Drawing 8 - Cylinders & Accessories
1. Cylinders    
I have chosen to make my cylinders from the cast iron casting that are available although I could have chosen gunmetal cylinder castings or raw material of either type. I will also be using cast iron pistons and cast iron rings. The castings were a reasonably good shape and I didn't see any chilled spots although I wouldn't have expected to with lumps this size, and with no small protuberances which are the usual casualties. The first job I did was to load them to the independant four-jaw chuck on the lathe and clean up the ends. If there were any chilled spots, my lathe tooling is more substantial and better able to cut through them. I wasn't trying to get anything accurate machined here, just to get a couple of flat surfaces to help subsequent operations.
They weren't coming off the lathe anything like square so it was over to the milling machine and loading to an angle plate, getting the casting reasonably square to the table and then cleaning these faces up again. Again, I wasn't after anything accurate here, just a starting point for clamping purposes. I used my large flycutter for this, 4" diameter with a carbide cutting tip rotating at about 400 rpm. Once I was happy that the ends were reasonably square, I then machined the back, clamping to the angle plate through the two rough bores and removing enough material to just clean up. These were now measured and checked properly for squareness and centre-ing which showed there was still about an eighth of an inch still to come off the length. To reduce the overall length, I opted to hold them a little differently, using my independant 4-jaw chuck mounted to the table, adjusting how much I took off each end to get them balanced.
The drawing shows that the centrelines of the cylinder and the valve bores are offset 7/16" from each other in the vertical plane but a close look at the castings revealed that the core for the valve bore is set about 1/16" too close to the front of the casting so I have made the offset 1/2" which will allow the cylinder core to be machined reasonably evenly and the valve core will lose 3/16" at the back and 1/16" at then front to bring them back to the 7/16" offset. I set them up as shown below. The plugs are small pieces of MDF wedged in and centered so that I could eyeball the amount of tilt required. After setting the scribing block to the centre of the valve bore, I set it on a piece of 1/2" tool steel and adjusted the casting to set the centre of the cylinder. Then it was bolted up, the clamp removed and the back faced off again until I had a full clean-up.
The next thing I did was to go back to the earlier ops and get the overall length finished to size and to get the centreline of the piston bore reasonably accurate to the back face. These were just repeats of the earlier machining. Then I drilled all the bolt holes on the back of the cylinders but I've also added a couple of extra holes that are not on the drawing. These have been reamed 3mm and I have also modified the mounting plate by drilling and reaming two holes and setting 3mm dowels into them. Although I am going to finish the bores using a between-centres boring bar, it's a lot of work because these bores have a lot of meat in them. Therefore I set up to rough machine the two bores and had to make a holder to carry the boring bar to enable me to load it to the 4-jaw independant chuck.
Here, I've loaded up the first cylinder and set the centre of the bore before locking the slide in the first direction and taking the gib screws up pretty tight in the other. Then it was just a case of feeding through under power, backing out, adjust the bar and go again. However, after the first few passes on the first cylinder, it became obvious that the setup wasn't quite rigid enough so I stopped machining and modified the lower clamp by bolting on a block with an M10 bolt through it. Once I had clamped the cylinder up again, I then adjusted this bolt down onto the slideway with just slight pressure and a piece of brass packing beneath and this took practically all the spring out of the fixture.
And this is where the dowels came in useful. I was able to machine the largest bore on the first cylinder, unbolt it and load the second cylinder and be confident that the two bores were in the same place relative to the casting. Then l used the same boring bar to rough out the valve bore before setting up for between-centres boring to finished size. Because I put a pair of locating dowels in my mounting plate, I was able to get everything clocked true and then load each cylinder in turn and continue with between-centres boring getting each bore to about ten thou undersize and then setting the boring bar to final size and finishing each bore in turn. I have left all the bores three thou undersize to allow for honing, which should be ample, but it won't matter if they need a bit more. The valve liners and pistons will be made to suit.
I was able to measure them in situ without removing the boring bar using a dial caliper gauge but forgot to take a picture at the time. Here is a picture taken later showing the type of clock I am referring to. They are very good comparator gauges and I have modified this one slightly with some pressed-on stainless steel feet. I set the size using a 1" - 2" micrometer. To drill the various holes, I took full advantage of the new DRO and used co-ordinate positioning throughout. Each cylinder casting was squared up on the mill and the centre of the main bore accurately found using a wobbler. Using the tables in the Zeus book for a twelve-hole setup, I worked out the X and Y co-ordinates of each hole and worked my way round centre-drilling and then drilling 3.0mm diameter.
