3

The two rectangular holes for the conjugated valve gear arms were cut out by chain drilling and filing to size. For appearance sake I also reduced the thickness of the frames to 1/16" where they protrude through the running boards at the front of the engine.

I could now think about assembling the frames, hopefully once and for all. To help keep everything nice and square I cut two pieces of steel angle and wedged them inside the front and rear horns using nuts and bolts with some packing to avoid damaging the soft gunmetal. Unscrewing the nut and bolt forced the angle against the machined face of the horns and ensured that both sides were perfectly in line. All the stretchers were fitted but the fixing bolts were left loose for the time being. Incidently, I don't drill and tap all the bolt holes in the stretchers straight away, just two each side, then if any adjustments to the hole positions are needed, it's easy to draw over the holes in the frames with a needle file. The holes for the remaining bolts are drilled and tapped in situ when the frames are completely assembled and I'm satisfied that everything is ok.

Using angle to line up the horns

Starting with the front buffer beam block, the bolts were tightened and the alignment of the horns checked with a square to make sure the frames remained square. I then worked towards the back of the frames, tightening the stretcher bolts one by one and repeatably checking the alignment. All was well until I reached the vertical stretcher behind the trailing axle. Tightening this stretcher caused the frame to twist slightly. This was found to be caused by one of the fixing holes being slightly out but was soon remedied by elongating the frame hole slightly. One of the fixing holes for the long horizontal stretcher was also slightly out of line. After all this I finally had a set of frames that appeared to be square and true and everything went much easier than I had expected. All that remains now is to drill and tap the remaining holes and fit the proper bolts and screws instead of the temporary ones used. I save all the old tatty screws and bolts for inital assembly as usually things have to be taken apart several times. This saves damaging brand new bolts etc.

Assembled frames

Quite a bit of time was taken drilling and tapping the remaining holes in the stretchers. Any fixings which would come behind the wheels were done with countersink head crews to avoid bolt heads fouling the wheels. I made up a little countersink tool from 3/16" inch diameter silver steel to make a slightly recessed countersink for the 6BA screws. This was made with a little pin on the end the same diameter as the clearing hole and this 'pilot' ensured that the countersink was truly concentric to the hole.

Homemade countersink bit with pilot pin

The hornstays were made next from 1/2" x 1/8" steel. They actually need to be 7/16" wide so the width was reduced by endmilling in the lathe. When doing repetitive jobs like this it's useful and time saving to set up a little production line process. LBSC was always describing ways of doing this in his articles. For instance, if you can provide a positive and repeatable method of positioning the components then it's a simple matter to machine them all to the same size using the micrometer dials on the machine used. For example, I roughly cut all eight hornstays to length with a hacksaw then milled one end of each square in the lathe using the vertical slide. I then turned the stay around with the milled end against a stop and milled the other end to length using the leadscrew to advance the cut. The micrometer reading was noted and the other seven milled to the same setting. The same procedure was used to reduce the width. One stay was then marked out for the fixing bolts and spring pins and the holes drilled. This was then used as a jig to drill the other seven. The hornstays were then clamped in position on the bottom of the horns and the bolt holes spotted through into the horn. The horns were then drilled and tapped 6BA for the fixing bolts. Thus, only one set of holes had to be marked out instead of 16 ( 8 stays and 8 horns ) and all components and hole positions are exactly the same.

Frames with hornstays fitted

The axleboxes are simple milling jobs from 1" x 1/2" mild steel bar. They were milled to fit the horns by clamping them in the machine vice on the vertical slide. Care was taken to make sure the bottom of the slot in the vice was square to the lathe axis and the jaws perfectly horizontal. It was then just a case of clamping each blank in the vice with the rear edge hard up against the bottom of the slot and the width reduced to leave a 1/16" thick flange. One side of all eight axleboxes were machined at the same setting and then the axleboxes turned around to machine the otherside. Each axlebox was carefully fitted to it's horn as there were slight differences in the width of the horn slots.

Milling the sides of the axleboxes

To drill and bore the axleboxes to take the bearing housings, they were soldered together in pairs using Carrs 188 solder paint. Drilling both together like this ensured the axles would be square across the frames and not end up at an angle.

A pair of axleboxes soldered together

One face of the 'block' was carefully marked out to find the centre and lightly centre popped. It was then mounted in the 4 jaw and adjusted until the centre pop ran true using a point held in the drill chuck.

Centreing the axlebox for drilling

The boxes were then drilled out in stages to the largest drill size I have ( 1/2" ) and finally bored to take the bearing tubes. The bores were made an easy fit on the tubes as they will finally be secured with Loctite. I had a few problems with the second pair coming apart a couple of times when I drilled them, probably due to being a bit heavy handed with the drill, but the rest were fine. The solder used is not very strong so the joint is easily parted if you are not careful.

 

Pair of axleboxes after boring

 

Next job on the axleboxes was to radius the inside of the flanges to allow the boxes to tilt when running on an uneven track. I decided to do this by milling in the micro mill using the rotary table. The problem was to get a large enough radius using the rotary table which was only 6" diameter. I eventually managed this by bolting the 9" faceplate from the lathe onto the rotary table and then bolting the Myford machine vice onto the edge of the faceplate. This gave me a radius of about 4-1/2" which was sufficient. Not the most rigid of setups but by taking light cuts the job was done ok. There is not much metal to remove anyway.

Setup for radiusing the axlebox flanges

The holes for the axlebox spring pins were spotted through from the hornstays and drilled and tapped 7BA to take the 3/32" dia. pins. These were made from stainless steel but unfortunately I only had enough to make 8 of the 16 pins so the rest will have to wait until further stocks arrive.

