Friday, 27 November 2020

The Chain Tensioners

As with the AJS K7 restoration and the AJcette project a Weller chain tensioner is required. The only difference with the V-Twin engine is that 2 are needed but, otherwise, the principles are identical to the earlier single cylinder K7 engine.

The first task was to generate and fix 2 strips of ertalon 4.6. This comes in round bars and must be machined into flat strips before it can be used.

If this were a piece of steel it would be possible to clamp it at both ends and machine one side in a single operation. Unfortunately the flexibility of the material does not allow that and it requires clamping in the middle and machining the flat in 2 operations on each side. After one side has been finished it can be turned over and the other side machined.

And, eventually, there are 2 flat strips of ertalon 4.6 ready to be fixed to the timing case.

On previous engines, these were fixed with copper rivets. This method of fixing could not be used here because of the difficult access to the lower fixing holes on the front cylinder that prevents riveting, namely that the rear cylinder part of the case gets in the way. In fact it was not even possible to drill holes easily due to the difficult access and these lower holes were spark eroded.

Each hole was tapped carefully turning the tap with a spanner from the inside of the casing, the strips were drilled and countersunk and attached to each surface with brass screws and a specialist adhesive.

As with the K7 and AJcette, the springs and blades were made by Alberta Springs and, with blade pivots and the spring tensioner retaining guides, aka the reaction dampers

The complete Weller chain tensioner could be installed

At this stage, there was still a dummy shaft and sprockets for the drive to the camshafts and the next task was to make the  bearing housings and camshaft to fit inside the Velocette camboxes.

Sunday, 22 November 2020

Timing Case Machining – Part 1

A couple of weeks after delivering the patterns to the foundry they called me to let me know that the castings had been made and heat treated. As with the crankcases, the material is LM25 Aluminium. They also mentioned that they had made a “spare” of each. I had already asked for 2 castings of each, just in case something went wrong with the machining, but I now had 3 sets!


All the machining went well, so I still have 2 spare sets ….if anyone wants to make a V-Twin like this one that please contact me for a good price!

I was very pleased with the castings and, after boring the 2 holes for the camshaft drives and the magneto in the inner timing case, I clamped the case to the crankcase with the engine in the frame  to check that everything was the right size and in approximately the right place.

It was.

The first machining operation would be to face the inner surface and to then drill and tap the timing case and crankcase for their securing screws. The inner surface was flattened by rotating on the plywood emery board and then the entire engine was built up on the milling machine table for drilling and tapping the holes for the securing screws.

After machining the first 4 holes, the timing case was put on the emery board for a few “laps” to clean up and start flattening the mating surface to the outer.

The effort in making the mating surfaces on the 3  "legs" of the pattern coplanar had paid off – the large “V” mating surface was, as cast, already nearly completely flat and just required a bit of cleaning up.

After machining the last 2 holes, the slot for the chain tensioner retaining guide (this is referred to as the “reaction damper” in the AJS workshop manual) and preparing the surface for the tension-side low-friction strip, the inner timing case could now be firmly fixed to the crankcase. The front cylinder "leg" of the timing case would not fit within the throat of my milling machine but luckily one of my buddies has a much larger Bridgeport and I used his mill to machine that part.

And with the cylinders, cylinder heads, camboxes, dummy camshafts and chains the whole assembly could be placed in the frame and again checked that everyone was in the right place.


So far, so good.

Wednesday, 18 November 2020

The Timing Case Patterns

Having previously made the patterns for the single cylinder AJcette, detailed here and here and which worked out pretty well, I planned to use the same approach for the V-Twin.

The first step was to make a number of resin copies from the K7 timing case moulds and to then cut-and-shut these to assemble new inner and outer V-Twin timing case structures.

The picture below shows the first 2 pieces that were amputated from resins and stuck together with JB Weld at an angle of 500 ….you have to start somewhere.


The plan was to make most of the inner timing case and to then match the outer timing case to the profile of the inner. It is a slow and laborious process making the patterns using bits of resin to get detail close to the original plus pieces of wood and metal, held together with JB Weld and Araldite and finished with P38 body filler.

