The blanks for the links were cut from 1/8" gauge plate and the holes for the pivot pins and the eccentric rod pivot drilled and reamed in the lathe using the micrometer dials again. Each blank was then mounted in turn on the jig and the slot cut. The slots were rough cut slightly undersize with an 1/8" endmill and then the final cut made with a 3/16" endmill. I assumed that so long as the 3/16"endmill only had a few thou to remove it would cut an accurate slot 3/16" wide. (Unfortunately I did not check this at the time - see later!)

Link blank mounted in the jig

Link blank after milling the slot

All four slots cut - note left and right hand pairs

The blanks were then mounted in pairs on the jig and the outside profile machined to shape.

Machining the outside edges of the links

After all four links were profiled I decided to check the curved slots in the links and found they were miles out and the edges were nothing like parallel. Much workshop esperanto (as old Curly would say!). Obviously the jig hadn't produced a true arc and the 3/16" endmill had probably wandered when finishing the slot. I could have remounted the blanks on the jig and trued up the edges of the slot but that would have left the edges of the slots a bit on the thin side so I decided to scrap the lot and start again!

I checked the jig before the second attempt and noticed that there was a bit of play in the pivot which would not have helped so I eliminated that by putting a piece of very thin shim brass around the pivot bolt. I also had noticed that the nut on the pivot kept coming loose due to the movement of the brass plate so I locked that so that it remained tight. Four new blanks were cut and the whole process repeated, this time taking more care! Also I did not rely on the 3/16" endmill to cut both sides of the slot at once and took an extra 10 thou of each side so that the slot finished up slightly wider. The die blocks will be made to suit so the actual width is not important. As a check I measured the slot width all the way along with a drill shank. This time all the slots came out within a few tenths of a thou. Much better than the first time! Motto - check your work as you go along!!



Expansion links after profiling


The two sides of the links are held together with an 8BA bolt top and bottom with spacers made from 3/16" bar to leave a 1/8" gap for the radius rod to pass through. The link pivot journals are turned from 3/16" dia. silver steel with a short 1/8" dia stub which is a tight fit in the link itself. These journals will be silver soldered in place. I was a bit concerned about getting the journals perfectly square and in line so I drilled each one 3/32" diameter so that a length of silver steel could be put through each journal on final assembly. This will ensure the two journals are perfectly in line. The final operation before assembly of the links was to machine two little slots in the outer links purely for cosmetic reasons. This was done by holding the link on the brass plate again but mounted on the rotary table in the micro-mill. I did not bother doing the same on the inner links as you can't see them anyway!

Machining slots in the outer links

Expansion links with one assembled ready for soldering the journals


The link journals were silver soldered in place with the links actually bolted together and a length of 3/32" silver steel threaded at each end to take a nut put through the journals, the nuts holding the journals in place. This ensured that the two journals would be exactly in line after soldering. The merest touch of solder was used to avoid getting any on the journals themselves or in the slot for the die blocks. The links were then cleaned up with carborundum paper and rotary sanding drums to remove all the black oxide caused by the soldering. That's one problem with silver soldering steel - the metal takes a lot of cleaning up afterwards. I am not sure if you could pickle it in citric acid but it's not recommended to use sulphuric acid for steel.

After cleaning up the links were tried in place and, as I suspected might happen, the bottom of the link fouled the connecting rod when the rod was at it's highest point. This was solved by milling a bit off the back of the inner link but I will change the final drawings and move the links and their brackets slightly further out from the frames. It would be a lot of work to do that on Helen at this stage of the game, it's easier to alter the links!

Trial fitting of the links

At this point I started thinking about the die blocks and what material to use for them. The obvious material is gauge plate I suppose but I was reading some back issues of Model Engineer a few nights ago (when I couldn't sleep!) and came across a letter in Postbag from someone who had used Nylon for the dies in a set of Joy valve gear. Apparently he had originally used steel for them but they only seemed to last a season of running. He then replaced them with Nylon and they were still going strong after 5 years! So, why not make them from more of the Peek material? I have deliberately 'beefed' up the valve gear components as much as I can to give a long life and the dies for the expansion links are 3/16" 'square' by 3/32" thick so the bearing surface is fairly substantial for a 2-1/2" loco (For example, the dies for the Flying Scotsman are only 1/16" thick). I think I will give the Peek a try and see what happens. It's certainly much easier to work with!

In theory, if the valve gear is well designed and built properly there should be little, if any, movement between the dies and the links (known as die slip) except when the cut-off is altered by raising and lowering the radius rod. Some valve gears are more prone to die slip than others. Walschaerts gear that has the Southern type lifting link for the radius rod in front of the expansion links will cause the radius rod to rise and fall as the rod moves backwards and forwards (unless the lifting link is very long) which will make the die move slightly in the link all the time and cause wear. The LNER/LMS type with the slot in the rear of the radius rod (which I am using) should not suffer from this problem as no vertical motion is imparted to the radius rod by the lifting lever when the radius rod moves back and forth.