The draincock holes were also drilled and tapped 1/4" x 40 and the final work on the mill was to mill the steam port and drill and tap the holes. The drawing shows that the face of the steam port is 28.5° to the horizontal and recommends drilling the steam inlet hole after assembly of the valve liners. Using angle gauges, I set mine to 27° and milled the face flat with an endmill. I also drilled the 4BA stud holes and the 5/16" dia steam hole at this stage and will drill through again after assembly. The cast pads were not exactly in the centre of the castings which is down to me for not getting equal amounts off of each end of the casting. However, I have made sure that the holes are central to the end faces. Finally, all seventy-eight holes were tapped freehand using a 4BA spiral point tap (except the draincock holes, of course) and I am relieved that all went well. To break a tap at the last moment would be disappointing to say the least.
Before I fit the valve sleeves, I needed to hone the cylinder bores because access may have been a bit limited later. The hone kit came from Cromwell Tools and did the job perfectly. If you've never used a hone before, the trick is to start AND stop with the stones in the bore (never pull the hone out whilst under power or you will probably destroy the hone), and run at high speed for twenty seconds or so, stop, clean both the bore and the stones, another squirt of light oil ( I used 3-in-1) and repeat. Follow this routine until you're happy with the finish. As regards size, you will probably need to work for half an hour or so to remove about three thou in cast iron. It's all about surface finish, not removal of material, get your size nearly correct before honing.
2. Cylinder Backplates    
I cut the cylinder backplates out a while ago using some 10 swg mild steel plate, offcuts of the frames material, and had marked out and drilled them. I have now countersunk the holes and screwed a few of the 4BA screws into place but one or two of the holes are not as accurate as I would have liked. These were made before I had co-ordinate drilling facilities whereas the cylinders were drilled using the DRO. I've opened the lower row of holes out an extra four thou and the screws are now pulling in, but I may remake these plates as there is not a lot of work involved. One advantage would be that I can get the dowel holes that I have added to the cylinder in the correct place.
3. Valve Sleeves    
I am making most of the parts for the cylinders from cast iron and I have now made the piston valve sleeves from some old sash weights. The material is not of the best quality and I got black as a coalman whilst machining it but the end product is quite acceptable. Unlike the majority of components in this build, both the bore and the O/D need to be very accurate and perfectly concentric because of how they are assembled. This would usually lend itself to making from bar stock, machining both the O/D and the bore to size and parting off to length as a single setup. However, my lathe is not the most accurate machine in the world and that is why I chose to do the job this way instead. The first job was to cut up and rough out the four billets, although they were faced to finished length at this stage.
Next I modified one of my between-centres boring bars to get down to this smaller diameter and also made up some brackets to hold one of my machine vices on the home-made vertical slide. Although a lot of time has been spent in preparing this, the vice arrangement will come in useful again in the future, I am sure. All four sleeves were finished in multiple passes and, therefore, marked with a felt-tip pen to ensure they went back in the same place each time. To machine the outside, a mandrel was made from some spare material and the sleeves loaded in turn and brought to final diameter in multiple passes, the final cut being about six thou. They were then reloaded to machine the slightly reduced diameter of the external part of the sleeve, and for the steam port groove in each one.
The last operation in the lathe was to load to the 4-jaw and clock the bore true prior to opening out the locating spigot (0.950" dia) for the end covers. Finally, it was over to the mill and the steam port holes all put in. The inlet holes are just a series of eight drilled holes of 3/16" diameter and the outlet hole a 9/32" x 9/16" slot. These will now be cleaned up, pressed into the cylinders making sure the outlet ports line up with the cast-in holes and the hone ran through each to get a perfect central join. The piston valves will then be made to suit whatever the final size turns out to be.
I decided to take the cylinders to a local garage and make use of their 30 ton press to push the four valve liners into place. The liners protrude from the main casting by 0.712" at each end and a ring of this thickness and with a biggish bore was made to rest the cylinder on and act as a depth stop. A packer with spigot was also made to place inside the liner to protect the liner from damage. The first one was aligned with the cast-in exhaust ports of the cylinder and pressed in to just past the finished depth. The assembly was then upended and placed on the depth-stop ring the second one likewise aligned and pressed home until it touched the other liner. A bit of extra pressure then moved both to the correct depth. This was repeated for the other cylinder and the whole job took no more than ten minutes.
