Rebuilding the engine of KTT 305 – Part 3

Most of the engine had now been rebuilt and there were just a few remaining tasks to complete the build.

The next job was to set up the valve timing. Much has been written about this in the past and there is a perfectly good description in the “Instruction Book for OHC models 1932 to 1935” so I won’t repeat it here. I always check valve timings after setup with a degree marker on the crankshaft. I have a few of these with different sized holes in the centre for different engines and they are made using an Excel spreadsheet to generate a 360⁰ pie chart, printing it onto an A4 sheet of paper, gluing it to a piece of cardboard and cutting a hole in the centre with a scalpel.

I use an indicator in the spark plug to find TDC (fine-tuned by getting the same piston displacement at +/- 60⁰ of TDC – it’s much more accurate than using TDC directly)

So, what should the valve timing be? The following table, taken from Fishtail 129,

summarizes the valve timings for all the Velocette cams. This data is also given in Phil Irving’s “Tuning for Speed”, shown below.

Acknowledgement to Temple Press Ltd, Tuning for Speed by Phil Irving

The data is generally in agreement although Phil Irving does not give data for the earlier OHC Velocette cams.

The “checking clearance” is very important for measuring the valve timing as the cams have long opening/closing ramps and totally misleading figures will be obtained by using the wrong clearance. I check when a 0.002” feeler gauge “nips up” and the results that I got are as follows:

IVO – 55⁰ BTDC

IVC – 68⁰ ABDC

EVO – 72⁰ BBDC

EVC – 32⁰ ATDC

I believed this cam to be a K-17/5 (There are no markings on any of my K-17/5 cams and I bought this one on eBay – so not entirely sure. The earlier K-17/2 = #24 are usually stamped for identification) but the EVO and IVC figures for a K-17/4 cam – the cam used in 1931 KTTs (which this is) are stated as 68⁰ BBDC and 70⁰ ABBC respectively which are remarkably close. These 2 values should be considered the most accurate because the IVO and EVC figures can be affected by the loading of one cam on the other during the valve overlap period; this is why Velocette advise checking with each rocker removed in turn. However, I did not want to do this because of the possible difficulty in getting them back in place without damaging the felts on either side of the rocker.

I was satisfied that the valve timing was OK and the final step was to turn the crankshaft to TDC firing (because both valves are closed), remove the cambox, insert the K-49 Vertical Shaft into the K-50/2 Cover, put the gland nuts on each end and reassemble without disturbing the timing. Much has been written about sealing of the vertical shaft: for better or for worse I have use small nitrile O-rings backed by a PTFE ring

and with a couple of rounds of asbestos-substitute string

soaked in graphite grease. It is important that the proper spanner is used for the gland nuts

to avoid mutilating them. I inherited a couple of these with some spares I bought many years ago and they are available from VOC spares (part number VSL364). Just for the record, the PTFE and nitrile rings came from a buddy of mine that also used to work at Ricardo and he used these successfully on his KTP and an early KSS.

The next step was to fit the K-72 magneto drive gear and the K-96 oil pump driving piece. However, I have now partially stripped my latest acquisition, KTT 581s engine, and found that the corresponding parts have a couple of steel disks sandwiched between the gear and the driving piece, shown below.

Why? I cannot find either pictures or part numbers for these discs in any of my parts books for up-to 1934 OHC or for Mk 2 KSS/KTS models. The KTT 581 engine has hardly been used – there is no discernible wear on any of the components, there are no nuts with rounded hexagons etc. and I am almost certainly the first person to strip this engine since it was built in the factory at Hall Green in 1935. This means that the 2x steel discs (total thickness is 0.108”) were factory fitted and I think I can understand why…

If I insert the magneto drive gear and pump driving piece onto the end of the pump on KTT 305 with no discs then the engagement of the gear with the crankshaft pinion is poor – see picture below.

and there is quite a bit of end float on the shaft when the cover is fitted. I tested this with the 2x discs from KTT 581 and the engagement was pretty well perfect.

I therefore ordered 2 packs (many more than I need in each pack – but that was what was on offer – and I have more engines to build - and they were remarkably cheap anyway) of 1.5mm and 1.2mm x 25mm diameter stainless steel discs

and fitted 2x 1.2mm discs (which gives the same total thickness as those used on KTT 581 to within a couple of thou)

with the following result.

Much more satisfactory engagement. There is still, but now much reduced, end float on the shaft and so I think this is a good solution.

If anyone can find a reference to use of these discs anywhere in Velocette documents, I’d much appreciate it – just out of curiosity.

In my entire collection of bits there was no correct oil feed pipe into the engine, part number K-119/2. Again, KTT 581 was able to provide a pattern to copy and it took only a couple of hours to make 3 of these – one for each of the engines being built.

and after chemical blacking as per the original…

The Magneto, a square ML, has been rebuilt but it had the wrong advance/retard plunger fitted and so a new one of these was made.

2 of the 3/8” BSW nuts on the magneto require reduced-hexagon nuts to be fitted – full hexagon nuts are too large for the available space - and so 4 of these were made and blacked.

A new timing chain was fitted

and the timing set up at 38⁰ BTDC fully advanced.

The AMAL TT carburettor was stripped, cleaned and rebuilt

and that just about completes the engine rebuild for KTT 305.