3

Context

I am fixing up a BSO where the rear cantilever brake is missing. I want to keep this project green and cheap, and there is a good supply of used cantilever brakes online. But they often come without the screws or washers for mounting to the frame, and my original ones are missing too.

Concern

It's easy enough to pick some screws and washers from my box of odds and sods, but it's a brake, not the place for a mistake (check out the rhyme - if a brake company use this line for advertising, I want a free brake).

The front brake is good so I observed it in action - it's even captured in this video I took. I pulled the lever several times and I noticed that the washer doesn't move when the brake pivots.

The photo below is just a snippet from the video. The photo is not very clear but the video is better.

cantilever brake - note black marker lines

Speculation about design intentions

It's a simple matter and it could be accidental but I can easily see this being important: it seems that the washer is designed to stay still, to reduce the risk of the screw becoming undone. It seems that the sizes, materials etc were picked so that the washer has more friction against the screw than the caliper. I believe the caliper face of the washer starts slipping under very low forces, therefore very little torque is then passed onto the screw.

Hypothetically, if the washer's friction was higher against the caliper than the screw, then every application of the brake would gradually loosen the screw on one side and tighten it on the other.

I don't want to get into the physics - I assume the factors at play are the materials' static and dynamic friction coefficients, contact area and it's distance from the pivot axis etc etc.

Primary question

So how do I pick the right screws and washers from the parts bin?

My guess at a strategy

My guess would be to use a trial and error strategy:

  1. Get a combination of screw and washer that physically fits the frame and caliper.
  2. Check if the washer moves with the caliper. If it does, go back to step 1. If the washer stays put, happy days!

My guess at selection criteria for step 1

And here is the list I can think of for getting a combination that fits:

A. Washer internal diameter needs to be close to screw thread external diameter.

B. Washer external diameter needs to be close to the size of the recess in the caliper.

C. Screw thread length needs to just under the depth of the female thread in the pivot/stud/boss.

D. Screw head size is a compromise between:

  1. the contact area with the washer wants to be big,
  2. it wants to keep clear of the caliper itself.

E. Get some threadlocker just to be on the safe side.

Secondary question

Is there something I can put on the caliper face of the washer? I wonder if it would encourage movement to the correct face of the washer. Just grease?

Closing remarks

Finally, I'm sure someone will spot two issues with this bike, but please let me know if you see more

  1. I know the brake pivot is missing on the non-drive side but I'm planning to use this answer for this.
  2. The front brakes seem to be set up for a poor mechanical advantage at the moment.

1 Answer 1

5

Cantilever brake studs are a standard dimension, and there are two different ways that cantilever brakes utilize them:

  • The frame stud is a dynamic surface. Each brake arm has a brass bushing running through that's just a little shorter than the stud is long, such that 0.1-0.3mm-ish of the stud extends past the top. The mounting bolt then tightens down the washer against the end of the stud. Any torque imparted in this design is very minimal. Yes the arm might contact the washer but the washer is not pressing down on it in any way. This kind of brake arm needs the mounting bolts torqued very low (2 ft-lbs appears on some torque charts) and should have threadlocker applied. This type is more common and is on the brakes you have.
  • The frame stud is not a dynamic surface. The arm parts extend past the top of it. You tighten the bolt and washer down on it directly (6 ft-lbs/8 Nm or so). The pivot is contained in other parts in the brake, which allows tighter tolerance control and control over the material and finish of both parts, not just the bushing. Many but not all higher-end cantis since the 90s use this design, as well as some mid-range models. There is not a consensus name for this type; I usually refer to is "integral pivot." Some examples are the Tektro Oryx and all the Paul cantilevers. Some brakes of this sort are particular about the size or shape of the bolt head, for example the Paul design where it nestles inside the spring tension adjuster, and others are indiscriminate about it.

For the design you have, where the stud is the pivot, what's really important and where things can go wrong with picking hardware is that the washer needs to fully cover the tip of the stud as best as possible. If the hole is a little too big, what can happen is that the washer gets deformed when you tighten it and start to contact the arm when it shouldn't be, which binds the brake up. Another piece of this is that some studs are nice and machined flat on top, which makes this problem less likely, and some are more rounded, which means the actual amount of contact area with the washer is very small at the best of times. All that said, a proper M6 good quality grade 8 or stainless washer from the hardware store and the matching M6 bolt should usually be fine, plus medium-strength threadlocker. For length, the manufacturer-supplied ones are anywhere from 16 to 24mm. Also if you have a common M6 disc caliper mounting bolt and washer around in the 16mm-ish length, that's fine too, it's basically the same thing as many manufacturer-supplied mounting bolts. Cantilever mounting bolts are visually prominent so they're usually domed and relatively pretty, but that's all that differentiates them from other M6 bolts with 5mm allen heads on bikes.

3
  • Great answer. I waited a day before reacting, but it might be hard to beat it substantially. It all makes perfect sense. I’m puzzled why this info doesn’t seem to be on Sheldon brown’s website, I really thought it contained practically everything, including some much more obscure detail.
    – pateksan
    Nov 23, 2021 at 23:57
  • Enough waffle: could you add (maybe edit in?) some detail about the second type, which you mentioned tends to be in “higher-end cantis”? Does this system have a name to differentiate it, like we have hyperglide vs uniglide? Just how higher-end do you mean? I ask because one of the cheapest listings I found online is for a Shimano STX brake which I think counts as mid-range, could it have it? And is there a similar rule of thumb for dealing with this type, similar to the washer situation for the other type?
    – pateksan
    Nov 23, 2021 at 23:59
  • 1
    One thing about Sheldon's site is the period he was writing is when basic low-profile cantilever brakes could seem like the only brake in the world, because they were on the vast majority of bikes in the early to mid nineties MTB boom. There were some cantis and v-brakes brakes with integral pivots around, look at the various XTR and XT v-brakes for example, but they weren't the norm unless you only touched race bikes or something. Nov 24, 2021 at 2:11

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.