primary chain adjustment - 1962 88SS

I understand it's from slight eccentricity in the sprockets. Worn chains are often blamed, but the chain length is the same wherever it is. But sprockets are often not perfect. The difference you describe sounds a lot, but it must be slack enough not to go bar tight anywhere when the cases expand when hot.

The gearbox mainshaft  could be bent , you should be able to rotate to see this .

My Dad taught me that it should be checked all round the chain run simply because the chain does not necessarily  wear evenly, especially if it has any 1 1/2 link sets riveted in . I have never seen sprockets wear unevenly.

If there was eccentricity  in a sprocket the tight spot would come around each rotation of that sprocket rather than the chain itself, and I think that if it was running out enough to do that then you would almost certainly know about it in other ways.

I run my primary chain  on the slack side  ,but not enough to touch the Alternator housing. My reasoning is that this allows the chain to dip into the oil even if its a bit low. Also if the box gets pulled back a little by the rear chain after adjustment it not so likely to damage mains/clutch/ box bearings  The clutch  frees better if its not pulled about by a tight chain . Never had a primary chain issue.   I usually check the tension  with a little pressure on the kickstart ,those with mag ignition need to be very carefull  of fingers !. Rear chains I like to tension a bit close , A slack rear chain on a DL Dommy can  clip the case  and throw off the spring clip  and break a link  .  Still we always have the ever efficient RAC  !    yeah right .  Better to carry a spare link than spend an evening in the cold and dark waiting  for nothing.

Usually I find more uneven chain stretching on the rear chain. I think that's caused by shock loads riding on bumpy roads or off road. Some modern high end chains are advertised as pre stretched. The primary chain is less prone to wear and stretching than the rear chain, It is enclosed and gets oiling. The pulling force is much less as it runs at higher speed than the rear one. I've found that even if you sets primary chain tension correct, using the right procedure, it could change a bit after a test run. A second adjustment might be needed. I believe that your primary chain should be replaced unless it is a good quality, fairly new one. Not worth the risk of hitting the alternator housing or wiring or having it tight spot when fully warmed up. My Manx still has the dents in the primary chain guard it got when the chain broke on the 6th lap at the -62 Junior TT. DNF.

My undertanding is that  chain wear is much more evenly distributed if you have the right combination of sprockets  to   give the "hunting tooth " effect.  I'm sure one of our clever  lads knows all ?.  Gives me a headache thinking about it.

... and I wonder if it was why some clutch sprockets (and rear wheel sprockets) had 43 teeth which is of course a prime number giving that hunting toothe effect? Begs the question as to why mostly they were 42 teeth however.

... the only person here who doesn't know what the "hunting tooth" effect is ...?

Hunting tooth, all you need to know is as long as at least one sprocket has an odd number of teeth then you have a hunting tooth. You get a better effect if both sprockets have an odd number of teeth but one will be enough. The hunting comes from the odd tooth sprocket moving the chain on by one tooth every turn. 

It's different with gears and chains. With gears, it's best to avoid tooth ratios which mean that the same pairs coincide frequently as the gears rotate. So its best not to mesh say 21 with 42 in a gearbox. With chains it's the ratios between chain links and gear teeth  and how many times they rotate per minute before the same tooth meets the same link. So a 20 tooth sprocket with 100 link chain means the same links meet the same teeth all the time, whereas with 102 links it takes more revolutions to meet again. (10 times?)

Inters and all but the last Manx bevel drives are poor in this respect because the vertical shaft has half as many bevel teeth as the bottom drive and top cam bevels. That makes it easier to assemble but increases the rate of wear both because the highest loads keep arriving on the same teeth  and also any defects in tooth profile repeatedly damage the same teeth.