Stuart Twin Victoria (Princess Royal) Mill Engine

No, I do t have a boring head.

Anyway, I made some centre plugs for the cylinders,





and checked them using the arc method, scribing a circle with dividers and going round with the odd leg callipers:



Most are pretty good, might re-do a couple, although they might be ok, but a bit oval.

Also divided the 1/2” square bar up for the feet. Just enough in the end, using short middle bits:



I milled them to consistent lengths, and they are a bit oversized so I can trim them once fitted:



Didn’t have much luck soldering them together, the solder only seemed to capillary into a tiny bit of the cleaned and fluxed joints. I tried to tin the mating faces, but they weren’t having any of it…a slight tap saw them fall apart…



Perhaps the wrong flux, or not enough heat (propane/butane mix)?

Anyway some progress at least. Thanks.

that flux should have been ok so probably not enough heat to get a the metal upto temp. Tin and sweat is the best method

How were you holding the pieces together Doc? As Jason says tinning the faces is best but the parts need to be in sufficient contact without being too tight together.

You could try this - a piece of 14-12 swg piano wire bent as a two pronged clamp. Gives just enough pressure to allow the solder to flow. Well flux the parts and put a blob of solder across the joint then heat gently and uniformly until you see the solder flow

Idealy it wants to lay on something non heat absorbent - this is a piece of Skamoflex -perfect for this sort of op.

Edited By Ramon Wilson on 22/06/2021 22:27:36

Thanks both. I was using screw clamps, with flux between the parts, and heating each joint and dabbing with solder until a blob formed, and - sort of - ran into/around the joint. I guess it was too tight a join.

I did wonder about some kind in sprung clamp in conjunction with tinned surfaces, but never got them tinned in the first place.

Not sure I’ve got any insulating material. I tried to retain heat by doing it on wood, but of course it started to smoke. Maybe an old concrete block?

Anyway, I’ll try again tomorrow with piano wire. Cheers.

In fact thinking about it: I’ve got a section of Space Shuttle re-entry tile material somewhere. That should be a good insulator if it’s big enough. I’ll try and find it tomorrow.

Soft soldering part 2. As mentioned, I considered using Space Shuttle heat shield insulation to rest the parts on (to retain heat within the parts). I’ve had it for years and it’s pretty useless:



Then looked at how much it goes for on EBay and thought again (apparently production of it ceased in 1993). Settled on pre-tinning each individual part held in air on a stand:



Which worked:



As per Ramon’s comment, needed to clamp the parts to solder them all together. If a normal clamp was used, when the solder melts, the whole thing would drop out of position. Rather than using piano wire, I spring-loaded a clamp:



Then hung on the stand and sweated it all together:



Initially used wood marked with diagonals to get the centre, but checking with dividers gave an offset, which I didn’t want:



Remembering Jason’s comment, I used hot melt glue to put a piece of aluminium in the middle, and drilled the centre using the DROs in the mill. Then it was spot-on:



Then centred in the 4-jaw Chuck using spreader plates:



And gradually opened up to size. The initial cuts were very rough for some reason, almost as if the surface was hardened by the sweating/cooling process. After the third or fourth 0.0025” cut, it became smooth.







Then un-soldered again:



Seems a good fit for the first one. I suppose I could have turned a chamfer on one side rather than filing it to mate it to the cylinder flanges, but whatever:





Still quite a bit to do, but so far so good this evening.

They turned out OK, I suspect the small burner on your torch did not help with the initial soldering problem.

Could be. When I tapped the wooden plug into the centres, one assembly cracked apart - and it wasn’t that tight a fit. Had to re-sweat it.

Never had an issue with soldering copper plumbing, or electrical stuff, but soldering steel never seems to be very strong the way I do it.

Nice result Doc

Just a point - do you intend to clean that primer off where the JBW is going ? If not it will only be as good a bond as the primer. Hope that's not teachng granny but worth mentioning just in case.

Before you JBW them in place file a good, rough, chamfer on the radii - gives an extra surface for the JB to grip onto

Thanks Ramon. Intention is to use a sanding cylinder or burr in the dremel to roughen up the mating surfaces on the cylinders and feet.

There is a lot more bulk in those blocks of metal than thin walled copper tube so it will take a lot longer to get the whole of the metal upto temp with that small burner rather than just melting the solder onto the surface.

Carried on machining the other two feet, which was straightforward, then marked out for the pockets, which I did with a slot drill. I marked the extents of the rads w.r.t. the edges of the arc cut-out:



After cutting both pockets I double-checked the centres with the DRO, and they matched the base hole centres, so as per the drawings/instructions:



The result (still needs the flanks chamfering):



The question is (there’s always a question!):

The base holes are spaced at 44.46 mm, the pocket centres in the feet are the same. So, unless the feet are attached to the cylinder castings precisely such that the finished axis ends up dead in the middle (seems unlikely), then the feet hole centres will not match the pocket centres. There will be some lateral mismatch. I checked a 4BA hexagon, and it gives about 2mm gap on radius between the points and the pocket…I can’t bring the pockets any further in toward the castings otherwise they’ll break into the cut-out arc (the edges need still need filing slightly for a snug fit).

