I've begun to assemble the things I've made so far; hopefully for good. The frames and buffer beams were not a problem after riveting the angle in place for the footplates. The first job was to fix the cylinder blocks in place, but before I could do that I had to devise a way of connecting the two together for the blast nozzle. I designed it so that there is a sliding joint on each side. This will allow the blast pipe to be able to be centralized.
Because I'd made an early change in the position of the cylinders, I'd had to drill extra holes in the frame. So instead of a simple two hole flange connecting the pipe to the cylinder, I made a full plate to cover everything up and make a steam seal with the frame. This pipe enters the cylinder at a 10 degree angle so it required some careful setting up when I silver soldered it.

The threaded end joins the sliding section with a knurled tube. The tubing I used was thick walled 16mm brass and so able to take a fine thread. Just to make life difficult, I dropped the die and it broke in two! The thread size was 5/8 x 26 tpi and the die fitted a 1" holder, so its walls were very thin. This meant either waiting for a new one or using the lathe. As I'd started it in the lathe, so I finished it.

I fabricated the centre part. The hard part being to solder the side bits in at the correct angle. Anyway, it's gone together, rather tightly, but having done it, I'm not happy with it. It's too complicated and I would prefer a better upward sweep angle. So I may well have another go. I haven't completely fitted it and I can see problems with sealing. PTFE tape sounds good but I think holding the sliding section while doing up the knurled bit is going to be difficult. Loctite might be the answer. Then there're the nuts to seal as well. Time for a re-think.

Couldn't do any more assembly until the piston rings had been made. I always make my own rings, usually because the bore size is rarely a standard size and having special ones made is expensive. Rings of this size, had I managed to bore the cylinders to a common size, were priced at a well known emporium a year ago at £12.80 + VAT each, making a total of £60.16 just for 4 rings! I find they are quite easy to make and the secret is in the tempering.

The first picture shows them being turned from a cast iron bar. The width and the thickness is roughly taken from the bore divided by 25. In this case, about 2.5mm. I always make them slightly oversize. The second picture shows them after parting off. They have been sawn (some people break them with a twist) and brought carefully to size. The way I do this is to put one edge in the vise and pinch the other up to close the gap round a fine file. By filing both sides at once you get a perfect fit. Each one can be tried in the cylinder until a good fit is obtained.

The rings are now held with a wedge between. This is 4 times the thickness of the ring i.e. 10mm. Some people bind the lot with iron wire and cover it with soap to prevent scaling. It is recommended that it is brought up to red heat and held there for 20 minutes and then allowed to cool down. I found this didn't come easily. Trying to keep a large lump the same degree of redness all over is not easy and requires a large burner and a lot of gas. So the third picture shows how I set mine up. Fortunately, I have a small kiln which I occasionally use for lost wax casting. So I put it in the kiln without all the wire and soap and brought it up to temperature. To mark the temperature, I took a small piece of aluminium sheet and folded it in an angle so that when placed on the kiln floor it could be easily seen when opening the door a crack. When this collapsed, I knew the temperature was right and it was easily kept at that temperature for the 20 minutes. The whole thing was just above dull red in colour. I have a laser thermometer which gave the temperature as 525 degrees C.(977 F)
When they came out of the kiln, now cool, I put them into the pickle which removed most scale. Wire brushed and 6 rings finished. The gap had closed very slightly. I can't claim all this as my own idea. I'm indebted to an article in 'Engineering in Miniature' in about the second copy they printed for the formulae and the info on tempering.

I make more than I need in case one is not so good and just to show how strong the rings are, the fourth picture is my dirty thumbs pulling an old discarded ring of about the same size to just about breaking point. The gap is nearly 25mm and I'm really pulling hard, so there is no worry when stretching over the end of the piston.

The assembly was going quite well until I came to fitting the pistons with the new rings. When I tried to fit the studs on the rear cover plate, I found they didn't line up. After some careful checking I found that the cylinders were on the wrong sides. Although identical mirror-wise, it meant the cover plate was on upside down with the machined side on the inside instead of the outside. Just one of those things sent to try us. Nothing for it but to start all over again. Funny how you can assemble it with bolts as a temporary measure, but when you fit the studs, they don't want to fit.

23 August 2006

Time has been in short supply this month, what with going on holiday and doing all the usual summer holiday chores. I also had to see the boiler man as it will soon be time for him to start and I had made a lot of major changes to the design after seeing the full-size photos. Most of the boilers he makes are quite simple with a tube into a round top firebox. Mine however, is a Belpaire. It took several trips to get everything right to satisfy both the boiler maker and Lloyds particularly with the welds. The good thing about the Belpaire is it allows plenty of room for the sight glass.