Now I needed to mount them to their backplates and get them located on the frames for spotting through. The cylinders on Britannia are inclined and it is not very easy to use co-ordinate drilling to mark the holes in the frames. The other problem is that, because the cylinders are castings, the centreline of the cylinders won't neccessarily be where I think they should be. On the frames, I had previously marked out the position of where the exact centre of the cylinder should be and scribed a line from a given dimension at the front of the frames to the axle centrline of the driving axle.
All I needed to to do was find the exact centre of each cylinder and transfer this to the cylinder backplate and I could then line everything up, clamp together and drill through. First I set up an angle plate on the mill, accurately squared off, and mounted the first cylinder and backplate assembly. Using a wobbler, I found the exact centre of the bore then swapped the wobbler for a scribing point and, without moving the "Y" axis, dragged it across the edge of the backplate. The assembly was then upended and the procedure repeated so that I had a scribed line at each end of the backplate.
The backplates were then removed and a line scribed on the plates between each of the two marks. Using a digital caliper, the half-way point was marked, followed by centre-punching and drilling 3/32". Back with the frames, I also drilled the previously marked cylinder centre point with the same 3/32" drill. The plate was then laid on the frame. a 3/32" rivet dropped through the holes and the marks on the end of the plates aligned with the scribed line on the frames. The whole lot was then clamped together and all the holes drilled through with a 4BA clearance drill. Clamping the two frames together and doing as a single item wouldn't work because the other cylinder alignment would probably be slightly different so he other frame was machined separately and finished the same way. I also scribed the position of the steam exhaust hole on the frames and these, along with a few other holes such as the saddle mounting holes will also be completed while I have the frames disassembled.
4. Rear Valve Guides    
The rear valve guides are usually machined from gunmetal castings but I decided to make mine from a couple of billets of cast iron which are 40mm dia Meehanite and 75mm long. The drawing is not dimensioned around the rib area and I have had to scale from the drawing to get the sizes somewhere near. The first part of the job was to turn the O/D and the front boss that fits in the valve sleeve. The hole for the valve stem was also drilled and then bored with a tiny boring bar to ensure concentricity. It chattered like a cage full of monkeys but finishing with a 7/32" reamer cleaned it up nicely, These were then loaded to soft jaws to machine the outer form of the guide rails and face to length.
Next it was over to the mill to make the final shape of the ribs. This entailed multiple operations, starting with hacksawing a large lump from the front. The 4-jaw chuck was mounted sideways using an angle plate and the centreline clocked and set on the DRO with all y-axis dimensions were referenced from this. A 12mm endmill was used to get the majority of the front-end material removed and a rummage in the "specials" box came up with this cutter to produce the angled form using multiple cuts and sneaking up on the shape until reasonably smooth.
Now that there was some working room, the holes for mounting the guide rails were next, drilling for 8BA. These will be tapped by hand later. Because of restricted access up near the chuck, removing the waste from the centre section meant making a holder to extend the reach of a 6mm end mill. The holder is a piece of 5/8" dia bar reamed 6mm, then cross-drilled and tapped M4 for a clamping grub screw. There was a small amount of chatter but easily dressed out with sanding drums in the dremmel.
The outside form was also machined with the same tool. Most of these cuts for both operations were plunge cuts, moving over about forty thou per cut and finishing with a ten thou facing cut to lose the plunge lines. That is as much as can be done for now. They need to go back in the lathe to machine the pocket for the valve gland and they will need some relieving under the guide rails to clear the slidebar casting but I will leave this until assembly time.
5. Valve Guide Glands    
Tthe valve glands have been made from bronze bar and this was just straightforward turning and parting off followed by loading to the 4-jaw on the mill and drilling the 8BA clearance holes. I then marked out the shape of the flange and the edges were hacksawed off and linished smooth. I have also now put the tapped holes and the pocket in the valve guide and this needed a couple of extended tools made to reach. The drill is jammed into a short length of brass tube but not soldered, remaining tight during the drilling of the four holes. I didn't have any tube suitable for the 8BA tap so I drilled a hole in the end of 1/8" key steel and banged that onto the end of the tap, which saved having to file some flats for the tap wrench.
The pocket was created using a 10mm long-series slot drill since I didn't have a 3/8" one and the gland turned to suit. These were then assembled but without the PTFE washer and it became apparent that I had an alignment issue. It wasn't much but enough to stop a 7/32" turned rod from sliding through, even when I loosened or turned the glands through one-eighty degrees. I must have messed up on my clocking up somewhere. I decided that the best thing to do was to increase the size through the gland and valve guide to the next standard size. 7/32" is approximately 5.5mm so I decided to lift that to 6mm, the advantage being that 6mm dia stainless rod is easily available as a stock size
The soft jaws were returned to the lathe and re-skimmed - the chuck had been off in the interim - and the guides and glands were assembled and marked as pairs, complete with orientation, just in case I have to disassemble them again. A special boring bar had to be made but this will come in useful for other things in the future so worth the investment in time. The home-made boring bar is a piece of silver steel which I have hardened and tempered. The assemblies were then loaded to the soft jaws and a few passes made taking three or four thou at a time until I could get the shank of a 5.7mm drill in the hole. This was enough to true them up and then the 6mm reamer was pushed through.
