Winson 5" Gauge Britannia


After the GWR Mogul I vowed never to touch another Winson or Modelworks kit but I was asked to have a look at this Winson Britannia chassis that the present owner couldn't get to run. He had been trying for ages and eventually had enough of it so asked me to have a go.

Anyone who has dealt with one of these kits will probably know about the Britannia Builder website where Johnny Johnson describes his build of a Modelworks kit and the various problems that he found and overcame during the build. It's essential reading for anyone trying to complete a part built set of kits.

The one that I have lumbered myself with is the earlier Winson version and there are differences between the two. A lot of the Winson castings are aluminium whereas the Modelworks uses more bronze castings. There also seem to be a few design differences.

This isn't going to be a blow by blow account of what I've had to do to the chassis. I didn't originally intend to record it at all so I didn't take any photos at the beginning but someone may find it of interest.


The first thing is that it weighs a ton compared to other locos that I've worked on and proved too heavy to carry into the workshop single handed so it stayed on one of the kitchen worktops to begin with! The finish on the motion was absolutely awful because whoever built it in the first place just fitted the motion as it came from the laser cutters and hadn't done any cleaning up of the parts at all. None of the burs etc. had been removed. I decided to strip all the valve gear off the chassis anyway so that I could get some accurate dimensions to put in the valve gear simulator and see what was what so I took the coupling and connecting rods off as well.

The grate is pristine and looks as though it has never seen a fire. Maybe the loco has never been steamed before? I don't have the boiler etc. for it so I can't say for certain.

I've never been very impressed with the laser cut finish on thick components such as coupling rods as it's usually dreadful. I think that the parts are cut very quickly to keep the cutting time down and the finish ends up very rough indeed. I'm sure that the finish would be much better if the cutters used a slower cutting speed but time is money I suppose. Personally, I think I would go for water jet cutting for anything cut from thick materials. You don't get the glass hard finish with water jet cutting either. Unfortunately, it is more expensive.

The slidebars had been milled to shape from flat steel bar and the finish on these was terrible as well. Also the spacers between the ends of the slidebars were two strips of thin steel instead of one solid piece. The crossheads were a pretty crap fit anyway so the gap between the bars needed adjusting. Consequently the slide bars and the aluminium castings that support them were removed for attention.

The first job that I did was clean up the coupling rods. I ordered a couple of packs of the small sanding drums that you use in a Dremel type mini drill from Banggood and set to work. I find these sanding drums really good for cleaning up rods etc and I like the satin finish that you end up with. I started off with 60 grit where necessary and then used 120 and 320 until the surface was reasonable. The drums don't last very long but they are cheap so that's not really a problem. I finished off with various grades of Garryflex sanding blocks that I also find very good.

The connecting rods were really bad and took a lot of work to get looking reasonable. I ended up using a narrow grinding stone to get into the flutes. The finish from the endmill used to cut the flutes was particularly bad on one of the rods and I actually remachined it. They are not perfect but look a lot better than they did!

I am not sure but all the bushes in the rods look as though they might be made from brass rather than Bronze. Anyone considering doing some serious running with one of these would be better to replace them all if that is the case.

The slidebars were given similar treatment and new spacers made for the ends to get a nice sliding fit for the crossheads. Before making these new spacers I remachined the running surfaces of the crossheads to ensure that they were parallel to the piston rods. The builder had done a bit of filing on the top surfaces which resulted in them not being flat anymore.

The casting for the right hand slidebar support had two sheared off bolts in the face that bolts to the frame so these had to be removed. Fortunately, they came out quite easily by drilling them and using a small 'Easyout' to unscrew them. I had acquired a selection of good quality ones from somewhere and this was the first time that I had used them. Don't ever buy cheap versions of these as they are liable to break off when you try using them and you end up worse off than you were before!


