Drawing 9 - Link Motion
1. Left-Hand Expansion Link Bracket  
The left-hand expansion link bracket, like it's counterpart on the other side, looks quite a complicated assembly at first glance. However, once you get into started, it's not so daunting. I am using the available gunmetal castings and they really are the pits with loads of material to come off in some places, not much in others and one part of the profile that can't be made as per drawing although it doesn't matter. They only faintly resemble the full-size article and there are fabrications available which are much more like the real thing. They also require a packing piece to be made that fits between the bracket and the frames because there is not enough material on the casting. It appears that the designer got his sums wrong and no-one has ever bothered to have the pattern modified. I shall be using offcuts of the frame material, 3mm mild steel, and need to account for this in the measurements.
Before cutting any metal, the first thing I did was to roughly measure from the as-cast back face to the centre of the reversing screw bosses to get some idea of how much needed to come off the back. I've set the height gauge for 2.065" and it looks like there's about twenty thou to come off the back. I then checked from the centre of the expansion link bearing to the top and bottom edges of the casting and it appears that the bottom edge needs just a clean-up (with the nib removed, of course) and can then become the starting datum.
Holding the casting for the first machining took a bit of thought but I found a way and then milled the back face flat, and also cleaned up the angled edge at the same time. I expected it to sing like a canary when milling the outer reaches but it was suprisingly quiet and no evident chatter. Now that I had a flat back face, the bottom edge was the next to be machined, followed by rotating round the fixture and milling the other two edges. The picture on the right shows how I quickly align the fixture using tee-slot packers and parallels when I don't need clocked-out accuracy. The casting was positioned using a scribing block to set the centres of the two reversing screw bosses at the same level for the first edge, and then a square used against this edge for the next one.
As each edge was machined, the flange thickness was also reduced to 0.200" at the same. This is not a dimensioned item on the drawing but they just about clean up at this size. It gives a flat seat for the lower mounting bolts and make it easier for clamping. Whilst machining the angled side, I also machined the area where the expansion link bearings go; one is cast in situ and the other gets bolted on later. Although there is a sacrificial piece at the outer edge, the pressure of the cut was bending the casting slightly so I propped it up with a wedge. The pieces at the bottom are a pair of wedges that I made as a ball-joint splitter about forty years ago, nice to see that they still come in useful. The clamps on the table are to stop the wedges drifting loose with the vibration of the cut.
Next up was the insides of the reversing shaft bosses where the left-hand side, or front of loco, gets machined away completely. These were machined to the finished size of 1.5/8" and the outside of the right-hand boss was also machined to 3/8". The top of the weighshaft panel was cleaned up at the same time. I decided to do the centre section where the weighshaft bush goes next and opted to do a full clean-up rather than the spot-face shown on the drawing. This provides me with a mounting face when I come to drill and tap the fixing bolt-holes on the back of the casting.
The rest of the faces were cleaned up in a similar fashion, just re-clamping in various modes to suit and I shall skip over these operations. These two pictures show the clamping arrangement for drilling the back. The drawing says "Mark off from frame" but I've calculated where to put the first hole - 5/16" up from the bottom and 1/4" in from the right-hand side - and used the frame drawing dimensions for spacing.
Once all the edges were cleaned up, I then clocked up the fixture and set the valves centreline to zero. From here I was able to drill the holes for the reversing shaft - a 3/16" pilot hole in this instance - and the slot in the front plate, doing the top half after the holes and then tilting the casting and doing the lower section. The corners of the slot will be filed square in the final dressing. The outside bearing carrier has to be machined at some point and was now separated from it's right-hand counterpart. They are a slightly different shape and won't fit properly if mixed up.
After machining all over to get flat surfaces the mounting bolt holes are drilled and tapped 6BA. The drawing doesn't give any dimensions and I have made mine at +0.475", 0, -0.475" from centre of casting, and marked the drawing accordingly - in a few years time, if I need to revisit this item, I won't remember the hole positions. Matching clearance holes were drilled in the expansion bracket and the picture shows the sacrificial waste to the side.
The two parts were then bolted together and the waste removed flush with the bearing carrier followed by the angled edge being milled to shape. The fixture was then set up one more time to drill the remaining holes, starting with the hole for the weighshaft bearing drilled and reamed 1/2" diameter followed by the expansion link bearing carriers.