The inner timing case was made by first constructing the lower section that could be clamped to the timing-side crankcase and then adding the upper portions according to the chain run. The whole assembly was then positioned approximately in the correct location in the frame to determine the length of the magneto chain section of the timing case. 

This process takes time – it took me about 6 weeks to make both patterns, because it is only possible to do a bit at a time before waiting until one or other of the adhesives has hardened sufficiently to handle the pattern and  attach the next bits.

On the outer timing case on the right of the above picture, I have machined the main flat part of the pattern from a piece of aluminium plate to give it structural strength. The circular section in the centre is formed from a pair of stainless steel Korean rice bowls bought on ebay for very few ££ and spaced apart for thickness of the casting. The main oil pump is screwed onto this face and a reasonably thick section is therefore required for the threads.

I found it useful to use tailored strips of wood for the sides and these can be clearly seen on the inner case pattern on the left. The wooden strips were made to order by AA Woodcrafts for a reasonable price and delivered very quickly.  I also found a local company, Finecut, that would make more sophisticated shapes in wood, in this case semi-circular sections. These, together with machined aluminium sections, were used at the extremities of the chaincases.

I wanted the castings to be as flat as possible at their mating surfaces as I do not have and do not have access to any machine tools on which these surfaces could be machined in one operation. The plan for ensuring  the patterns were as flat as possible was to first eliminate any bowing by weighing down the patterns in the appropriate places, as shown below, and to eliminate high spots by carefully rotating the patterns on a flat sheet of plywood to which multiple sheets of coarse emery cloth had been stuck. The flatness was checked using a thick sheet of glass.

The picture above is taken prior to attaching the Korean rice bows. Additional thickness to surfaces that will eventually be machined has been added in the form of thick cork gaskets cut to shape; these can be seen at the holes at the end of each chaincase section.

Eventually the patterns were finished and ready for the foundry.


As the patterns are fairly fragile – they would definitely be damaged if dropped, a bespoke box was made to store and transport them. The surfaces to be machined have been painted red to assist the foundry in positioning the sprues and risers in the mould.

….and there were plenty of offcuts of resin timing cases left over for scrap.



Saturday, 14 November 2020

The Camshaft Drive

The chain drive to the overhead camshaft follows the same principle as that for the single cylinder K7 and AJcette engines and was covered extensively in previous blogs. The main difference is that the V-Twin requires an additional sprocket to drive the camshaft for the 2nd cylinder, resulting in a shaft with one 40 tooth gear driven by the crankshaft and 3 sprockets, one to drive the magneto and 2 for the camshafts,

Blanks were made for the 20 tooth crankshaft pinion and the 40 tooth gear as before and the teeth cut on the milling machine

and then heat treated and, together with 3 sprockets, were keyed onto the shaft.

These are a tight interference fit and although it is highly unlikely that they would ever shake lose, the gear and each of the sprockets were laser welded to their neighbours. The extremely fine fillet welds can be seen in the picture below.

To minimise the width of the assembly the location of the teeth on the middle sprocket is staggered with respect to the sprockets either side to avoid the rivets on the chains interfering.

The outrigger bearing support was made in a similar way to the previous K7 engine and the whole assembly could now be mounted on the timing-side crankcase.

For the first time, it was now possible to try the engine for size in the newly-acquired frame and to put together a crude mock-up of a V-Twin motorcycle. With the addition of brass plates on the end of both camboxes (where the bevel drive on the Velocette usually resides), a dummy spindle to replace the camshaft and a couple of chains the engine could now be placed in the frame to see if the combination would work.

It seemed that it would although space would be very tight in a number of areas – the proximity of the rear cylinder inlet tappet and the front cylinder exhaust tappet and exhaust port to the frame and the small space available for a gearbox.

It was also fairly obvious that the rear cylinder exhaust and front cylinder inlet were in such close proximity that there would be zero chance of this combination working as it would be impossible to fit either an exhaust pipe or a carburettor.

….and, in any case, I didn’t want this bike to be a Vincent lookalike! And so another plan was hatched….