Drilling the sheared bolts


Bolt removed with 'Easyout'


I also took the oportunity to drill and tap the mounting holes a bit deeper to take longer bolts as the originals only went into the casting about 0.125" which I didn't consider enough given the load that is on the slidebars.

One thing that surprised me is that all the fixings seem to be BA. The later Modelworks kits all used metric and I was expecting this to be the same.

When I came to refit the right hand slidebars it was obvious that they were too high in relation to the piston rod and I had to fit a 0.025" shim between the top slidebar and the casting to line everything up properly. I could have elongated the bolt holes in the frames to lower them I suppose but that seemed a bit of a bodge way of doing it. Also, the slidebars were too far out from the frames so I had to machine some off the back of the support casting to move them further in. When all this was done the crosshead slid nicely from one end to the other without any binding.

The left hand slidebar support seemed fine and the slidebars on this side lined up without any shims or machining. I did tap the mounting holes deeper though for longer bolts as on the right hand side.


Chassis after rods and slidebars refitted



Note that I won't be doing any repainting on this job. It will be purely mechanical.

You may have noticed that the chassis has appeared on a stand. It just wasn't practical to work on it on the kitchen worktop so I cobbled up a movable stand using scrap materials that I had lying around. I was given four large diameter casters some years ago and these were ideal for this job. A couple of hours with some scrap wood and two bits of Dexion angle and job done. It's a bit rough and ready but it makes the job so much easier.

When looking at the cylinders I noticed that the pistons are about 0.020" smaller in diameter than the cylinder bores and so rely on the O rings to centre them. This was the same with the Mogul and I made new pistons for that that were a proper fit in the bores. The piston should be a good running fit in the bore so that the loads from the connecting rod when running are carried by the piston and the crosshead. If the piston is a loose fit then some of the loads are liable to be taken by the piston rod and the packing gland in the rear cylinder cover. This is not good as the piston rod may get scored and the gland will wear rapidly (you see this quite often on locos that have done a bit of running). Ideally, the piston rod gland should be floating so that all it does is seal the piston rod. However, this time I'm going to leave the pistons as they are and see how they go. After all, I am only supposed to be sorting out the valve gear, not the whole chassis!

The steam passages from the valve ports to the end of the cylinder bore are tiny! They are just a single hole of maybe 0.125" diameter which seems far too small to me considering that the cylinder bore is 1.720" diameter. I wonder how much that will affect the performance? The Mogul was better in this respect - at least that had two holes!

The piston valves are the original Winson design but have been fitted with PTFE heads. I was told that the loco was fitted with these when the latest owner got it and possibly they may be the ones that Doug Hewson sold as replacements for the really crap brass rings that came with the kits. Unfortunately, the PTFE heads have been damaged and are pretty rough with the edges rounded off so they will have to be replaced. That could be part of the problem with the loco not running as the timing will be miles out. The timing is determined by the edges of the rings and they need to be square to ensure the accuracy. The rounded edges on the fitted valves will make a nonsense of any accurate valve timing.



While I am working on the valves I will fit my favourate design of bobbin to make setting the valves much easier (see Helen Longish and the 75000 valves for more detail). With the present design you have to adjust nuts on both sides of the valves which is a real pain to do. With the design that I use the setting is all done from the front end of the valve.


I've been making a bit more progress on the chassis.

I decided to make a new weighshaft as the original one was a bit of a bodge. I am not sure if this is how it came with the kit ( I would hope not!) but it is not a straight shaft. Instead, it has been cut and the centre section has been offset, presumably to stop it fouling the underside of the boiler barrel. It had been very roughly welded together and was shaped like a dog's hind leg! I made a new one by machining the centre section from some flat bar, silver soldering in the actual weighshaft made from silver steel and then cutting out the centre of the bar. A bit of tidying up and shaping and job done. The shaft still needs to be cut to length but I'll do that when I rebuild the valve gear and can check what length it should be.


