Drawing 21 - Tender Frames, Axles
1. Frames    
The tender frames have been made from guillotined pieces of 3mm mild steel plate and were marked out by hand and all the cutaways made with hacksaw and files. Because I didn't have my larger mill at the time, all the original holes have also been marked out by hand or spotted through the various components. The half-ovals were chain-drilled first, followed by removing with a cold chisel and filing the shape to finish, and the large hole at the rear of the frames was made using a hole saw. Because most things are dimensioned to fit inside the frames, I kept these dimensions and compensated for the thinner frame material (two lots of seven thou) by making the horns a little wider.
Looking at photos of the prototype, there is a row of rivets that appears to run the full length of each frame just below the sole plate. By counting rivets between the spring hangers, and using the drawing dimensions between hangers, I was able to calculate the spacing. The frames were set up on the new mill to add this detail and a series of 1/16" holes drilled between the water filter and the rear steps. The water feed pipe hides any rivets that may be towards the front and have not been drilled. The picture to the left has been modified form one of the Nigel Fraser Ker photos.
A further detail I have chosen to incorporate are the rivets for the rearmost top frame stretcher which is a series of twelve holes, also in 1/16". The front stretcher is behind the filter box so wouldn't be seen and not worth copying. The final detail I have added is the triangular-shaped bracket directly above the centre spring and is cosmetic only. After removing from the mill, all the holes were countersunk on the rear. The dummy rivets were then fitted and hammered flat at the back, followed by dressing back to flush, particularly in the area of the top stretchers. The triangular brackets were also fixed and the frames are now ready for priming.
2. Front Beam    
The drawing shows the leading beam as a two-part assembly with 1" x 1/2" x 1/16" channel behind the buffer plate, riveted together. I had a short length of 1" x 10swg box section available and made my own channel from this, whilst the buffer plate was cut from 3mm mild steel sheet. The drawing also calls for the frame-mounting plates to be silver soldered to the ends of the channel but I felt that some form of mechanical fixing was also required and made a pair of angles to fit in the ends of the channel and riveted to the buffer plate. I started by sawing the box section longways to produce the channel, dressing the sawn edges and milling the ends square and to finish length. The two angle pieces for the ends of the channel were cut from the waste and the holes for the rivets marked out and drilled. Next, I marked out and drilled the twelve rivet holes in the front plate, countersinking quite deep, and all the bolt holes for the bufferstocks and the pivot block. On the channel, I marked out one rivet hole at each end and then riveted all the parts together, remembering to include the angles. Now, it was easy to hold the channel in a machine vice and spot through all the other rivet holes, finishing off by setting all the rivets and filing the front face flush. They can just be seen in the photo although this was taken later because I hadn't put the larger holes in at this stage. All the other small holes were drilled through at this stage also.
The bufferstock location holes were made next, followed by the drawbar slot which I stitch-drilled and chopped out, then filed to size. I didn't have my larger mill then or I would have cut the hole with a small slot drill. All the holes that needed threads were done at this stage and care was taken with the pivot block holes as four of them foul the channel behind. The reason for this is because the pivot block on the loco has the mounting holes in a different position to the tender but I had made mine both the same and had to modify the tender dimensions to suit. The end plates were made from offcuts of the 3mm mild steel sheet. The two forward holes were drilled 6BA clear and the four rearward holes drilled and tapped 6BA. After drilling and tapping the angled ends of the channel, I was able to bolt the end plates on prior to soldering. The two front screws go through the frames and the end plates, screwing to the riveted angle pieces, thus providing my mechanical fixing.
3. Forward Buffers    
The buffer housings on the leading beam of the tender are drawn as a fabrication silver-soldered together but I found it easier to make mine from solid. There is little need for accuracy required here other than a nice sliding fit of the buffer in the tube. Because my 3-jaw chuck accepts 1.3/16" dia material in the bore I cut a 4.3/8" length of EN1A, enough for two housing plus parting and facing. With half of the material hanging out of the chuck, I faced off the bar and then turned the O/D down to 0.650" for the 2" length shown. The bar was then reversed in the chuck, shoulder up against the jaw face and the operations repeated, followed by parting into two components. Leaving the first one in the chuck, I drilled a 31/64" diameter hole to within 1/8" of the bottom, taking advantage of the flange to prevent the material moving. The hole was then reamed 1/2" diameter as deep as possible and the front faced to leave a flange thickness of 1/16". This was repeated for the second one.