6. Front End Caps    
The front caps for the cylinders have been machined from blanks sawn from bronze plate using offcuts from the smokebox door and ring. I started by marking a circle on them using a cardboard disc and felt-tip pen, then aligning the rough billets in the independant 4-jaw chuck. These were then faced off and the outer diameter turned to size, a small radius being filed on also. The opposite side was turned using soft jaws in the 3-jaw chuck. Care was taken to ensure an accurate size for location in the bore, and good finish on the cylinder mating face although a gasket could be used if neccessary.
The spigot depth was controlled to exact size using the compound slide to set the tool. The bolt holes were produced on the mill using the DRO and the data from the Zeus book. The self-centering 4-jaw chuck was mounted on the mill table and the first workpiece accurately centered. The covers use a twelve-hole solution but with only eleven used because of the cutaway on the reverse to allow steam in and out of the cylinder. After finding the position of the first hole, this was marked with a felt-tip pen and NOT drilled. It's too easy to forget ones self and mess it up at this stage by drilling twelve holes.
Each of the rest were centre-drilled, drilled and countersunk at each of the positions alternately. It's a little laborious keep changing tools but no more so than winding back and forth three lots of times. Before removing from the chuck, a 6BA blind hole was drilled and tapped at the 0,0 co-ordinate for fixing the outer cover to, and the cylinder end cap is not seen once all is assembled. Finally, the steam access cutaway was machined by holding vertically in a vice and using a woodruff cutter to create the recess. By finding the centre of the workpiece and loading on two drills, the cutaway has to end up in the correct place. I only had a 1/16" wide cutter so four plunges were taken in forty five thou steps to create a 3/16" high recess.
7. Rear End Caps    
I have made the rear cylinder caps from cast iron although the glands are from broze offcuts. I made the glands first because I wanted to fit them to the cylinder caps before final sizing. The bore is just pilot-drilled at 1/4" diameter in the photo. To make the cylinder cap, I started by turning the O/D and forming the locating spigot for the cylinder bore, leaving it twenty thou above finish size. A 1/4" pilot hole was drilled through the billet as well.
They were then loaded to soft jaws and the second operation undertaken. It seemed a shame to waste so much cast iron machining the gland housing so I removed a ring of material using a hole saw to create a radial groove in the face and then parted the ring off. These will come in useful for something one day. The rest of the machining was then completed, ensuring that the gland was a good fit in the housing. Moving over to the mill, all the holes were drilled on their respective PCD's in a similar fashion to the front caps (covered earlier) and the gland fixing holes tapped 8BA.
8BA studs were then fitted to the end caps and the glands inserted and nutted down tight. The assemblies were then returned to the lathe, held in soft jaws, and the piston rod holes bored and reamed to size. The cylinder locating spigot was also skimmed to final diameter, plus a couple of thou of the face, which ensures that the whole assembly is perfectly concentric with the bore of the cylinder. Finally, the steam inlet recess was formed in the same manner as the cylinder front caps. The underside of the glands and the housings have been notched to aid correct repositioning. A piece of 3/8" dia centreless ground material has been set in place to check the fit.
8. Valve Front Cover    
I have made the valve front covers from cast iron, in this case from some old but better-quality sash weights. I machined the O/D and the spigot for the valve first, and drilled the valve rod guide hole to depth at the same visit. The parts were then reversed in the chuck and the rest of the machining completed as a single operation.
9. Pistons    
The pistons have been made from cast iron billets, whilst the piston rods are made from an unknown grade of stainless steel and I am using two cast iron piston rings rather than the single "O" ring specified on the drawing. I started by cleaning up one face and the O/D of the billets, then prepared some soft jaws for subsequent operations. The O/D was left thirty thou up for finishing after assembly but the bore and thread were finished to size at this visit. My billets were just over an inch long and I was able to get a piston ring from each one but I missed a trick by not ordering slightly longer billets to enable two or more rings to be made from the surplus. These were made first, then the piston body was faced to length.