The radius rods were machined from 1/8" gauge plate again after the holes had been drilled and reamed in the lathe. The slots for the lifting pins were machined out with a 1/8" endmill. I deliberately left the rods slightly too long in case I have to extend the slots later if they prove to be too short for the valve travel.

The machined radius rods

The driving pins that fit in the die blocks are simple lengths of 1/8" silver steel. The holes in the radius rods were reamed slightly undersize so that these pins would be a tight fit and they were tapped into place with a spot of Loctite on them to make sure they would not come loose. These pins are often silver soldered in place but I decided against this in view of the mess it makes and the cleaning up afterwards! In any case, so long as they are a tight fit they are not going to go anywhere. The sides of the expansion links would prevent them coming out anyway if they did come loose.

Die block pins fitted to rods

Next items were the return cranks to drive the eccentric rods and again these were machined from 1/8" gauge plate. The holes were again drilled and reamed in the lathe to accurately locate them. The crankpin holes were reamed slightly undersize to make sure the cranks would be a tight fit. The bottom of the cranks were slit with a slitting saw and a hole drilled to take an 8BA clamping bolt. The bolt is a clearance fit in one side of the slit and tapped into the other side.

Return cranks after machining

The journal for the eccentric rod is turned from 3/16" silver steel and held in place by an 8BA countersink head screw put in from the back of the crank and threaded into the journal.. Loctite was again used to hold everything secure. This again avoided having to silver solder the journal in place. The screw was left sticking out the front of the journal to take the retaining nut for the eccentric rod.

Return cranks with journal and clamping bolts fitted

Once the position of the return cranks is accurately set they will be secured by a pin half in the crankpin and half in the return crank to ensure that they cannot move again and upset the valve timing.

After finishing the return cranks I had a look at the drawings of the valve gear and for an awful moment I thought I had made a cock-up! With piston valves (inside admission) the return crank follows the crankpin when the loco is moving forwards but with slide valves (outside admission) it leads the crankpin. I thought I had got it the wrong way round in which case the distance between the crankpin and the eccentric rod would have been wrong. I had visions of having to scrap the cranks and start again but fortunately I had got it right after all. Phew!!

Return Crank fitted to end of crankpin


As a bit of light relief I decided to make some proper pivot pins for the valve gear components. There are several ways to make these. They can be a plain pin made as a press fit in one part and a running fit in the other, a special turned bolt,or a pin threaded at each end to take a nut. Pressed in pins are the easiest but can work loose ( I've seen it happen!). The problem with turned bolts is getting a decent finish on the actual bearing surface and with the threaded pins the problem is threading both ends due to the very short length of the pins. Most of the pins in Helen's valve gear are only 1/4" long!

I decided to use a mixture of all the methods! The actual pins are simple lengths of silver steel which gives a good bearing surface. These are drilled and tapped 8BA all the way through and threaded onto an 8BA bolt which secures the pin in place using a nut on the outside. The head of the bolt stops the pin coming out once fitted. The bolts for the combination lever pins are countersunk head as there is limited clearance behind the top pins. The bolts for the conjugated lever link pins are normal hex head.

built up pivot pins for the valve gear

Just a quick apology about some of the latest photos - it's getting difficult to get decent photos due to the lack of natural light at this time of the year so a lot of the photos have been taken using artificial light. My digital camera is a fairly basic one and it struggles to focus properly sometimes!


As mentioned previously I decided to make the die blocks out of some more of the PEEK material. A piece of the 10mm rod was drilled and reamed 1/8" to fit the pins on the radius rods and then four 3/32" thick slices parted off. The blanks were then trimmed to shape by mounting them on the same jig used for the expansion links which was mounted on the rotary table in the micro-mill. This enabled the curved sides of the dies to be cut to the same radius as the slots in the expansion links. The top and bottom of the dies were then cut to size in the same jig. This job went very quickly as the PEEK is so easy to work with.

Milling the die blocks to size in the micro-mill

The die blocks were milled very slightly oversize and carefully fitted to the expansion links to give a nice fit without any play.

A trial assembly of the valve gear so far showed that the slots in the ends of the radius rods were plenty long enough so the ends of the rods were finally trimmed and shaped. The trial assembly also showed that the travel of the dies along the expansion link slot was not quite enough to give the calculated valve travel due to the edges of the rods fouling the spacers at the top and bottom of the links. This was solved by filing a bit off each spacer and also milling a cut out in the top and bottom of the rods where the die blocks fit.

Finished die blocks with the radius rod

Previous Page Next Page

Page 1 2 3 4 5 6 7 8 9 10 11 13 14 15 16 17 18 19 20 21 22 23 24 25 26