Greene King (S15) Cylinders


As mentioned previously, I passed the 3½" gauge S15 over to my brother for his next project but he only has a mini lathe so some of the jobs have been passed back to me! I've already machined the driving and coupled wheels for him and now it's the turn of the cylinders.

These are piston valve cylinders using the castings for Doris (LBSC's Black Five) with a few alterations. The castings came from Blackgates but they only had the blocks in stock at the time and no end covers etc. That's no big deal as it's easy enough (and probably easier!) to make these from brass bar. Apparently the valve liners they supply are only bits of gunmetal bar anyway.

I intend to fit PTFE rings to the pistons and also PTFE heads for the piston valves in view of past experience with both of these. I'm also going to try a little experiment with the valve liners as well. When looking through some old ME's, I came across an idea for a 3 piece liner (middle and two end pieces) instead of the one or two piece. The central part has the valve ports milled into the ends so it looks like a castellated nut at each end. The two outer parts then just butt up against the central part and hey presto, there are the ports. No drilling holes and filing them square. No flimsy wall between the pieces which may get crushed if they are too tight a fit when you press them in. More of that later!

And so to work:

Cylinder Castings

The major difference between the S15 and Doris is that the distance between the cylinder centre and the valve chest centre is increased from 1-3/16" to 1-1/4" and this makes things very tight. There is very little metal to spare at the top and bottom of the casting (on these castings anyway). I printed out a paper template and tried it on the end of the casting to see how much metal there is to play with.

Paper template on end of casting

To utilise the maximum height of the casting it's necessary to work from the outside edge. As you can see, the core holes are now miles off centre and there will be a fair bit to machine off the bolting face. You can see how much will be left after boring in the photo below - not a lot! I can't move the valve chest bore down any further or the cylinder wall will be too thin at the bottom.

Not much to spare!

Martin (Evans) says to machine the bore to 1-3/16" IF the casting will allow but I have my doubts that I will be able to go this big without leaving the cylinder wall a bit thin. I'll have to see how it goes during the boring process.

The first job then is to machine the bolting face and I thought it easiest to do this in the mill.The outside edge of the casting was cleaned up with a file so that it would sit flat without rocking and the line of the bolting face scribed on the end using the paper template to get the position. It's not just a case of using a scribing block on the surface plate as the outside of the castings are at a slight angle to the bolting face i.e. the cylinder block is wider at the top than at the bottom by 1/32". (You can see the line of the bolting face on the photo below).

Clamping something like this onto the milling table is quite awkward if you need to machine all the top of the casting and I used 2 bits of square bar through the core holes held down with the normal bolts and clamping bars.

Using 2 square bars to clamp the casting to the table

To get the bolting face parallel to the table, a piece of suitable thickness packing was put under the cylinder bore end of the casting to tilt that end up slightly. A scribing block was used to check that the scribed line on the casting was indeed parallel to the table. The clamps were then carefully tightened up until the casting was held firmly but without distortion. As an extra 'security' measure, I put clamps on either side of the casting to prevent it moving on the table under the milling cuts.

Milling the bolting face

I used a 1/4" carbide cutter to remove the 'skin' of the casting with the intention of changing to a larger diameter one to finish off but I actually did all the milling with the 1/4". The face has two large slots in it (one for the exhaust passage) so there's not really that much surface to mill and it didn't take long at all.

Bolting faces finished

Next, the boring bit!


Well, I spent nearly a day contemplating how to machine the bores of the cylinders! What a nightmare. The first option was bolting them onto an angle plate on the faceplate of the lathe as suggested in the write up. I do have a Keats type angle plate that I bought to do Helen's cylinders but it was too short really to hold the S15 castings. Next I considered using an ordinary angle plate but having tried that I decided that was not practical. None of the holes in the angle plate or the faceplate matched up sufficiently to ensure adequate bolts to be used and clamping the casting to the angleplate would have been fun! In any case, I don't think there is enough clearance in the lathe bed gap to allow the edge of the angleplate to clear. Also, I would have had to bolt a substantial 'counterweight' onto the faceplate to balance the whole thing up. All these methods sound easy on paper until you actually try doing it for real.

One of the difficulties is that these are piston valve cylinders and I've only got one machined face to act as a datum. If they had been slide valve cylinders I would have machined both the bolting face and the portface prior to boring and then I would have had two faces to play with.

Last year someone at the club gave me an R8 boring head so I even considered boring the castings in the mill but decided against this as the vertical feed on the quill of the mill is not too clever and getting a fine enough feed might have been difficult.