I think I’ll have to figure out some method of placing the un-bored cylinders onto the properly spaced feet before fixing them (ie using the centred wooden plugs).

Probably overthinking again or maybe underthinking. You need to remember it's a casting you are trying to replicate and that would never have had micron accuracy.

Understand the concept of it looking like a casting, just that there’s hardly any wiggle room at all laterally before the bolt/nut points would hit the inside of the pocket’s vertical wall. The longitudinal position of the holes is easy to get perfect simply by making a flat spacer to fit between the foot blocks as a jig.

Anyway, it’s probably been done hundreds of times in the past, so there we go. I’d rather ask the question than have to do re-work or make some kind of compromise. Thanks.

You could always put a shallow 1/16" hole in the underside of the cylinder and a matching one in the middle of the feet which would help locate them using a pin when gluing up.

I’m ok with the location onto the outside of the cylinder, but the final position of the axis isn’t defined because they’re not bored yet. That’s really the issue, because if they were bored, I could easily line everything up on the mill bed and clamp in place while the JBWeld sets. Then d&t for good measure.

basically my assumption is that the bored axis won’t end up being concentric with the outside of the castings, therefore there will be some lateral error. I have no idea how much, but if it’s more than about 1.5 mm it might be an issue in terms of fastener alignment.

Hi Doc, I think you are jumping ahead of the game again by making the feet to finished dimension before you fit them to the cylinder or, to put it another way your cylinder dimensions are now going to be defined by the feet.

As Jason rightly says in reality the cast feet would not be symetrical and in all probability, nice as it looks, have neatly milled pockets to surround the nuts. (Think of the type of spanner required if so)

Were it me I would mill those feet to finished dimensions - and relative to the finished bore - once securely JB'd in place and also do away with those pockets. A more likely prototypical design would be for the top faces to be slightly tapered downwards in an outwards direction and the nuts seating on inserted but protruding level pads - a little extra work but far more in keeping to a prototype.

Something on the lines of this perhaps?

Just a thought - hope it helps

Ramon

Thanks Ramon. I was really wanting to stick to the original geometry, purely for the sake of proving to myself I can build parts like this from plans and stock material. I don't think the 10V had anything like that.

Must admit I'd assumed that by "machining" when in place (i.e. treating as one casting), you guys primarily meant the flats on the base and the fastener holes, rather than the overall form of the feet. I don't think - for the plan parts - that it's possible to machine the pockets and chamfers once assembled to the main casting due to clearance issues. The pocket ends would certainly be extremely close to the cylinder sides, and the cylinder end flanges would obstruct machining of the chamfers.

Thinking more about it, I've got the centred plugs in the cylinder bores, and these effectively will be the axes of the finished bores, because they are what I'm aligning the boring bar with. So there must be a way of setting the cylinders on the feet with these as datums while the JB Weld sets. Not quite figured out how yet, but it should easily get me within the necessary tolerance.

Cheers.

If you mark a ctr line on a flat surface, the feet and then stand them on it with your end to end spacer and hot glue or clamp them in place so ctr lines are lined up you can then apply your JBW and put the cylinder in place then use a square to check your pop marks on the plugs are sitting vertically over the ctr line you should not be far out.

If your nut clearances are as tight as you say then think how you will tighten them, I you stick with that design may need one size smaller hex nuts or hold nut in place and use screws up from the underside through claearence holes if you can't get a spanner in.

Also worth noting that like me ramon tends to run a Dremel over the surfaces to give them a bit of texture and also knock off all the external corners, makes the difference between a painted "barstock" engine and one made from "castings"

I was thinking put a bit of aluminium plate on the mill bed, co-ordinate d&t four holes to match the bed, then drill the feet blocks close clearance 4BA, at the same time I'm milling the pockets. Might as well mill them to length too. I can also spot two small holes on the plate, and scribe a centreline - as you say.

Then bolt the feet in place and fettle everything so the cylinder seats nicely, and fits laterally using the square to check alignment (within a couple of mm). The height doesn't matter so much because that will be achieved by milling the foot pads later on.

When I'm placing the cylinder, I'll have to check that the valve plate is horizontal, and will clean up once fixed. I suppose I'll have to use the scribing block/surface gauge for this. Might invest in a vernier height gauge at some point, which would probably be easier to set to a precise height for scribing/checking.

So I suppose I'll end up doing half it like a whole casting, and half like an assembly.

BTW a 7/32" socket fits in the pocket, and matches the bolts (which will be nuts on assembly), so tightening shouldn't be an issue.

Thanks.