone who professes not to like silver soldering (I don’t either) they are very neat .

Thanks, Jon.

Until I started the boiler, I had very limited success with silver-soldering. Having Wilf as a partner in that section of the build changed everything; we both improved our skills and, more importantly, our confidence. Also, I (we) discovered the need for a good flux and how important it is to a successful joint. I find the flux that I use suits most situations and is 100% reliable. I used to approach silver-soldering with much trepidation, now it's just like any other workshop skill. Practise, practise, practise! It really does make perfect.

Bob: thanks for your kind comment.

Regards, Steve

[92220]

Hi Jon.

Yes they are drafting vents.

Bob.

[springcrocus]

Before making the firehole door handles, I fixed the rails to the backhead so that I could check all the hole positions. The backhead was clamped onto a lump of 20mm MDF and the six 8BA holes drilled and tapped.



The handles were cut from 16 swg mild steel sheet, then bent at the appropriate positions. The pivot bosses, as drawn, are on the wrong side of the handles. I silver-soldered mine on the outside, the bosses having a 3/16" diameter locating spigot. The slots for the door pins were done after bending. The angles have been adjusted to suit the firehole doors.



The drawing shows the linking bar as a fabrication of three pieces but I made mine from 3/16" square brass bar and, after facing to length, drilled the two holes for the cotter pins. Then I milled the two slots using a 65 thou slitting saw.



Then I milled away the top of the centre section to more closely resemble the prototype. I put some 1/16" packing pieces in the slots and dropped the bar into the vice. I'm only holding on 1/16" but it's enough for this.



To make the clevis pins, I used 1/8" diameter iron rivets, holding them in a collet and threading them 5BA. Because there is practically no head to hold on, I loaded them into a block with a 1/8" drilled hole through it, then offered it up to the collet.



That's good enough to set them square and run a button die down.



After making the pivot pins, I assembled everything to check it all worked as it should. Doors closed...



... and doors open.



However, I didn't like the look of the handles, I felt they appeared too wide and clumsy. I decided to thin them down to 1/4" wide and also dress the heads of the clevis pins. I also made the brass top for the handle, which is soldered into place, and the door latch .



And I couldn't resist screwing things into place for a preview of what the backhead will look like. There are a few mods I need to make but nothing difficult, the tray for one! Half a cup of tea each at that angle. The door latch holds the doors slightly open for secondary air.



Steve