3½" Gauge Clarkson Princess Royal



A couple of months ago a couple turned up at the club on one of our thursday work days and brought along a very nice looking Princess Royal, which I recognised as the design by Clarksons. They had had this loco on the sideboard for 30 years after buying it from the estate of the deceased builder (I presume). The builder had been a miner and the loco had apparentally only been steamed the once before the builder passed on. The new owners had decided that it would be nice to run the loco again and had brought it to the club to have it looked at with a view to getting it running again. It's only a two cylinder version and I presume that the cylinders had been enlarged to compensate for this. The motion work is all very well made with no wear as would be expected.

One of our club boiler inspectors had tried to do a bit of a pressure test on the boiler but all the fittings leaked very badly and it was impossible to build up any pressure. At this point, I arrived at the club and one of the members approached me and asked if I fancied having a look and sorting it out. Being a glutton for punishment and always willing to help, I said ok but it would have to wait for a bit as I had other jobs on that needed finishing first. The owners were quite happy to wait and about a month ago I brought it home.

Unfortunately, I didn't get any photos before I started to attack the loco but will get some as it's reassembled.

The loco had no boiler certificates so I decided that it would be best to give it a full hydraulic test to twice working pressure, just to be on the safe side. The easiest way to do this was to remove the boiler from the chassis and remove the fittings. Removing a boiler always tends to be a major operation and this was no exception! You never know what needs to be undone or removed before the thing comes free. Anyway, I got it off eventually. This made the chassis a lot easier to handle as well, as the whole loco is quite a lump.


Once the boiler was free, I removed what fittings I could, fitted blanking plugs, and fitted a gauge and pump. The boiler was filled with water and I started pumping. Nothing much happened except for a lot of water escaping from the steam pipe header on the front tubeplate. That didn't surprise me as I suspected that the regulator would leak. I guessed it was probably a dreaded disc in tube type and I wasn't wrong! To be fair, it hadn't been used for 30 odd years so could have been corroded or similar.

A set of drawings would have been useful so I could see how the regulator was constructed but I assumed it all fitted in from the front tubeplate as the backhead bush was a solid bush silver soldered in with just a gland for the regulator rod. The steam pipe header on the front tubeplate was an oval casting held on with two nuts and I assumed that this also formed the regulator body. I removed the two nuts but it was a right struggle to remove the casting. It appeared to be sealed to the tubeplate (there is no separate bush) with something akin to Araldite and was well and truly stuck. I eventually got it free without damaging anything but I think the builder had had a job to seal the joint, hence the liberal quantity of 'goo'


Front tubeplate showing regulator/steam header position

The regulator turned out to be the usual affair with a flat disc with holes in 'seating' on the regulator casting mounted on the end of a tube. The regulator rod runs through the tube and has a spring to press the disc onto the regulator port face.

Regulator casting and disc valve


Both the port face and the disc didn't look very flat so I carefully lapped them on some fine carborundum paper on the surface plate. The disc actually had quite a deep score where the builder had scribed a circle to mark the position of the four holes which would have caused the regulator to leak even if the surfaces had been perfectly flat.

I thought that getting the regulator body to seal on the tubeplate again could be a bit of a nightmare so I modified the casting to take an O ring to take care of that. I mounted the casting in the lathe and cut a groove with a parting tool. This meant shortening the outer tube slightly as well to allow for the O ring.

Regulator body modified to take an O ring seal

The regulator was reassembled and was ready to put back. I did notice though that one of the studs on the tubeplate was very loose in it's thread which I wasn't happy about. There wasn't enough room to re-thread the tubeplate and fit a larger stud so I made up a special bronze sleeve nut that fitted on the stud behind the tubeplate and made sure that it could not pull out.


Bronze sleeve nut to secure stud

The regulator was then refitted and we were ready to have another go at a pressure test. This time I managed 40psi before the dreaded drips appeared! Not from the regulator this time but from one of the longitudinal stay nipples on the front tubeplate. I assumed that these would be the usual double threaded type beloved by LBSC etc. so I hoped it would just be a case of unscrewing the nipple and resealing it. Wrong! When I tried to unscrew the nipple, the one on the backhead turned as well which did not bode well. I got the nipple off by putting a spanner on the backhead one to stop it turning and discovered that the threads in both the backhead and the front tubeplate were completely stripped. They are tapped straight into the copper without bushes and probably stripped when the stay was first fitted. It makes you wonder if the stay was ever steamtight?

