I've spent the last few days trying to sort out the exhaust manifold which has proved very tricky due to the lack of space to fit all the pipes etc. in! The original idea was to bend up some 1/4" dia. copper pipe which would have flanges on one end for fastening to the frames and middle cylinder and the other end would be fitted into a central piece which would extend into the smokebox and carry the blast nozzle. After a few unsuccessful attempts at bending the pipe to a small enough radius I gave up on this idea and had another think. I am not very keen on the 'T' shaped exhaust pipes as I am sure the right angle bend where the exhaust pipes join the central blast pipe must restrict the passage of the steam but it is a very simple system. I eventually decided on a slightly refined version where the sharp angles were replaced by 45 degree ones which should ensure a smoother exit for the exhaust steam and reduce the back pressure.

The connections to the outside cylinders are made by bolting the exhaust pipes to the inside of the frames and the steam passes from the cylinder to the pipes through a hole on the frame, the connections being sealed with 'O' rings. To ensure that the inside pipes would line up properly the flanges and central piece of pipe were machined in one piece from brass bar. Two holes were then machined at 45 degrees at each end of this central piece to take two short pieces of 1/4" copper pipe to connect to the central collector/blast pipe. Each end was recessed to take an 'O' ring to seal the ends against the frames.

Machining the 45 degree holes for the copper pipes

Four holes were drilled around the exhaust hole on the frame to take 8BA countersink head screws to bolt the flange of the exhaust pipe in place. The central exhaust connector was then clamped in place and the holes spotted through into the flanges which were then drilled and tapped for the screws. The connector was then sawn into two seperate pieces and the ends chamfered at 45 degrees to match the copper pipes and to give clearance for the middle cylinder valve rod.A flanged pipe connector was then turned up to fit the exhaust connection on the middle cylinder and also drilled at 45 degrees to accept another piece of 1/4" pipe.

The collector/blastpipe was then turned up from brass bar. This has three holes drilled at 45 degrees at the bottom to take the pipes from the outside cylinders and the middle cylinder. The top portion was turned down and threaded 3/8"x32 to take a nut which will seal the pipe inside the smokebox using a shaped washer. It took two attempts to make this part as the first time I got the 45 degree holes too close together and it would not fit properly.

The final job was to silver solder all the bits together and I decided to do this in situ using the frames etc. as a jig. This was done using Easyflo with a small needle flame after fitting everything in place and protecting or removing any vulnerable bits. I wouldn't recommend doing it this way but it saves the hassle of making a seperate jig!

Exhaust manifold after silver soldering the bits together

The flange on the middle cylinder connector needs filing to shape still and I'm not sure I will be able to fit any bolts to hold it in place. I may just leave it as a push fit and seal it with a suitable gasket cement. It will be held in place by the exhaust assembly anyway and the pressure should be very low.

Whilst I was faffing around with the exhaust manifold I also conducted a bit of an experiment. There has been some discussion about needle roller bearings on the model engineering discussion board triggered by me saying that I was going to fit my Simplex with needle roller bearings and silver steel axles. Quite a few other people are either building Simplex or are about to start and one or two are interested in also fitting needle roller bearings. However, some have expressed doubt about the suitability of silver steel as an axle material as the bearing manufacturers recommend hardened steel axles. Now, quite a few modern writers in Model Engineer such as Neville Evans use needle roller bearings in the axleboxes of their locos and all use silver steel in it's natural state. Some have even used ordinary ground mild steel with good results. The point is that the loading on the bearings and axles when used in our models is considerably less than the loadings used in industry so the rate of wear is also much less. The general opinion is that the axles will outlive the loco!

As a matter of interest I decided to test a bearing/axle assembly as used in Helen. I fastened a length of 10mm silver steel to the end of the shaft of a small induction motor and fitted a 10mm needle roller bearing to it which had been pressed into a piece of bar to simulate an axlebox. On this 'axlebox' I hung a bag containing 16.5lbs of scrap metal to put a load on the bearing assembly. The motor was then left running over several days after lightly oiling the bearing.

Bearing test setup

The motor speed was 1440rpm and assuming wheels of 3.375" diameter this equates to a loco speed of 14.6mph. If this axle was fitted to an 0-6-0 the weight on the bearing would equal a total loco weight of 99lbs (6 x 16.5) which is obviously considerably higher than a 2.5" gauge loco in real life. Helen's axle loading will be less than half this. The motor was left running for 69 hours which means that our 'loco' would have travelled 1000 miles!

