Chapter 8. Time to Rebuild

My blog is very quickly catching up with my real-world build but after last week’s heady success its back to a reality check with the realisation there is still a long way to go. So, I have made a “punch-list” of what needs to be done. It is a bit like doing your revision timetable at school, rather than actually doing any revision. Its already filling a page of A4 notepaper and it does not include the obvious items such as “paint the tank”, buy “footrest rubbers”, etc. 

I am still enjoying the moment however and have followed my usual practice of mocking-up the remaining fabrications, whilst I push the bike around whilst it’s still on wheels and admire it – it will soon have to come apart for painting. Before it disappears as an entity, albeit temporarily, here’s Marge with primary chain guard, battery box and even the brake pedal in cardboard.

This week as promised I will provide a bit more detail of the engine rebuild as there are some items of interest. As purchased, and once I had determined there were some internals, I was confident that it would be in good condition. Why rebuild it in the first place – I hear you ask? Although the timing gears and cams were present and correct the mainshaft pinion key was missing (see note taped on the timing case – just in case I forgot) as were the pushrods. This was sufficient reason in my mind to carry out a rebuild, mainly as a check everything was as it should be down below.

The first step was to measure the Compression Ratio – or more exactly the Volume Ratio. At some time, we all ought to get together and have a chat about this. It’s a bit like Tony Foale’s book on frame design, the more you get into it the more complex it gets. As I may have said before my part-time work involves running courses on gas turbines, and being “heat engines” they work on many of the basic principles as an internal combustion engine. One of the ways to get more power from a gas turbine is to increase the pressure ratio. I’ve recently come across an interesting fact about the compression ratio of the Concorde Olympus engine. I would share it with you but since Perry and I started up the Freddie Clark T80 under the Concorde wing, unfortunately when a guided tour was starting, we’re possibly persona non grata. Our excuse that it was raining and not nearly as loud as that ****er didn’t go down so well (I didn’t actually say the last bit but certainly thought it).

Back to the BSA B31-2; the tried and tested way to check the CR for a non-running engine is to measure the cylinder and clearance volume. I know you can do it by inverting the head, filling it with paraffin from a burette, then likewise the cylinder with the piston in place, but I used a simple syringe from Boots and engine oil which is close enough.

I first set the engine so the spark plug hole was approximately vertical, rotated the crankshaft to TDC and added 49 millilitres of oil to the bottom of the thread. I had removed the head once before and knew I had a plus 20 piston, so working from BSA catalogue date calculated the displacement from the bore and stroke (63.2 x 77.7mm) then calculated the CR = (displacement + clearance volume) divided by clearance volume as 6.1. The additional 0.020” piston actually increases the CR marginally from 5.974 to 6.054. 

This ratio is very low by current standards but reflects typical values for the day when petrol was a much lower octane rating. My plan has always been to get the bike running well with a standard engine then look for performance improvements later. I have already found a supplier that stocks a higher compression piston so that could be an easy first step.

Removing the head and barrels revealed nothing untoward. I inspected the valve seats using a spring compressor and the valve collets and decided to leave the head as it is. The engine has had a re-bore and a new piston, in fact it all looks in great condition.

 

 

I laid the engine timing side down on blocks, released the previous owner’s home-made engine studs and started to lift the LH case. The parting joint had been sealed with a very fine smear of Red Hermetite, not my favourite jointing material and I used something different on the re-build. A few gentle taps with a hide mallet were sufficient to separate the two halves. The mainshaft fit to the bearing felt about right – i.e. requiring a few gentle taps on the crank-shaft end as I eased the case up. Similarly, the crankshaft to timing side bearing was well located. I had checked the big-end play on the way down and decided that as everything looked in great shape to leave well alone.

 

I think I may have already mentioned that I had set the engine up on my lathe and rotated it slowly – 70 – 80 rpm. This allowed me to see that the oil pump was working as the tell-tale plunger came out smoothly as I increased the revs. Although dropping the pump from the case I had discovered only one pair of gears in the pump -i.e. there was a pressure feed to the crankshaft end but no scavenge loop from the sump. I knew the oil tank was integral to the front of the engine case but had wondered how the oil was returned from the crankcase to this tank.

With the engine apart all became clear. As you can see, particularly from the timing side pic, there is a scraper ring close to the flywheel and this is the return mechanism. It remains to be seen how well this works in practice but there’s a lovely little oil supply control on the timing case which should allow me to fine tune the system. 

Of course, having got this far, I then came across some period BSA literature that described this engine type in sufficient detail to have a full understanding of how it works. You will notice on the drive side case there is a small port just behind the crankcase mouth – this feeds another control valve for the primary chain lubrication. There is also a drilling through the mainshaft, coming out under the engine sprocket internal diameter – presumably to lubricate the cush-drive? Anyway, an interesting journey of discovery and re-assuring that I required no new engine parts and the condition was generally excellent. I guess I must have completed the strip-down and rebuild in one day – oh for the simple life.

I fabricated new pushrods using some used items I got at Kempton and replacing the alloy rod with longer sections, nominal 8mm diameter, pressing them into the steel pushrod ends. This was more difficult than I had imagined; getting the correct interference fit, the rod fully home in the pushrod end and resulting in the correct length. So I had first made up some dummy rods and used small washers as shims to get the exact length with the valve gear in place. 

I had also researched into whether aluminium alloy or thinner high tensile steel tube would be better and came across an interesting  thread referencing some testing by Peter Williams and Professor Gordon Blair; https://www.accessnorton.com/NortonCommando/pushrod-strength-test-alum-vs-steel.21936/. Some of these on-line “conversations” are guys who are not experts but trying to prove they are. This one at least gives the impression that it’s serious bike racers talking. It would tend to suggest that HT steel tube is best, but since the BSA will not be revving that hard I went with the alloy and I’m keeping my fingers crossed that they won’t bend or flex too much in use.

With the assembled engine on the bench I checked out the valve timing, using a Test Dial Indiactor (TDI) mounted above the head and a printed timing disk on the crankshaft. I could not find any BSA timing data for a 1932 BSA B31-2, but worked from Phil Irving’s Tuning for Speed and fortuitously a well-timed (pun intended) VMCC magazine article. I then ran an on-line calculator to produce the printed disc.

These are the figures I used with the measured data in brackets;

Inlet Open 25 BTDC (26.5) Max 110 ATDC (107.5) Close 65 ABDC (67.5),

Exhaust Open 65 BBDC (55) Max 70 ABDC (74) Close 25 ATDC (22.5).

The exhaust is a little way out but given that the cam wheels have 36 teeth – i.e. moving it one tooth back would be equal to 20 degrees of crankshaft rotation  - I could not get it better without making a stepped key. It is more important to use the maximum lift than the open / close point and that’s only 4 degrees out. I purchased a couple of end-mills and made up a pinion gear key on my lathe and have Loctited the nut in place. The photo shown is actually machining a stepped key for the magneto pinion. I’m less concerned about my home-made mag key shearing than I am the cams

 

 

After my last blog article someone said I must have the patience of a saint, it came from an “unknown” address, but thank you anyway. The reality is I love being in my workshop sorting out these problems. I also know I’m very lucky having the time to do this, and the tools, and the workshop itself. It’s now many years since I rebuilt a Triumph engine in a mate’s one room flat and carried it down several flights of stairs before sticking it back in his bike that was sitting on the pavement, and it was raining, and in the wrong part of Croydon, but I have been there.

All the best stay safe

Gareth