Old versus new weighshaft


I've had a look at various chassis on the Station Road Steam website archive and I can't see any that have had this modification done to the weighshaft so I don't know if it is necessary or not. I don't have the boiler so can't check so have to assume that it is.

I was talking to a friend at the club the other day and she has a complete set of kits for the Britannia and told me that the Winson/Modelworks design is a copy of the Norman Spinks Britannia. That doesn't surprise me as their 14XX tank loco is a copy of the Neville Evans Didcot. Their 9F is probably a copy of the Les Warnett one.

I don't have a copy of the Norman Spink drawings but fortunately a chap building one to these drawings ( Steve Sharman) has a website and has put photos of the various parts he has made on top of the drawings so I can check some of the valve gear dimensions against the drawings. It's obvious from the drawings that the Winson design is indeed a copy.

The motion brackets that support the weighshaft and the expansion links are again aluminium castings and one of them had two bolts sheared off in it. These were again removed successfully by drilling them and using an Easyout. I think that the fixing bolts had been too long and had bottomed out before they were tight to the frames and the builder had sheared the bolts trying to get them tight. The weighshaft bushes had been poorly fitted and one was loose so they were removed, cleaned up and then refitted with Loctite.

The expansion links didn't look too bad but I gave them a good clean up to make them look a bit better. There are some deep scratches in them though that I can't really get out.

They are supported by two trunnions that bolt to either side of the link. The bolts were a bit of a sloppy fit through the expansion links so I drilled a 0.125" hole through the shafts of the trunnions in the lathe so that I could use a 0.125" bar through both trunnions to get them lined up with each other when they were refitted.

The original die blocks were brass and a terrible fit in the links so I replaced them with some new roller ones from hardened 0.25" silver steel. These will run on a 4mm pin in the end of the radius rods. The original brass ones had a pin on each side that ran in the holes in the radius rod and the only way to fit them was to force open the ends of the radius rods and fit them over the die block. You then have to squeeze the ends of the rods back together again. This is how they are shown on the Spink drawings as well but I think it's a terrible way to do it. Hence my separate pins.


The trunnion pins seem to be a reasonable fit in the bushes in the expansion link brackets and the bushes in the tail of the links seem ok so I'll leave those alone. I did have to fettle the ends of the slots in the expansion links to get the new die blocks to run smoothly right to the ends of the slots but it didn't need very much taking off.


The next items to be tackled were the radius rods. The originals were terrible and I decided that it would be easier to make new ones rather than try and fettle those. Also, as mentioned above, I wanted to change the method of fitting the expansion link die blocks to the ends of the radius rods.

I compared my measured dimensions of the original rods to the drawings and they were pretty much spot on. I ordered some 3/8" square mild steel as I hadn't any in stock and made a start.

The holes for the various pins were drilled and reamed in the mill using the DRO to get the centres exact. The sides were then milled to shape by bolting the rod blanks to a piece of flat bar held in the machine vice to act as a jig.



This is the sort of job where the power feed for the x axis would have been very useful but I haven't fitted it yet! Lots of horrible needle sharp chips as well.



The 'jig' was then modified by drilling and tapping two holes in the top so that the rods could be mounted on the top for maching the parallel sides between the two holes at the expansion link end of the rods. I made some buttons from hardened silver steel to prevent me machining too much away!

To do the tapered part of the rods I drilled and tapped two more holes but this time the hole to hold the combination lever end of the rod was moved forward by 0.05". This made the rod at an angle to the axis of the jig so that more metal was removed at this end of the cut. The offset of the holes was taken from a CAD drawing that I made of the rod.



More horrible needle sharp chips!



The next job is to round off the ends of the rods by hand using filing buttons. The rods will then need fluting and the slots cutting in the ends.


Another year bites the dust. They seem to go faster and faster!