The axle was then examined and the only visible effect was that the bearing surface had become a dull grey colour. I measured the diameter but could not detect any wear with a micrometer or digital caliper so I'm quite happy that Helen's axles should be ok for several thousand miles. I doubt very much that she will ever reach anything like that!

Axle surface after 'travelling' 1000 miles


Although it's two weeks since the last update I have been working on Helen, mainly stripping and re-assembling.

Just a quick update on the needle roller bearing test - I left the test running for another 160 hours which means that the axle under test has covered the equivalent of 3,400 miles ! On inspection the bearing surface looked no different to the 1000 mile point and again I could not measure any wear. There didn't seem a lot of point in carrying on any further as I think the motor bearings will give out before the axle does !

Back to Helen and I've spent time assembling the cylinders with proper gaskets made from thick paper (printed out with CAD) and all the 'O' rings fitted. Before doing that though I finished the cylinder 'castings' off by cutting and fitting the little fill in pieces on the outside ends. They were cut from 1/8" brass and soft soldered in place using Comsol. (see photo of drop links)

The crossheads were temporarily fitted to the piston rods and drilled and reamed to take a single 3/32" taper pin for securing them to the piston rods. They were then removed again and the bearing surfaces case hardened using Kasenite compound to improve the wearing properties. They were then cleaned up again and fitted to the rods permanently with the taper pins. I've adjusted the slide bars with shims to give a slightly tight running fit as a good fit between the slidebars and the crossheads takes any sideways load off the piston rod.

I've also made a few small bits such as some proper nuts and fastenings for securing the connecting rods on the crankpins. The ones for the front crankpins are in the form of flat headed screws which screw into the end of the crankpins and have two holes in the head for tightening them up. The heads are quite thin as they have to clear the inside edge of the connecting rods. The 'nuts' for the two rear sets of crankpins are top hat shaped with two flats milled on the side to take a spanner.

The last items to be made before I fitted the cylinders to the frames were the two drop links fitted to the crossheads to drive the bottom of the combination levers. These were simple milling and filing jobs from 1/8" thick gauge plate. To ensure accurate positioning of the top and bottom holes they were drilled and reamed in the lathe by mounting them on the vertical slide and using the cross-slide micrometer dial to set the distance apart. I'll use this method for the holes in the other parts of the valve gear as it's a very accurate method of setting out.

The links are held to the crosshead by the nut which secures the small end pin and also a 10BA bolt to prevent the link moving.

Drop link fitted to the crosshead

The next operation was to fit the cylinders back into the frames. This was done one cylinder assembly at a time so that any tight spots associated with that cylinder could be dealt with before the next was fitted. It was at this point that I realised that a definite sequence of assembly had to be followed if problems getting at fixing bolts etc. were to be avoided. There's very little room between the frames to get fingers in! The first item to be fitted has to be the exhaust manifold as the screws securing it fit from the outside of the frames and are hidden by the outside cylinders. Next the outside cylinders can be fitted and then the middle cylinder. The steam manifold has to be fitted last otherwise you can't get to some of the outside cylinder fixing bolts! The only way to do this is to not fit the inlet stubs to the outside cylinders until the steam manifold is to be fitted. The stubs can then be pushed inside the manifold, the manifold put between the frames, and the stubs pulled out and screwed into the cylinders. It's a bit of a nightmare really but it's difficult to envisage these problems until you actually come to assemble the bits!

One or two problems were encountered with the outside cylinders. It was found that one connecting rod was slightly bent. This was soon sorted by careful bending in the oppposite direction to straighten it. Also one driving crankpin was found to be very slightly at an angle to the wheel. This was sorted by a few judicious taps with a hammer to straighten it up! I'd read about this method of truing up a crankpin before so it wasn't my idea! After all three cylinders were fitted the motion was quite stiff but with no real tight spots so I decided to run the chassis in a bit to loosen everything up. After a bit of thought on how to drive the wheels I hit on the idea of mounting a rubber roller (from an old printer) in the lathe and resting the loco wheels on this, the roller driving on the two front sets of wheels. I wasn't sure if there would be suficient friction between the roller and the edges of the wheel flanges but with a bit of help by turning the wheels by hand it worked quite well. I let it run for about an hour, stopping now and again to apply oil to the various bearings etc., and at the end of this the chassis was noticeably freer, although still a little stiff. I think I'll leave it for now and then repeat the operation when the valve gear is fitted.