I have spent most of the workshop time assembling and fitting the valve gear in its permanent place. This is a surprisingly long and tedious process. I re-made several pins as I thought they were a bit too loose. The hardest part was the actual timing, finding forward and backward dead centres accurately proved quite difficult. I seemed to get a different place each time! Having replaced one pin I found this and a bit of slack where things were put together in a temporary way, was enough to show that the eccentric return crank rod was a tiny bit too short. In the end I made a complete new rod.
Before I could set up the accurate placing of the components, I had to make the reverser. Unfortunately, the full-size one is bolted to the inside of the side tank. I could do this, but it's likely the plate would be too thin to be of use. Besides that, it would have meant making a side tank first. So I've had to make a bracket and hope it's in the right place. I haven't been able to finish the steam chest and cylinders because I've yet to make the connector from the steam chest to the inlet pipe as this has to be fitted before bolting everything down. Rolling the chassis along by hand to make sure the ports opened correctly both in forward and reverse, is physically hard. Everything is so stiff. Several days of doing this seems to have loosened it quite a bit. I still have to pack the glands and then attempt to run it on air.

I decided to have a break from all this and have another go at CNC work. I needed nameplates for the Romulant and always meant to make them. Not being satisfied with the software I'd already got, I came across some new stuff at the club exhibition. I bought the software and of course had to learn how to use it. It's called 'Vcarve' and I found it very easy to learn and my existing setup worked well with it. Getting it to cut is another thing altogether. An expensive learning curve in cutters was the first thing. Lack of experience and no information on plunge depth rates and feed rates was the main cause. However, after a few broken 1mm diameter cutters, and I was happy with the result. The first efforts were cut in acrylic and were perfect, brass was a different story!
On first inspection the result looked perfect, but then I spotted one or two anomalies. Some of the problem came when changing cutters, finding Z zero accurately was difficult. Then for some reason the program jumped, so I'm told, and cut in a different place. Not sure whether I believe that, because the N & T, which are correct, were cut after the other letters and the left hand boss, followed by the holes and the outer bezel which are correct. So it must have jumped back again! Anyway, I show the result so far. Not good enough to fit on the engine.
I'd run out of half-hard brass 1/8" thick and the only source I could find was selling a one square foot piece for £35 which I thought a bit expensive, but did include postage. I bought locally at £17 a square foot. The internet doesn't always find the cheapest.

It's been more then a month since my last update but things should improve with the colder weather coming.

2 October 2006

Hopefully, I put all the bits together in the right places and having finally satisfied myself that I had the timing right. I cut up some studding and fitted the steamchest studs. Having done so, there seems to be a little bit of over-crowding. Whether I should have reduced the number or the size of the studs needs some thought. Too late now. The front endcaps were also put on using studding.
Once assembled, it was ready for testing on air. All four glands were packed using PTFE tape, everything oiled especially the cylinders. I know some people swear by 'O' rings, but when they need replacing it can mean a major stripping job whereas the tape can be replaced in situ. The discarded draincocks from the Romulant now came in useful as I'd made sure they could be fitted to this engine.

I'd kept the pipe arrangement which I'd made for testing the Romulant and only a little bending was needed to get it to fit. I used layers of masking tape as a gasket. I blocked up the front end to give the wheels clearance making sure I also put the right sized blocks in the suspension otherwise the motion would have locked.
So all I had to do was connect to the compressor. Of course, nothing happened, just the sound of air escaping from the exhaust. Much easing round gently and then turning on the air. The gland packing made everything stiff, but without it, the air would have escaped. Eventually I got the first lurch, half a turn and finally complete rotation. It still has one stiff spot for which I can't find a reason - yet. However, it still ran when the compressor pressure had dropped down to 50psi with a rather jerky action.

It became immediately obvious that I'd have to re-think the reverser. The bracket wasn't rigid enough and the reach rod flexed under load. Certainly not helped by having bends in it. Because of this, the lever on the weigh shaft kept moving backwards and forwards; not enough to stop rotation but not good enough. At first, reverse didn't work at all, so the timing wasn't quite right. It still worked forward when notched back three, so I shortened the rod by one and a half turns. Reverse worked perfectly.

First I strengthened the lever bracket by silver soldering a 5mm thick triangular gusset on the underside which stopped it twisting, at the same time as changing the 5mm bolts for 6mm. The reach rod was completely replaced from 6mm rod to 8mm. The whole lot was assembled and again tested on air. Quite quickly the stiff spot vanished and it was very satisfying as well as a relief, to find it ran both forward and in reverse.

I have now assembled the running chassis and put the bogie back. The motion work is now complete so time to think about the next phase.

11 October 2006

HOME . NEXT . BACK