I decided to do the fluting on the radius rods first while the rods were still solid and easier to hold. They were bolted to the jig again and the flutes cut with a small Woodruff cutter. This is one that I have used before for this purpose and has the corners of the cutting edges rounded off slightly so that the bottom corners of the flutes have a slight radius to them rather than a sharp corner. The jig was angled slightly to make the top and bottom edges of the rods parallel to the bed so that the flutes came out tapered the same as the profile of the rod.

The ends of the rods were then rounded off by hand filing using the silver steel buttons used before as filing buttons.

The last operation was the slotting of the ends of the rods and this was done with a 4mm carbide end mill. The DRO was used to measure both the length and the width of the slots. A bit of cleaning up and job done.


New rods versus the old at top


The old rods don't look that bad in the photo but they are a lot worse than they look.

The pins for the new die blocks are short lengths of 4mm silver steel with a short thread on one end and a screwdriver slot in the other. One side of the radius rod has a 4mm clearance hole and the other side is threaded 4mm to secure the pin in position.

I don't think I've mentioned before that most of the pivot pins in the valve gear are a pretty sloppy fit in their holes in the rods etc. Some of the pins are undersize and some of the holes are oversize. To remedy this, I'm going to ream out all the holes in the valve gear to the next size up and fit hollow silver steel pins secured by a nut and bolt through them. The original holes are supposed to be 5mm so I'll be opening them out to 5.5mm. I didn't have a 5.5mm reamer so got one off Ebay for a reasonable price and also some 5.5mm silver steel which I didn't have in stock either.

To ream the holes, I clamp the particular part to the milling table ( I used the small mini mill for this ) and then use a length of 5mm silver steel in the mill chuck to locate the hole centrally. The rod is then replaced by the 5.5mm reamer and the hole reamed out. The photos below show this being done to one of the combination levers.




Yes, I need to get a proper clamping set for the Mini Mill! I got one for the large Mill soon after I got it but never bothered to get one for the Mini.

Hopefully, the rest of the valve gear will just need cleaning up and then it can be reassembled with new pins.

The only items that should need reworking are the eccentric rods which need shortening to correct the valve events. As mentioned at the beginning, I had measured up the valve gear and put the dimensions in the Wallace valve gear simulator to see what the valve events were like. They were not very good but reducing the eccentric rods by 0.075" makes a huge difference and the valve events become very acceptable in forward and reverse gear.

The eccentric rods on this loco are 5.475" between centres which is close to the drawing dimension of 5.460". However, according to my analysis, they need to be 5.400" between centres. I don't have the drawing dimensions for the position of the expansion link brackets though and only have my dimensions measured from the chassis. It's possible that the brackets on the chassis are not correct to the drawings.

I also noticed from the dimensions of the combination lever, which agree with the drawings, and the dimensions of the valve bobbin, which also agree with the drawings, that the lead seems excessive at 0.0428". I think that I will reduce this to say 0.020" when I make the new valves by making the valve lap longer by 0.028".


I spent a bit of time today making a start on cleaning up the combination levers. These are quite fiddly with lots of curves and straight bits. I was halfway through doing the first one when I noticed that it didn't look straight. Putting a ruler against the edges of the top two holes showed that the bottom hole was out of line by about 0.25"! I checked the drawing to make sure that this wasn't intentional but all the three holes should be in line as I expected. Putting the top of the lever in the vice and hitting the other end with my 'precision' soft hammer soon had it back as it should be.

When I checked the other one, it was even worse!



I'm not sure what has happened here. Either someone has deliberately bent them to try and change the valve timing or there's been an almightly crash when someone has tried to run the chassis on air. I did notice when I dismantled the valve gear that the retaining plates on the top of the valve rod guides are bent upwards on both sides. I did wonder at the time if something had hit something else and caused the damage.



Looking at the photo above of the bent combination lever it has been bent backwards. I wonder if the bottom end has hit the rear cylinder cover? Oh well, I may find out when I fit everything back together!

(no, I've just checked and the bottom of the combination lever shouldn't come anywhere near the cylinder cover)


Previous Page........... Next Page