Longitudinal stay with nipple and stripped threads in tubeplate

Another revelation was that the stay itself is stainless steel which is not a good idea in a copper boiler due to electrolytic action between the dissimilar metals. Not so bad for fittings that only see steam maybe but not good where the stainless will be in the boiler water. I decided that the best course of action was to replace both stays with bronze and make new oversize nipples and retap the boiler plates. With this in mind I removed the second stay. The threads were ok on this one but the stainless was showing signs of rust pitting in several places.

I put this job to one side as I needed to order some taps and dies to make the new nipples and some 1/4" bronze bar for the new stays.

Whilst waiting for the taps etc. I tried running the chassis on air. It ran beautifully in forward and reverse but had a constant blow from the exhaust. This could either be a valve leaking or the piston packing leaking. I disconnected both valve rods so I could move the valves by hand and the blow seemed to be mainly from the lefthand cylinder. Even when the valve was central over the ports, closing them off, the blow was still there. This suggested the culprit was the valve so off came the valve chest cover and the valve chest. Getting the valve chest off was a pain as you have to dismantle the steam pipes to get everything free. They are the usual concoction of threaded tees and pipes and are a nightmare!

Once the steamchest was off, it was easy to see from the wear on the valve and portface that the valve was only touching in certain areas so no wonder it leaked. In fact, the valve actually rocked slightly on the portface.

Valve and portface showing uneven contact

It turns out that the portface is actually slightly convex. It's too much to cure by lapping so the cylinder will have to come off and the surface skimmed again. Once I got the cylinder off, I did give it a quick rub on some carborundum paper but it definitely won't clean up that way. There's some quite deep scores on the face and the ports are not too clever either. It looks like the builder make a mistake with the ports as some holes have been filled in. I think I'll try cleaning the port edges up in the mill and then alter or remake the valve to suit.

Port face after initial rub on carborundum paper

When I was running the chassis on air I noticed that the oil wasn't going down at all in the lubricator so I disconnected the feed pipe from the union where it goes into the steam pipe. Nothing came out of there when the lubricator was turned but I noticed that the union was completely blocked with gasket goo. Even if the lubricator had worked, the cylinders would not have got any oil. Maybe that's one reason the port face was scored badly. Also, the end of the oil pipe that connects to the check valve on the oil pump was loose and wouldn't tighten up anymore so that would have leaked badly. I'm beginning to wonder if this loco has ever run successfully at all.


That's pretty much progress to date. Works stopped at the moment due to other jobs (more on the 94XX tank, and finishing a couple of Tich boilers)


Finally managed to get back on this! I decided to try and get the boiler sorted first before doing any work on the chassis. First job was to give the boiler a close inspection and it was then that I noticed that the front tubeplate in the firebox had a distinct bulge between the tubes and the foundation ring. It was easy to see why - there were no stays between the tubeplate and the throatplate leaving a large unsupported area! You can perhaps just see this in the photo below:


If the boiler was going to be used, the only thing was to fit some stays in this area. I tapped the bulge flat with a soft hammer and decided to fit three threaded stays which would be rivetted over and then sealed with soft solder (This is how the firebox stays are done). Unfortunately, this meant removing the boiler cleading which was all soft soldered together in situ. This was removed fairly easily with the gas torch but will be a pig to fit back again. Also it will have to be repainted! This did give me chance to inspect the outside of the boiler which, although not a shining example of the boiler makers art, seems sound enough. It's mainly brazed? construction with soft solder caulked stays. This would have been the normal method of construction when the loco was built. I would imagine that the soft solder will be ordinary 'tinmans' or similar low melting point stuff. They didn't have Comsol in those days!

The original stays are 3/16 inch diameter copper so I drilled through the tubeplate and the throatplate and tapped the holes 2BA. This was not easy as the holes had to go through both plates at an angle, hence the reason for rivetting them over rather than fitting the usual brass nuts. I should say here that the boiler differs from the original design in that it does not have a combustion chamber and the firebox tube plate is not parallel to the throatplate as it should be. This is why the holes had to be drilled at an angle to the plates.

The next problem was that I didn't have any 3/16 inch copper rod so I finished up turning down some 1/4 inch square bar! I did have some 3/16 inch phosphor bronze but this would have been very difficult to rivet over.

The new stays were threaded in, cut to length, and then rivetted over. Finally, they were sealed with high melting point soft solder.