A small chuck was set up on the table of the Cowells to enable machining of the square flanges, setting the centre as the zero datum and milling an equal distance out from centre for each side. The four mounting holes were centre-drilled at the same time but drilling was done free-hand on the pillar drill as the Cowells doesn't have the headroom for drilling. As you can see, it's a bit cramped on the mill table. The corner radii were done freehand on the linisher and with files. The retaining slot was made by holding the workpiece in a small machine vice and milling with a 1/8" diameter slot drill. The retaining pin is yet to be made. The plungers, or buffers, were made from 1/2" diameter EN1A. The hole for the spring was drilled first and the O/D polished with emery to give a nice sliding fit in the buffer housing. They were than reversed in the chuck and the front form created by turning the shoulder and using a form tool to create the radius.
4. Pivot Block    
The pivot block was covered on Drawing 2 as part of the loco rear beam but is included here for completeness. For anyone making Britannia to this set of plans, the mounting holes are different on the loco compared to the tender and causes a problem if made to the loco dimensions. The central stand-off bracket, or pivot block, is a fairly rough gunmetal casting and a little thought went into the machining sequence. Because the casting sprues were so prominent, I decided to mount the pivot block on a small angle plate, setting the side square to the table, and milled the top edge with my small flycutter until cleaned up. Now that I had one clean face to work from, all further work was done in the milling vice, starting with the bottom face and then some more off the top to get the webs and drawbar slot balanced.
The next job was to face the back to the nominal size. To get it level, I set four screws into the tender front beam (rear beam of the loco in the picture) and adjusted them to the same height. This then went into the bottom of the milling vice and the casting settled on top.
The back was then machined with a flycutter. There is no size given on the drawing but scaling from the drawing, which is normally inadvisable, worked out at about 1/8". It wont matter as long as I remember to make the drawbar to suit, not just to drawing. The sides and the large radius were dressed clean on the linisher - no particular size, just to please the eye - and the mounting holes were marked out and drilled 6BA clearance. To finish off, it was given a quick session in the shot blast cabinet to make a good surface for painting.
5. Rear Beam    
The rear beam has been made in a similar fashion to the front beam, again using a section of 1" x 1/8" box section to create the channel as well as my belt-and-braces angles at the ends of the channel. This time, however, the channel is set level with the top of the buffer plate and the plate is screwed on after other parts have been fixed. Because of this, the angles were fixed in place first, and a very light skim given to the ends to dress them square. As before, the side plates were made from offcuts of 3mm MS sheet and bolted on before soldering together. All the holes in the plate were marked out, centre-popped and drilled freehand. The brass angle pieces were cut from stock material and drilled also. I fixed all the dummy rivets first, followed by the angles and, finally, the drawhook stregthening plate.
The drawing calls for the buffer plate to be riveted to the channel but I have chosen to use 6BA countersunk screws, hence the deep countersinks. The screws will be filled prior to painting. On the back of the buffer plate, all the rivet holes were countersunk and the rivets have been hammered in and filed flat to enable fixing to the channel. I managed to make a bit of a pig's ear of the buffer stock alignment holes, as can be seen in the picture, having to file them over to get the stocks to bolt on. It's not important, though, just annoying. The screw holes are in the correct position. I am not going to cover the buffers and buffer stocks because I made these on a Colchester Tornado CNC lathe eighteen years ago and there is not much point in describing the process on here even if I could remember it.

