Jonneh

So I've set up my singlespeed with a rear brake (Shimano 105), alongside the front brake it already had. The trouble is, even when the brake shoes are positioned at the end of their range, the pads rub on the tyre, which isn't going to end well.

I suppose that short of buying a different brand of brakes, the only solution is shortening the chain. At the moment the rear axle is sitting right in the middle of the track ends. Anyone have a rule of thumb to estimate how much a single chain link will alter the horizontal position of the axle?

Broctoon

1/8" axle movement for each chain link removed.

Shortening the chain could be your simplest solution. But how much space is there between the tire and seat tube? Alternatively, you can get a brake with longer reach. It's not about brand, but model. Tektro R559 is a long reach caliper that has worked well for me. It's not crazy-long. Something like 57-67mm, I think.

hairnet

It's 55-73mm, what I use for 27" to 700c conversions that require a fair bit of reach. OP hasn't told us just how much the brake is rubbing the tire. If he only needs a few mm then the Tektro559 might be too long. The 359 is in between TEKTRO BRAKE SYSTEMS - Products, but we need more info from OP.

Broctoon

Jonneh

Thanks for the rule of thumb Broctoon. Intuitively, I thought the movement would be more than that. As far I intuit, it should be one pitch (1/2 an inch) of axle movement per link removed for a cog and chainwheel of zero size. The deviation with common ring and cog setups gets us all the way to your 1/8th inch rule of thumb. Surprising, but that's what these rules are for!

To give some more details, the top edges of the pads make a mark on the tyre about 4 mm beyond the edge of the rim, and there's about 3/4 inch of space between the tyre and the seat tube. It's a thin tyre at the moment (21 mm) which I might at some point replace with a 25 mm one.

Jonneh

Hmm, so I went ahead and removed a link. All's well now with the rear brake - that was more than enough to offset those 4mm. On the other hand, the bolts have moved exactly half an inch forward in the track ends. Now that I think about it, the size of the cog and chainwheel shouldn't affect the predicted movement, which leaves me wondering why the rule of thumb isn't just 0.5 inches of axle movement per link removed, which is what happened in practice?

Yan 

Another suggestion for Tektro brakes.

When I went shopping for them I thought the price was for a single pair, so I ordered two. Turns out the price was for two pairs, so I ended up with four pairs of brakes. I couldn't believe how cheap they were. Very well made too.