It appears that the positional relationship of these two holes is important and this is where the DRO really pays off. Because of the flimsiness of these, packers and wedges were used to support them. After drilling and reaming the main hole, the four flange holes were put in the top carrier, then the carrier removed and the lower one drilled. I will now spend some time prettying these up with the Dremel and then make the various bushes and bearings. Also, the front reversing shaft hole was reamed to 9/32" diameter and the rear one tapped 5/16"" x 40 ME
I have made the expansion link bushes from phosphor bronze because they are going to get a fair bit of wear and the four bolt holes were drilled on the mill using co-ordinate drilling. There are two for each side and they are the same both sides. I've also made the reversing shaft bushes and the weighshaft bushes but these have been made from brass as they won't get anything like as much wear. The backing plate has also been made from some 10swg mild steel and all the parts assembled together although I still have to make the reversing screw which is a left-hand square thread and will need some thought. The assembly has been temporarily bolted to the mainframes and work will now start on the right-hand bracket.
2. Right-Hand Expansion Link Bracket  
The right-hand expansion link bracket is very similar to the left-hand bracket, although a mirror image, so there is no point covering the same machining operations all over again. This is the starting point and, like before, the castings are very rough and also require the packing piece behind them when assembled. The one thing that is a little different is that the amount of material available between the rear mounting face and the inside of the expansion link bearing face is a bit mean. I am going for a full clean-up but may need to increase the packing piece thickness to compensate to get the link and valve centre-lines the same.
This bracket also carries the restoring spring mechanism and this has been fabricated out of some brass sheet, a brass bung and a length of 15mm copper tube. The bung and tube have chamfers to create a fillet and were a tight fit together, with the three parts then being soldered together.
3. Expansion Links    
The starting point was to cut a pair of blanks from some black flat bar. On these drawings, all the expansion link parts are drawn at twice full scale so I can't trace the outline and tranfer to the material. These have been machined all over to give a flat plate of the final thickness (1/4") and, using the centre of the trunnion pin as the x0,y0 datum, a locating point drilled and reamed 3/16". This was followed by working out the absolute positions of the three 6BA clearance holes for bolting on the trunnion brackets using the pitch circle co-ordinates in the Zeus book and, finally, the hole for the eccentric rod. Although this should be 1/4" diameter reamed, I've left it at 3/16" dia for the moment because I have a filing button of this size for the outer form.
I had a couple of pieces of 2" x 1/2" flat bar going spare and used these to create the machining fixture for the next few operations. At the lefthand end I have made a 5/16" diameter pivot point which gets bolted to the table and the DRO centred on the pivot. At the right-hand end another piece acts as a pressure pad with a fence spaced a few thou over the half-inch bolted on top which holds the working plate captive but free to slide back and forth. The working plate is a ten inch length of the same material with two holes drilled and reamed 5/16" at the left-hand end. The first is for the outside radius of the link and the second is for the radius of the slot in the link. At the other end are three holes drilled and tapped 6BA and a hole drilled and reamed 3/16" to take the locating pin. I have also drilled and tapped a couple of M6 holes to take a pair of clamps. Finally, I have made up a crude leadscrew arrangement from some M8 studding and a couple of bits of brass to allow controlled movement of the working plate.
The leadscrew works because the two brass bits are free to pivot in the respective plates and there is a small bit of plastic under the grubscrew in the leadscrew nut that reduces the backlash of the thread to a minimum. It all looks a bit Heath-Robinson but it does work and the next photo shows completion the first operation. I roughed out using plunge cuts and finished by taking two radial (side-cutting) passes at ten thou and a final one at five thou to finish at the correct dimension. The fixture was then adjusted and set up using the second pivot point and everything re-centred.
I then marked out the top and bottom extremities of the link by clamping the workplate and moving the table to plus and minus 1.218" on the DRO and scribing a very short line by dragging the chuck with a centrepoint loaded in followed by resetting the table to zero and pivoting the workplate till the mark matched the centrepoint at the edge. The workplate was then clamped, a 1/4" dia cutter loaded, the table move out and the angle formed with multiple passes until I reached 0.125" on the DRO (half the cutter diameter). A similar procedure was used to mark out the extremities of the dieblock guide and four 1/16" diameter relief holes drilled before milling the slot. Using the pivot block and adjusting to centre each time ensures that all the holes and slots are in the right place but trying to mark this shape out from dimensions only would be a nightmare.
The final operation with this setup was to mill out the slot and for this I first punched the locating pin in the workplate down below the surface and then roughed out with a 3/16" dia slot drill using plunge cutting, removing the leadscrew arrangement and just moving and clamping the workplate by hand. Then I changed back to a 1/4" dia cutter and repeated the plunge cuts. Finally, I dropped the leadscrew back in and ran the cutter back and forth in the slot to clean the sides.