I do recall that I made the stocks in a single operation from round bar, leaving only the mounting holes and square shape to finish by hand. The buffers were turned and parted off in a single operation, the parting tool being used to get most of the way though, retracted and then following the circular path needed to form the shape, just leaving a small pip to linish off. Also in the picture is the drawhook which was made completely freehand with a few drilled holes, a hacksaw and files. And then I remembered that the tender buffers cannot be made to work in the same fashion as the loco buffers because the frames clash perfectly with the centreline of the buffers and a different means of retention was required! Therefore, a 1/8" slot was milled in the underside of each stock for a distance of 3/8" and a 3mm threaded peg made that screwed into the buffer after assembly, in the same manner as the tender-to-loco buffers.

6. Bottom Stretchers    
The drawing shows the two bottom stretchers as a fabrication, made from 18 swg mild steel, but I have chosen to make mine from solid. The drawing also shows them riveted into position but I am going to fit mine with 8BA screws instead. To be any use, these need the length to be reasonably accurate and I don't feel I could bend sheet steel to this level of precision. Milling them to length is the obvious answer. Two pieces were first sawn from 40mm x 10mm black steel and these were then cleaned up all round on the mill, finishing at the exterior dimensions as per drawing. Following this, the two circular and one rectangular relief sections were milled away to leave a solid version of the stretchers.
Then it was time to pocket mill the interior sections and this was done with a 10mm carbide end mill, twenty thou depth per pass and full-width cutting around the periphery before clearing the centre. All previous work had been done on the little Cowells but I had my larger mill when it came to do this operation. As can be seen, I have left the side wall thickness nearer to 1/16" rather than the specified 0.048" but the ends are about 1/4" thick to allow for making the fixing holes. When the milling was complete, the stretchers were clamped to an angle plate and the holes drilled and tapped 8BA. Here are the two finished articles ready for painting and assembly.
7. Top Stretchers    
The top stretchers have been made in a similar fashion to the leading and rear beams by using 1" square box section for the cross-member. However, I made these as a fabricated item requiring no soldering because they are just a spacer. To make the end pieces, I faced a pair of 1.1/2" dia mild steel billets on the lathe, then milled them to the shape shown. The holes in the ends of the channel were drilled 6BA clear and matching tapped holes put into the blocks, which are a snug fit into the channel. I also drilled the tank mounting bolt holes in the top of the stretchers but had to redo them when I decided to change the position of the tank bolts for ease of manufacture.
8. Horn Cheeks    
The horn cheek castings come as a pair joined together but I chose to separate mine and machine them separately. There was nothing out of the ordinary in machining these, being held in a machine vice for all operations. The only tip I can offer regarding these is the little clamping piece I made up to help hold the work. With twelve cheeks to do, this came in very useful and easy to use - beats juggling with other forms of blocks or pads. When I mount the vice on the small Cowells mill table, the fixed jaw is square to the table within a couple of tenths of a thou so I put a double thickness of baking foil under the fixed-jaw end to create a tiny deviation from the vertical for the fixed jaw, such that the finished components were "toe-in" at the outer ends when mounted on the frames.
Once all the milling was complete, the holes were drilled and countersunk, and each cheek marked with it's location for strip-down and re-assembly. It is usual to use rivets to fix the horn cheeks in place and then final machine but I have fixed mine with cheats rivets - 8BA dome-head screws without slots. This gave me the opportunity to adjust the cheeks using a 1.1/4" square bar and feeler gauges, without the need to machine them in situ. Once they were screwed firmly into place, two holes were drilled through the frames and into the webs of each cheek and a pair of silver steel dowel pins inserted. This should stop them moving in service since, unlike the main locomotive horns, they are not subjected to a reciprocating action.
9. Horn Stays    
The horn stays were made from some 1/4" square mild steel, cut to length and the rear milled away to create the required shape. Although not shown on the drawing, I have also milled the fluting that is evident on photos of the prototype using a small Woodruff cutter. A pair of tapped holes were put into an offcut of mild steel sheet to allow bolting of the components prior to machining.