This was followed with the piston rods, machining to finished length and forming the thread. Because I only have one 5/16" x 40tpi button die in carbon steel, I didn't want to risk it with stainless steel so looked for an alternative. I have got a HSS M8 x 1.0 die, though, and it was cheaper to use this and buy a matching tap for the piston body. The rods were screwed into the bodies and pulled up really tight. The assembly was returned to the lathe and set up between centres prior to finishing the outside diameter to 1.749", and the two ring grooves formed. I have made these 1/8" wide by just over 1/16" deep since this seems to be a standard size.
Returning to the rings, the soft jaws were remachined to the new diameter and the billets reversed in the chuck. The front ends were skimmed to true the face, the inside bored out to 1.625" and the O/D finished at 1.750". The rings were parted off at 0.125" wide. Two more rings were made from raw stock and all four were then carefully broken at a single point by holding in the vice and being given a sharp tap from a hammer and punch. The inner faces were filed smooth and square to help the next part of the job.
Next they had to be expanded for an easy fit over the piston and to give a spring contact to the bore of the cylinder. They were sprung onto a length of 3/16" thick flat stock and heated until cherry red, at which point they dropped off the flat stock and were left to cool. After a light clean-up with some wire wool, they are ready for assembly onto the pistons.
10. Valve Spindles    
Because I had to enlarge the hole through the valve guide, I have made my valve spindles from 6mm dia stainless steel with a M5 thread at the slidebar end and using M6 x 0.75 for the bobbin locking screws rather than the specified 7/32" x 40 tpi. The drawing incorrectly suggests using 2BA nuts on these threads. The central area has been reduced to 5mm to ensure it clears the root of the thread. I also drilled a 5/16" dia hole through my button-die holder mandrel to allow the spindle to pass further through the die-holder. The nuts were made from some 3/8" hexagon brass and used as a gauge to set the size of the male thread.
11. Valve Bobbins    
The valve bobbins are made of cast iron to match the valve liners although the drawing calls for stainless steel if using gunmetal liners. I made both bobbins from the same stick of meehanite using the centre section for holding in the chuck. I started by cleaning up the front face and roughing out the O/D to thirty thou oversize. Next I machined the centre reccess to finish diameter and accurately finished the front shoulder dimension to 0.312" and the gap to 1.625" with a wide parting tool . Because I am using 6mm for the valve spindles, rather than 7/32", I then drilled and reamed the through-hole at 1/4" diameter. After turning round and repeating for the opposite end, I then parted off the two valves a few thou over-length and finally faced them to exactly 2.000".
A 1/4" diameter mandrel was made and left in the chuck to ensure concentricity and each bobbin loaded in turn. The O/D was then slowly reduced until just a couple of thou above finished size and then slowly polished down with an abrasive block until they were just starting to go in the valve bore. Then I used the hone in each bore, polishing until I had a slightly stiff, sliding fit. Each bobbin was matched to it's respective cylinder and marked to ensure they didn't get mixed up. The two bores are about half a thou different to each other. They have been loaded to their respective rods and will be adjusted for position at the appropriate time.
12. Cylinder Cladding    
I've been playing around with the cylinder cladding and am using the same number of fixing points as the prototype. The drawing calls for 22 swg brass or copper sheet but I'm using 28 swg (eighteen thou) because it doesn't look so bulky. First, I marked out the cutaway for the steam pipe and flange, leaving some material to make into tabs. I've clamped the two pieces onto some 12mm MDF and am using a 1/8" cutter to remove the waste. Once that was complete, I reclamped the job and drilled the various rivet holes, 1/16" for the fixing rivets and 3/64" for the inspection cover-plate, or whatever it is.
The positions of the rivets were guessed with reference to a photo I took at Bressingham twenty years ago. The larger holes are for the cylinder cocks. One can see where I'm attempting to make the tabs to support the steampipe cover part of the cladding. To make a perfect circle in thin material is not easy and I chose to screw a pair of blanks to a block of roof batten and set it in the rotary table. I only had to find the centre of rotation - centre of the chuck - and not worry about the postion of the material.
After drilling the four 1.2mm holes on a 3/4" PCD, I then loaded the 1/8" slot drill and wound out to 0.531" on the "Y", 0 on the "X" axis. After plunging through, I was then able to wind the table round to create the 15/16" diameter circle needed. After getting about half-way round, I added a couple more screws and washers to hold the discs firm. I then returned to datum, set the cutter down to depth and finished the cut. The photo to the right is the fireman's side and clearly shows the rearward offset of the circular cover plate.
13. Next item...