Cylinders and Valve chests

Valve chests

The inside valve chests come cast as two parts each – the port block, with the steam ports cast in, and the chest itself, and circular affair with some flats cast on for the steam inlet and valve entrance. You also need to cut the cover from ¼” thick steel.


The valve chest covers were laser cut as per the drawings, before the castings were obtained, but the supplied chest casting did not match them. It was 3.75” diameter, whereas the drawing quotes a 3.5” cover (and shows the casting as being 3.5” diameter). The drawing also shows the internal diameter as 2.5” as cast, but, again, the supplied casting was nowhere near this (I measured it at about 2.2”).


Now, I had an issue with the valve travel at full cut-off – the valve would be extremely close to the valve chest wall at the extremities – and, what with my abilities, could well hit it. However, because the drawing shows a 3.5” o/d and 2.5” i/d chest, and the casting is 3.75” o/d, I reckoned that there would be no harm in boring out the chest to 2.75” diameter. This still ought to be strong enough, yet avoid any possible accuracy-related problems with the valve travel. The cover securing bolts were put on a slightly larger PCD to compensate.


In order to align the valve rods correctly, my thoughts were that the best method would be:

  1. Drill and bolt port block to frame
  2. Drill and bolt valve chest to port block, measuring height of valve rod above the front axle, and positioning valve centrally above the ports for alignment
  3. Fit valve rod, inner motion plate and guides. Adjust for best fit

The first 2 were relatively straightforward, but the last left the guides floating in all 3 planes. I thought this would allow me to adjust, clamp and drill. In fact, it proved impossible to find a ‘best fit’.


So, I measured the distance from the frame to the valve chest (3”), and fitted the valve rod guide to the inner motion plate at that distance. This left me with much less to float, and made fitting the motion plate much easier. Then I cut the gaskets, and fitted the port blocks and valve chests to the frames. Then, I set the valve gear into the proper position for mid gear, and drilled and fitted the roll pins which hold the expansion link hangers in place. At the time of writing, I have made one valve buckle, so one set of inside motion is complete. I could spend hour just turning the wheels to watch it in motion.


First, I milled the mounting faces flat, so that the center of the cored out bore was approximately the correct distance from the face. This was a marathon - about 2.5 hours on the milling machine per cylinder and a large pile of debris afterwards. That was bad enought for two of them, but when the supplier's poor quality casting led to me doing it a third time I could have screamed!!

I had two ideas for boring the cylinders out:


As I have access to a large lathe at Swanley, the latter seemed the better way to go. On the internet, I found an article which said that, to avoid deflection, a boring bar should never overhang more that 3.5 times its diameter. As these cylinders have a large mounting face which the tool must pass, I needed to be able to overhang nearly 6”. Therefore, I bought a 14” length of 1.75” diameter bar (the cored out diameter of the cylinders is only 2”, so it’s going to be close), and milled off two flats 180 degrees apart as clamping faces at one end, and a ledge about ¾” long at the opposite end, for half the diameter, as a seat for the cutting tool. This is a carbide insert, which I bought cheaply on Ebay, and uses a 6mm bolt to secure it.


As mentioned earlier, the casting has a mounting plate at the back, which overhangs about 1” beyond the bore. Even the 4-jaw chuck for the big lathe didn’t have jaws long enough to hold it securely, so it would have to be attached to the face plate. I drilled some holes in the casting to assist holding it to the angle plate (either side of the bore, where they will be hidden by the cylinder wrapper plate). This was bolted up in the lathe, as close to true as is possible with such a rough casting. I did this by wedging a piece of wood in the bore, then drilling and fitting a long length of 1" diameter bar. This was then set to run as true as possible.

Then I took a pinch of courage, and started boring. With the carbide insert, I could easily take 40 thou deep cuts, and used the auto-traverse at a fairly coarse speed, so the initial cuts were done very quickly indeed. Before taking the final cut, I checked that the tool bit wasn’t damaged in any way – I’d paid so little for them that a new tip wouldn’t break the bank, but it was fine. So, I wound the last 20 thou on, put the auto traverse on as finely as it would go, and off I went. It took about 25 minutes to fully traverse the bore (it seemed like a week), and the result was a bore with a wonderful finish. Then I faced and counterbored the piston rod end of the block. This I did twice within about 5 hours.

Unfortunately, when I removed the second block from the faceplate, I found several blow holes in the bore. There was nothing for it – it would have to be replaced – and a whole day’s work was wasted (1/2 day boring out and 1/2 day milling off the mounting face). Send it back to the supplier, who promised to replace it and refund the postage. After 2 chasing phone calls, I got a new block back but I’m still waiting for the refund of the postage….


Then I machined the first cover. This went fine, until I drilled the hole for the piston rod – about ¼” into the operation, the drill hit and hard spot and was deflected. In order to fix this, I chucked it in the 3-jaw, and bored the hole out true. Then I fitted a bronze bush at the cylinder end to make it the correct size again. The slide bar mounts were then machined flat, parallel and equidistant from the bore on the milling machine, using a fixture (plate with a 3/8” rod sticking up). The cutter always passed down the same side, this was machined, then the casting turned 180 degrees, and machined again.


The front covers were simple – laser-cut circles of steel, with a small recess turned into them.


The last thing to do was the pistons – again, the castings had enough spare on them to make a third. This meant a lot of machining to get the correct size. The drawings I have are annotated several times as piston ring sizes have changed – and Reeves claimed that they couldn’t get any. A quick browse of the internet located the Clupet piston ring company (as specified on the drawings) who sent me 4 of the correct size. The drawing is unclear how to space them – I just subtracted 3/8” (2 piston ring widths) from the width of the piston, divided by 3, and that gave me the land thickness.