10. Top Plate    
The area behind the leading beam is taken up by a strange ensemble of plates and spacers that provide support for the drawbar pin supports, the brake and scoop bellcrank assemblies and also the upper platform and deflection plate. The top plate, made from 16 swg brass sheet, sits between the frames and is flush with the top. Soldered to the underside of this is the upper drawbar pin support, made from 1/2" dia brass bar. The hardest part of this component is deciphering which of the mish-mash of dimensions on the drawing relate to this and which to the lower plates. After making each of the items as per drawing I found it very awkward to assemble the parts, wishing for a second pair of hands. Subsequently, I made a modifications to make life easier which was to discard the lower front spacer in favour of three brass bushes soldered to the underside of the top plate since these cannot be seen once assembled.
11. Bottom Plates    
There are two bottom plates joined by a length of 7/32" flat bar. The front one of these carries the lower drawbar pin support which, like it's upper counterpart, is soldered to the plate. The lower one carries the scoop and brake bellcrank brackets below, and a pair of spacers above for connecting to the top plate. The drawing suggests riveting the two plates together through the flat spacer bar but I have fixed mine with 8BA bolts and nuts. The brake brackets are made from 3mm mild steel plate and are bolted to the plate with the centre screws going into the 1/2" square spacers.
I'm not modelling the scoop and assocciated gear but it can be added later if I change my mind. These two views are the lower plates primed and in position.
12. Platform    

The platform is made from 20 swg brass sheet. It was cut out with a hacksaw and then filed to size prior to marking out the bending position. The material was left a little long to assist with bending and the two notches at the front were cut away. Bending was performed the old-fashioned way of supporting between two hefty lumps of steel and gently peening over with a mallet. The platform was then set up against angle plates on the mill and the height of the platform lip machined to size.

I tried it laid flat and using the end of a milling cutter but it was difficult to support the edge and stop it ringing like a bell. Using the side of an endmill moves the cutting forces down into the material, rather than at ninety degrees, and the lip becomes self-supporting. The next operaion was drilling all the holes and the plate was clamped on a piece of MDF, trued up and the holes made by using co-ordinate drilling.
13. Deflection Plate    
The deflection plate is also made from 20 swg brass and was prepared in much the same way as the platform. The holes at the base of the deflection plate were drilled in a similar manner, the tee slots on the table being deep enough to the allow the material to sit nicely on the 12mm MDF and squared up against the edge of the tee slot.
14. Wheels    
After a small amount of fettling to clean up the spokes a little, each wheel was mounted in the independant four-jaw chuck, front face inwards and holding on the tread, and adjusted until the boss ran reasonably concentric. The back was then faced off till clear of witnesses and a further ten thou taken. The bore was drilled 1/8" undersize, followed by boring with a small boring bar to 1/64" undersize and finally reamed to the correct size. The O/D of the wheel flange was also skimmed to clean up but not to size. The wheels were then mounted in soft jaws on the three-jaw chuck and the bosses faced to clean up followed by machining the face of the wheels. The object here was to try ang get the correct look of the wheel-rim against the spokes without removing too much metal and leaving the wheel thickness undersize. Once I was happy with this, I then continued facing off the boss to the correct distance from the rim. The tyre was then turned to about sixty thou above finished size to get it concentric for holding on the next operation. I'm missing a couple of pictures here, and I've checked all my backups, but a wheel in a chuck doesn't look like much else so no great loss. The wheels were then reversed in the chuck and the backs machined down to finished thickness. The O/D of the wheel-rim was also skimmed to finished size and a radus filed onto the rear edge.
For the tender wheels only, I made a fixture which allowed me to mount each wheel on a mandrel correctly sized to the axle holes, held with a nut and washer, and with a driving pin to provide positive drive through the spokes. A piece of tool steel was ground to produce the root radius of the tyre and flange and the compound slide set over to two degrees. Using this tool, itself set over to the correct five degree angle, the flange was faced off to the correct thickness and the saddle was locked at this position. The compound slide was then retracted and the tyre machined to the correct angle using the compound slide instead of the carriage. Care was taken near the end of each cut to get a nice blend with the root of the flange. Once finished, I filed the radius to the edge of the flange and the wheels were complete except for further fettling of the spokes.
15. Next Item...