Eventually, I decided the only real option was to clamp the castings onto the cross-slide of the lathe and use a boring tool in the chuck. I know this is a 'standard' way to machine large cylinder castings but I'd never done it this way before so was a bit apprehensive. More nightmares followed! I tried mounting the casting with the bolting face resting on packing to bring the cylinder bore up to height. The Tee slots in the cross-slide were all in the wrong place to fit suitable clamps! I tried using an angle plate bolted to the cross-slide and clamping the casting vertically to that. Similar problems with the clamps again! By this time I was getting fed up with the whole job and wished I had never seen it!

I decided the only way to hold the casting easily was using G clamps and the angle plate bolted to the cross-slide. The angle plate was set parallel to the lathe axis using a square against the face plate to ensure the cylinder bore would be parallel to the bolting face. The bore centre was then set to the lathe centre height by suitable packing under the top edge of the casting which was pretty flat and true. The casting was then clamped against the angleplate using the biggest G clamp that I had and a piece of 1/4" steel plate to spread the load and avoid distorting the casting (easily done with gunmetal) The plate ensured the casting was clamped on the thicker ends of the casting rather than the thin wall of the bore. I would have liked to have used two G clamps, one at each end, but there was only room to fit the one.

A second fixing was set up to clamp the casting down onto the packing on the cross-slide, again using a bit of steel bar to spread the clamping pressure and make sure it was on the thicker ends of the casting. Finding somewhere to put the bolt for that was fun as well.

The whole affair looks very 'Heath Robinson' but actually worked very well and is very rigid. Also, it was easy to remove the casting and replace it in exactly the same position again should the need arise (as it did!)

The ideal boring tool is a between centres boring bar but I didn't have one (of course) so I resorted to the method I used for boring the spindle casting on the Kennet Tool grinder i.e. a square shank boring tool held in the 4 jaw chuck! The 'experts' will frown at this but actually it works very well. Putting the cut on is a bit fiddly as you have to adjust the chuck jaws every time and it's difficult to move the tool an exact amount but I find it good enough. I suppose I could have used the R8 boring head held in the chuck as the R8 shank has a long parallel section at the end but the overhang of the tool from the lathe spindle would have been excessive and I think I would have found it difficult to get a good parallel finish on the bore. The tool in the 4 jaw is very rigid.

Heath Robinson would be proud!

Before getting to this stage I had turned up a Go/No Go gauge from mild steel bar to check the finished diameter of the bore. I had decided to limit the diameter to 1-5/32" rather than 1-3/16" due to the lack of room for error in the casting.

To make the boring less 'boring' I set up the electric drive to the leadscrew so I didn't have to feed by hand. This works really well and I must make it less of a 'bodge' sometime!

Crude, but very effective!

Going well

The boring went without incident, despite the fact the core holes were considerably offset from the required centres, and the metal cut like butter. Well, that is until I got to within 1/8" of the finished diameter. The tool started making 'clicking' noises as it got just past the centre of the bore and yes, you've guessed it, BLOWHOLES!!

Blowholes in the casting - all that I needed!

Much 'Railroad Esparanto' followed (as Curly would say) but as I still had a 1/8" to go I carried on. The holes got worse and then better but were still there by the time I got to the finished size. I decided to leave this casting there and have a go at the second one to see what that was like. To be honest, the holes weren't too bad and probably wouldn't have had any effect on the finished bore but these weren't for me and I didn't think dear Bro would be very happy with them. My idea at this point was to return the casting and get a replacement. This was also my brother's idea when I told him!

Anyway, the second casting was set up and machined perfectly with no problems at all. No sign of any holes etc.

What to do with the first one? I didn't have anything to lose now so I measured the thickness of the cylinder wall in the thinnest part and decided there was enough metal to bore it out to the full 1-3/16". Back in the lathe it went and another cut taken to see what would happen. As luck would have it, this removed the worst of the pits so I finished the bore with three cuts at this final setting which left a good finish. The few pits left are quite small and so I took the casting for my brother to have a look at. "Is that it?" he said, " I thought there were great big holes in the bore, not these tiny little pits!" He was quite happy with them so that's saved me the hassle of getting a replacement casting and having to go through the whole machining proceedure again. In any case, a replacement casting may be as bad, or even worse.

What I have to do now is bore the second casting to the same size which is no big deal. I've left the boring tool at the same setting as the final cut for the 'faulty' casting so it's just a case of putting the casting back in the lathe and taking another cut. With the setup as it is, putting the casting back in the same position as it came out is dead easy, as mentioned before.

Funny enough, at the club yesterday we were discussing these castings and one chap had had a similar problem with some Rob Roy cylinders. What he had done was to fill the blowholes with silver solder and then machined the bores again. I've read that LBSC suggested using soft solder to fill them but not silver solder. I do wonder if you could use a high temperature epoxy such as JB weld with similar results? The obvious answer is to get a replacement casting but that may not always be possible if you've bought the loco part built or you've picked up a set of old castings that are no longer available.

Previous Page Next Page