I decided to complete the assembly of the leading axle at this point as this would be the most complicated of the four. The two short bearing housings were glued into their hornblocks using Loctite and the oil/dust seals ( simple O rings ) put in place. I had already decided to fit the wheels to the axles with Loctite rather than use a press fit so the crank axle wheel seats were reduced slightly in diameter to an easy push fit in the wheels. I also decided to remove the centre part of the axle ( left in while the crank axle was assembled ) now rather than after the wheels had been fitted.

Next job was fitting the wheels and getting the quartering ( thirding? ) correct. I used the same method as for the Flying Scotsman i.e. mounting the axle between centres and using accurate pillars to set the height of the crankpins. I draw the layout of the crankpins in AutoCad and calculated the height the pillars would need to be above the top of the cross slide. As it happens, I was able to use the pillars I had made for the Scotsman as the crankpins are the same diameter. All I had to do was extend the pillars for the outer crankpins by exactly 1/16". The main difference is that for Helen all three angles are 120 degrees as the middle cylinder is in line with the motion centreline instead of being at an angle as in the Scotsman.

At the end of the day it's not that important to get the angles exactly 120 degrees and a degree or two either way won't make any difference. The important thing is that all the outer crankpins are set to exactly the same angle. If not, you will never get the coupling rods to fit without binding unless you give the bushes excessive clearance. It is important to get the order of the crankpins correct when using the Gresley conjugated valve gear as the motion for the middle valve is derived from the two outer ones. When going forward, the RH crankpin should lead, followed by the LH, then the middle. Looking at the details for the original Helen, the order is RH, Middle, then LH, but she uses a seperate set of valve gear for the middle cylinder so there's no problem there.

Firstly the RH wheel was fitted to the crank axle with a pillar under the centre crankpin and one under the RH wheel crank pin. The LH wheel was then fitted using pillars under the two outside pins. The only problem was that the topslide is only just wide enough to take the pillars for the outside cranks.

Setting the centre crankpin relative to the RH wheel

Setting the LH wheel to the RH wheel

The only things to be careful of when using Loctite is not to hang about too long putting everything together ( the smaller the gaps, the faster it goes off ) and make sure you don't get any in the axleboxes! If you glue the axleboxes to the axles. you'll have to heat everything to 250 degrees C to get it apart again!

One axle down, 3 to go!

The next few days were spent cleaning up the remaining six wheels with needle files so that the other axle assemblies could be completed. The flash on the wheel spokes caused by the molten iron running along the joint of the two part mold used was very bad and took quite a few hours to remove. I recently picked up a set of six driving and coupled wheel castings off Ebay which turned out to be original Dave Goodwin castings for Black Five ( although I didn't know what they were until they arrived! ) and they were much better quality with no flash at all. Pity I didn't get them sooner as I could have used them instead. I can always build a Black Five I suppose!

In between fettling the wheel castings I was thinking about how I would construct the cylinders and spent a few hours on the computer trying different designs ( whilst the blisters from all that filing subsided! ). I had already decided to use piston valves and as no suitable castings were available, the cylinders were to be built up from seperate parts. I had earlier obtained a length of 1-1/2" dia. x 3/4" bore cored gunmetal bar for the cylinders and some 3/4" dia. gunmetal bar for the valve chests. The idea is to roughly machine the cylinder and valve chests and then silver solder them together along with mounting lugs and an external exhaust manifold. At a pinch I could probably have used the piston valve cylinders from Black Five for the outside cylinders but I wasn't sure how big I could take the bores out to with those. I would still have to fabricate the inside cylinder anyway.

I did not fancy turning seperate liners for the piston valves and wanted to use the much simpler LBSC design where the ports are turned directly into the wall of the valve chest using an internal boring tool. I drew this design out at first but wasn't too happy with the small bearing surface this gives between the valve and the bore and I can see that wear would be fairly rapid if the lubrication was not 100%. Using a seperate liner increases the bearing area somewhat due to the ports being a circle of square holes with lands inbetween rather than an open ring. My thoughts turned to the design which Henry Greenly described in ME Volume 62, issue 1516. His design uses a double ported piston valve in which each end of the cylinder has two ports, one for admission and one for exhaust, spaced a distance apart. The piston valve head is consequently longer and the bearing surface greatly increased. HG's idea was to reduce the chance of steam leaking past the valve straight from the inlet to the exhaust although looking at both drawings I can't see that there would be any difference between the two variants.

The two types of piston valve

The valve for the double ported design is a lot longer but I drew out a design that kept the overall length of the valve chest the same.

I had also been thinking of using PTFE for piston valves but decided that it would probably be a lot of trouble to use this material instead of the usual bronze or stainless steel. The big problem with PTFE is it's high coefficient of expansion which makes it difficult to get the diameter of the valve right. The valve needs to be of such a size that it is the proper running fit when the cylinders are hot and this means it needs to be quite a loose fit when cold. There is a lot of trial and error involved in getting this right. However, I was looking on the RS website and came across a material called Tecapeek PVX which is a high performance thermoplastic containing 10% Carbon, Graphite, and PTFE. This has a much lower coefficient of expansion similar to bronze or stainless steel. It sounds a very promising material and ideally suited for making piston valves. It's low friction, even with no lubrication, will withstand superheated steam, very strong at high temperatures, and easily machinable. It's not cheap but I think I will give it a try. Nothing ventured, nothing gained!

Anyway, back to the filing!

 

Previous Page Next Page

Page 1 2 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26