69 BSA with the same tls front brake,
It judders a fair bit under heavy braking.
"back in the day" it was routine to turn the brake drums when they started to do this. However, other than sending it out to some shop in Ca with a 6 month waiting list, I can't find anyone locally who has the equipment, much less the experience to clean up the drum.
From your linking your Bonnie's brake with one on your '69 BSA, I'm assuming no confusion on your part about your Bonnie's 'model year' and we aren't talking about a '71 'model year' conical-hub front brake. Therefore, I'm
what "some shop in Ca", (special?) "equipment" and "experience" has to do with fixing the problem? If it isn't a conical hub, the drum's braking surface is parallel with the spindle and any competent machinist should be able to check the drum's braking surface for ovality and cure it if present?
The only possible caveat is the machinist's lathe spindle should be high enough from the lathe bed to take the complete wheel - although some drum brakes can be pulled out-of-round by irregular spoke tightening, Triumph drums were/are substantial; "back in the day", Meriden didn't use specialist wheel-builders but afaik
skimming drums after wheel-building wasn't normal practice; however, if you can find a machinist with a deep enough lathe bed, no sense in dismantling the wheel for the sake of it. Fwiw, last time I needed to have drum-skimming done, a local agricultural engineer had the necessary deep lathe bed and knew exactly what was required.
I had an idea that perhaps I could take some old shoes and glue sharp sandpaper to them, and perhaps clean up the drum a bit by forcing some pressure and rotating the wheel manually
Even if this was practicable, if the drum is oval, by definition, the sandpaper has to be a constant distance from the spindle
, not the drum - you want the drum surface to end up concentric with the spindle
; so the pressure necessary on the brake linkage, shoes and sandpaper would vary depending on how far a given point on the drum was from the spindle.
It is not the shoes - I've tried several different new ones, but actually found the original ones (complete with asbestos) to work the best.
Did you 'arc' any of the new shoes to the drum before testing? If not, you're assuming new shoes' braking surface will be concentric with a half-century-old drum; given the problems we have with new parts, is that a wise assumption? While finding "the original ones ... to work the best" wouldn't cause me to eliminate the possibility of the drum being oval, I'd bear in mind "the original ones" are the shoes most-likely to have worn most-concentric with the drum so should "work the best"?
Unless finding a machinist with a deep-enough lathe bed to take the complete wheel is easy, imho first locate some vernier calipers capable of measuring at least 8" ID, remove the wheel from the bike and measure the drum ID in at least three places 60 degrees apart; only have the drum skimmed if those measurements show the drum to be wildly-oval.
Once you know the drum braking surface is (pretty-much) round, use the aforementioned sandpaper to 'arc' a pair of shoes - sandpaper glued to the drum braking surface so it rubs on the shoes' braking surface - colour the shoes' braking surface with something so you can see when it's all been removed and the shoes' braking surface is concentric with the drum. However, one thing is important - each time you tighten the spindle brake plate nut, the cable must be connected and you must be squeezing the handlebar lever; reason is the shoes 'float' between pivot and cam and you want the 'arced' surface of each shoe to be concentric with the drum when the brake's applied
. Fwiw, I've found it 'easiest' to clamp the drive-side end of the spindle in the timing-side slider, attach the brake cable, apply the brake, tighten the brake plate nut; yes, it's a faff but, when the brake's on a triple, it's being asked to pull up 70 lb.(?) more than a 650 twin ...