New stays in throatplate/tubeplate

The next job was to make and fit the two new longitudinal stays. These are 1/4 inch phosphor bronze bar with bronze nipples at each end which screw into the backhead and front tubeplate. Again, this would be the norm when the boiler was built. This went fine except that I had to make the nipples twice. I made the mistake of threading the outside of the nipples before tapping the holes in the plates. The first set of nipples were too loose a fit in the threaded holes so I had to remake them!

The new stays fitted very easily and I decided to seal the nipples with soft solder again. They won't need to come out again (hopefully!)

Tomorrow I'll give the boiler another hydraulic test and we'll see what happens. I think the boiler may have to be derated to something like 70psi working pressure as the firebox sidestays are a little too far apart for my liking to run at 80psi. There's quite a large unstayed area at the front of the firebox wrapper. The problem is that the original builder deviated from the original drawings as mentioned above. The builder had not only left out the combustion chamber but also shortened the inner firebox for some reason. This left a big gap between the front of the inner firebox and the throatplate which meant that no stays could be fitted to the firebox outer wrapper in this area (there was nothing to fasten the inner end to!). It wasn't possible to fit any full width stays between the sides of the firebox outer wrapper ( like the cross stays over the top of the firebox) as the tubes were in the way. However, the builder had increased the thickness of the firebox wrapper from 3/32" to 1/8" which helped make up for the lack of stays. It could probably do with another cross stay fitting at the front of the firebox wrapper in line with the originals but I'll show the boiler to our club inspector first and see what he thinks.


Well, I gave the boiler a hydraulic test yesterday and what a disaster that turned out to be! It leaked like a sieve. I think the only stays that didn't leak were the new ones that I've just fitted. I should have expected it really as all the soft soldering was very poorly done. I don't think the metal had been cleaned up enough before the soldering. It's times like these that you feel like giving up but after a cup of tea and a bit of a break I started the job of cleaning up all the joints and resoldering them. Cleaning was very difficult and I resorted to heating the old solder, cleaning it off, and then wire brushing the area, finishing off with a sanding drum in the Dremel. The difficult bit was the area just around the protruding stay head and I finished up using a diamond point in the dremel. Using plenty of acid flux I managed to resolder all the firebox side stays but I only did the outside of these to begin with. I guessed that some would still leak inside the firebox but one thing at a time!

Another test showed that the joints I'd redone were dry but some of the nipples on the cross stays were now leaking. I actually managed to easily remove all but two of the nipples without melting the original solder so you can tell how bad the soldering was. Cleaning up was easy this time - a sand with the dremel on the firebox wrapper and a file on the heads of the nipples. The nipples were then refitted and soldered again.

The next test showed these to be dry (getting there!) but there were a couple of drips from the front corners of the foundation ring. A close look showed a couple of very small pin holes in the brazing. These had been soft soldered, but again, poorly. These were sorted as well as a slight weep on the top of the firebox wrapper where a rivet had been used to hold the wrapper to the throatplate during construction.

There was also a drip from the regulator flange where it bolted onto the front tube plate but I wasn't too fussed about that at the moment.

The next test was much better. I managed to get up to 150psi without any trouble but, as I expected, some of the stays inside the firebox were weeping. I decided that there was no way I would be able to get these clean enough to resolder so decided to try the Loctite solution. I cleaned up the inside of the firebox as best I could with a wire brush and then drew a vacuum in the boiler using a large aquarium air pump used to suck instead of blow. A drop of Loctite 290 ( a thin wicking grade) was then applied to every stay head (may as well do all of them just in case) and left to hopefully soak into the cracks. I'll leave the boiler overnight and then try another test (yes, I know it's Xmas day tomorrow!).


The Loctited stays proved to be bone dry on the next test (Phew!) and I also gave the regulator flange the same treatment as I didn't really want to have to take it off again. That did the trick and sealed it. A couple of very small pin holes appeared on the top of the firebox again so I resolderd those. One of the stays on the lower backhead weeped slightly (I hadn't resoldered these) so that got the Loctite treatment as well which cured it. None of these later 'leaks' were serious and would have probably sealed themselves eventually. The boiler was now bone dry at 160psi. Thank heavens for that!


Next job is sorting out the cylinders and valves.

I'd just like to mention that I received an email from Martyn Myer in Australia after he had seen this page and he very kindly sent me copies of the original H P Jackson Drawings. Martin is renovating a 4 cylinder version of Princess Royal. Many thanks Martin, the drawings have been a great help.


Previous Page Next Page