Now that the slot was finished, it was just a case of getting the exterior shape reasonably correct. The working plate still came in useful as for milling the lower angle because it just needed to be set to the thirty degree angle and the lower section milled away. The rest of the outside shape was marked out using one of the unfinished trunnions to scribe the rear circular shape and digital vernier to mark out any other dimensions. They don't show up particularly well but this photo gives the general idea. Look closely and you will see some lines.
From this point on, it was mainly hand-work sawing off the excess material before filing, grinding and linishing to get the final shape. And here are the two links: they just need the eccentric rod bearing hole opened out and a PB bush installed plus a little more polishing. I'm quite pleased with how these have turned out, and it just goes to show that complex shapes can be made with the simplest of equipment if one just takes the time to work out how. I will probably avoid laser-cutting for a while longer, it's far more satisfying to make it ones-self.
4. Expansion Link Trunnions    
The expansion link trunnions have been made from 1.1/4" dia mild steel as complete articles although the drawing seems to indicate differently. All four billets were turned to 1.3/16" diameter and faced off (two of each size) and then held in soft jaws for the second and subsequent operations. This was just straightforward turning and doesn't require description. Then they were drilled on the mill using the same technique as for the expansion links and then the backs milled to create the clearance for the radius rod.
The inner pair, with the short axle, were drilled 6BA clear with deep countersinks and the outer pair were drilled and tapped 6BA. Because of an issue with sloppiness, these were later remachined to cater for 1/8" BSW. To the right is a composite photo showing three views of the nearly-completed link assembly. The hole for the eccentric rod has been opened up and the phoshor bronze bush has now been inserted in the link.
The various small jobs on the expansion links have now been done, the expansion link brackets have been line-reamed and the expansion link assembly put up between centres and the journal turned to exactly 0.250". The brackets were disassembled and then reassembled with the expansion link assembly in place. They rock too and fro but are a little stiff at the moment and will ease when in use. The right-hand bracket is shown here.
5. Lifting Link Bellcranks    
The drawing shows the lifting links being fabricated from some pieces of 14 swg mild steel mounted on a central boss and "riveted over" but I can't see how this can be very accurate, or rigid for that matter. I decided to try soldering the parts together and started by cutting out the arms from some offcuts of 2mm M.S. sheet as there is only a couple of thou difference in size. The longer arms are the same size on both sides of the loco so the four pieces were machined as a single item, first milling the sides to clean up at 11/16" followed by drilling and reaming the 7/16" hole at one end and the 3/16" hole at the other. Because of how I held them in the vice, I added a top clamp (moved aside for the photo) to ensure that they remained held together but drilling this way gives more support to the point of the drill and ensures a reasonably round hole.
On the lathe, I made a pair of filing buttons and the two central bosses, noting that they are different sizes. The arms were then marked out and the basic shapes formed using files and sanding drums. When it came to soldering time, I played with various ways on holding the assemblies together whilst soldering and this gave me the most trouble. The arms to the reversing screw are at 87 degrees to the lifting arms and the arm to the return spring on the other side is at 83 degrees. I made various spacers to fit between the arms and hold them square, set them up on the mill and fixed the angle using clocks but one problem kept getting in the way.
To keep position, the assemblies had to be kept tight and to get solder in anywhere they had to be loose. After many frustrating hours trying to make this work, including scrapping one boss and remaking, I decided to abandon the idea of soldering and use pins instead. To get this under way, I started by making a simple fixture on the mill to load the centre bosses to and drilled two 1/16" dia holes through. The pin was then sleeved up to 7/16" to suit the long arms and a further locating pin added to the fixture. The arms were then drilled with a pair of 1/16" holes in the boss end at right-angles to the arms.
The fixture was then modified again to accomodate the other three arms. To set these at their respective angles, I worked out the hole positions for the locating pins using sine tables and set the position on the DRO. The holes were then drilled at the same position as all the others. Finally, the holes in the outer arms were countersunk and the various parts were riveted together with some home-made rivets cut from 1/16" dia brass bar. These will now be put aside until assembly time when a cross-hole will be drilled at forty five degrees through each of the bosses and the weighshaft, and pinned with taper pins to keep them in position.
6. Radius Rods    
The radius rods, or expansion rods as they are called on my drawing, were made from some 10mm square black bar which was cleaned up all over to finish at 3/8" square. I then mounted them to my pair of angle plates and drilled and reamed the three 3/16" holes using the DRO for precise positioning. Using the same style of setup, I then necked the rods using a 12mm end mill. I prefer to rough out using a plunge-cut, which can be seen from the finish at this point, followed by a sliding pass to clear the high spots and a second pass at finished depth, about an extra six thou in my case. There is less stress on the machine and, as a bonus, I have loads of 12mm endmills and use them where I can.