Chassis driven by roller in the lathe

I took a short video of the chassis running with the digital camera although it's not brilliant quality. For those interested it can be downloaded here but it's quite a large file so could take some time to download if you do not have broadband!


The first items made for the valve gear were the combination levers. These are simple milling and filing jobs from 1/4" square gauge plate. Both levers were made from one piece of bar and were not seperated until the final shaping which made it easier to hold the blanks for milling etc. Once again the holes for the various pins were drilled and reamed in the vertical slide on the lathe cross-slide using the micrometer dials to ensure accurate positioning.

Drilling and reaming the holes in the combination lever in the lathe

The bottom half of the levers was reduced in thickness to 1/8" by milling and then the two levers separated. The top of the levers were then slotted with a slitting saw and the ends rounded off using filing buttons. The bottom half was finally reduced in width by milling to give a more authentic shape. I was in two minds whether to flute the front of the levers but thought it would be better to leave any 'fancy bits' until the valve gear was completed and found to be satisfactory. I did not want to spend ages making the bits look pretty and then find I had to remake them!

Next items were the anchor links and these were made from gauge plate again in similar fashion to the combination levers. I had originally intended these to be forked at both ends but found that the end that fastened to the droplink would have fouled the motion plate. Instead I changed the design so that the combination lever end was forked but the droplink end was solid and fitted on the outside of the droplink.

The valve crossheads are a bit complicated in that as well as having the combination levers pinned to them they have to drive the conjugated levers for the inside valve. This was achieved by having a forked lug on the inside of the crosshead which drives the conjugated levers via a simple flat link. The crossheads were machined from some 1/4" mild steel plate as I did not have any gauge plate big enough. The holes for the driving pins will be case hardened before final assembly of the valve gear.

Temporary assembly of the combination lever, anchor link and valve crosshead

Top view of the valve crosshead and drive link to the conjugated lever



The expansion links are basically the same as specified for the A1 (Flying Scotsman) and made from two pieces bolted together with spacers at top and bottom to leave a slot for the radius rod to pass between them. Each 'side' of the link has a curved slot on the inside to take the die block (two per link) and a pin silver soldered into the outside to act as the link pivot. This is perhaps slightly easier to make than the three piece LMS type which requires a slotted radius rod to pass either side of the single die block. However, care must be taken to ensure that the curved slots in either side are parallel to each other so that the two die blocks move up and down the slots exactly in line.

Cutting the curved slots in the links can be quite difficult. Some writers suggest that they can be marked out and drilled and filed to size using the die as a gauge but that method does not sound particularly accurate to me! In any case, that method cannot be used for these two part links as the slots do not go all the way through. That left milling them out with an endmill. One option would have been to mount the blanks on the rotary table in the Micro-mill and cut them that way but I did not think the mill was rigid enough to give a satisfactory result. The only way left was to mill them in the lathe using a suitable jig to cut the curve. Martin Evans described a suitable jig in several of his articles and I believe plans were available to construct it. Basically it is just a length of bar pivoted at one end and a screw at the other. Turning the screw moves the end of the bar up and down and causes the bar to rotate around the pivot. The blank for the link is mounted on the bar at a suitable distance from the pivot and the slot cut with an endmill held in the lathe chuck. I adapted this idea using a piece of 2"x1/4" brass bar and the two vertical slides mounted on the cross-slide. The rear vertical slide carried a fixed bolt to act as the pivot point and the front vertical slide carried a 'loose' bolt which raised and lowered the end of the bar as the slide was raised and lowered. The brass bar was drilled and reamed for two 'dowel' pins to accurately locate the blanks for the links. One of these dowels corresponded to the centre hole in the links for the pivot pins and the other to the pivot pin for the attachment of the eccentric rod. The latter dowel was removable so that it could be fitted in two different holes. This was so that two of the links could be machined opposite hand to the others. Some holes were then drilled and tapped to take 5BA bolts for holding the blanks in position.

The improvised jig for milling the curved slots in the expansion link blanks

The G clamp in the above picture is to hold some spacers which prevent the pivot bolt moving sideways in the T slot. The second T bolt is free to move in the T slot of the front vertical slide as the slide is moved. Obviously the pivot bolt could have been mounted on an angle bracket instead of the rear vertical slide but the slide made it a doddle to set the pivot bolt at lathe centre height.


Previous Page Next Page

Page 1 2 3 4 5 6 7 8 9 10 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26