My next job was to make a fixture to bolt the rods to and this was made from some 1" thick-wall box section, milling square then drilling and tapping the fixing holes. I used 2BA as it's a good fit in the 3/16" reamed holes and the DRO ensured that the holes were in exactly the right place. The end sections between the two holes were done first because this part is parallel to the centreline of the rod and the fixture was then repositioned to machine the tapered sides of the rod. I worked out that I needed to offset one end by about forty thou to achieve this, and feelers were used to act as packing between the fixture and the table slot wedges. After machining the front, the feelers were moved to the opposite end so that I could mill the other side.
The flycutting bit that I had made for the connecting rods was too large for these radius rods and I made a new tool bit from an old centre-drill. I had planned to turn the fixture ninety degrees and then mill out the flutes but I forgot to allow for the bolt holding the tool bit in place so plan B was adopted which entailed lifting the fixture up somewhat. The angle plates were brought back into use and the feelers placed under one end to set the correct angle for the first pass, then moved to the other end for the second pass. I described the basic setup back at the con rods page and won't repeat that here. While I had the angle plates set, I also took the opportunity to rough out the slots and set the finished depth of the slot. The long end,in particular, gets a bit flimsy further down the job.
To complete the opening out of the slots to the finish size, I held them in the vice and surrounded the sides with a pair of fences to give extra support and then gently milled them out with a 3-flute slot drill, working outwards from the cross-holes all the time. Because the side walls are so thin, coming in from the front end courts disaster as the endmill will probably pick up the side and bend it inwards. Not worth risking after this amount of work. And then, to finish, a bit of time spent deburring and polishing up. Getting into the flutes is not easy because of their small size but I was able to clean them up a bit more with the file shown in the last photo. The tool I used to create the flutes is shown also.
7. Union Link    
When I made the radius arm machining fixture, I drilled and tapped the two 2BA holes for the combination link at the same time. The pair of holes in the combination link were drilled and reamed 3/16" in a vice and then bolted to the fixture so that the rods could be necked, as before. A pair of 3/8" dia washers under the clamp-down screw heads acted as visual limit markers to ensure that I didn't get too close to the ends with the endmill. This time there is no fluting of the rods, just a central relief with angled edges and this was done with an endmill that I had in the "specials" drawer, although it could just as easily been filed to shape. They were then loaded back into the vice and the slots milled with a 3/16" dia cutter, no special support being required this time as they were much more rigid. A quick clean-up all round and these now get stored until assembly time.
8. Combination Lever    
The combination levers were once again made from black bar milled all round to clean up. There is no fluting on the items but the centre section is thinned from 3/16" to 1/8" as well as being tapered from 1/4" down to 3/16". The drilled and reamed holes on these are precisely dimensioned on the drawing and it was obviously important that they were accurately positioned, otherwise they would have been dimensioned in fractions. After drilling and reaming the holes in the milled blanks, the drill jig had three extra holes put in it at the same positions and clamped square to the table. The hole for the radius arm pin was set as the zero datum and a variety of roughing out operations were then undertaken by calculating the various finishing positions for the cutter. Once again, the washers acted as visual limit markers in the uncalculated axis and the next picture shows one of the combination levers after the first round of cuts.
The fixture was then moved over at one end and the taper formed on the central section. Flipping the workpiece over allowed for maching the opposite side without having to adjust the fixture a second time. It can be seen that the basic shape is starting to develop here in the picture on the right.
The waist of the combination levers were thinned by 1/32" each side next and then the fixture was turned ninety degrees and a variety of other operations performed by mounting the levers on the side of the fixture.
One of the other general-purpose fixtures that I made a while back came in useful because of it's height, a matter of moments to drill and tap a 2BA hole high up allowing me to drill the oiling holes in the end of the levers. As always, the final work was to clean things up with files and rotary sanding drums and the final picture shows them nearly finished, along with a bunch of pivot pins that I made on the lathe whenever the mill was able to run using the power feed. Since taking that photo, I have also drilled and tapped the 8BA holes shown on the drawing. I was debating leaving them out until I realised that the die block on the valve guide uses headless pins that need to be held captive and this addresses that issue.
9. LiftingLink    
The lifting link connects the lifting link bellcrank to the radius rod and is very similar to the combination lever but with one less hole and were made in the same manner. I'm only including these with a couple of pictures for completeness. The only variation to my usual way of making these was to form the raised pivot bosses by bolting the workpiece to a mandrel in the lathe. Based on the drawings, this is the only one of these parts that can be turned this way.
10. Next item...