There has been a lot of excitement on the BOB, Kogswell, and Rivendell lists recently about the new (but not yet available) Tektro 556 long reach dual pivot brake. This is a pretty cool brake with a reach of 55-73mm, a good 16mm more reach than the now common 47-57mm reach dual pivots. Longer reach allows for fenders and bigger tires.
Here is a nice drawing of the brake from Kogswell:
The red line shows a fender and the purple shows a 38mm tire. Kogswell is so excited about this new brake that they plan on building all of their future frames around it. I'm not that excited. Here's why.
Long reach caliper brakes -- design problems
Long reach caliper brakes have much longer arms than the 39-49mm and 47-57mm reach brakes that we're all used to. The longer arms either need to be thicker (and thus heavier) or they are more flexible. You might think "Alex, whats the deal, the brake is only 16mm longer". True, it is only 16mm longer, but the arm length increases by more than 16mm. The width of the brake must be increased to keep the same mechanical advantage (and to fit around wider tires and fenders). If only the brake reach was increased then the mechanical advantage of the brake would drop. You can see that the whole brake is scaled up if you compare a 39-49mm and 47-57mm Shimano dual pivot.
How much longer does the arm get when you increase brake reach by 16mm? I didn't know before writing this post, so I measured. I have some Shimano 47-57mm reach dual pivots and some Diacompe FS-E 57-73mm reach dual pivots. I measured the length of the arm holding the left pad from the pivot point to the bottom of the pad slot. On the Shimano brake it measured 117mm. On the Diacompe it measured 157mm. So 16mm of reach cost us 40mm in arm length between these two particular models. That is 40mm of extra arm that can flex.
The longer pad slots of over 57mm reach brakes also cause problems. The section of the arm where the pad is located has to be constant thickness and the thickness is determined by the design of standard threaded pads, not the brake. There will be considerably more flex at the bottom of the slot than at the top. You can easily see the difference in this photo:
Note how long the thin pad-mounting area of the black brake 73mm reach brake is compared to the silver 57mm reach brake.
Cantilevers
Okay, we've discussed why I'm not crazy about super long reach brakes. Why is there a demand for them? It is because many cyclists don't like the most common alternative, cantilever brakes.
There are a couple of reasons that people state for disliking cantilevers:
- Complicated setup
- Squeal
- Asthetics
- Fork design
Cantilevers do have a complicated setup, especially the designs from the early 90s that most people are familiar with. The pad setup is complicated because most deisgns use a single fixing bolt adjusts five different variables -- pad height, rotation around the pad, toe in, distance from the brake to the rim, and angle compared to the rim. If you want to adjust one of these you end up adjusting all five. Modern cantilevers use threaded pads which reduce this to four variables (distance is fixed) and just eliminating that one variables makes setup a lot easier. Some good brakes (such as the Avid Tri-Align) from the mid 90s had nice designs that used multiple bolts per pad and reduced the number of variables to two or three per bolt.
Pad setup isn't the only complicated factor about cantilever setup. Cantilever brakes use a straddle cable and the straddle cable height changes the mechanical advantage (power) of the brake. This is both complicated (because you need to understand the relationship) and very flexible, because you can tailor the brake to your exact needs. Most people don't care about this flexibility and are happy with the one mechanical advantage given to them by dual pivot calipers. Modern cantilevers typically use a "link cable" instead of the straddle cable. This fixes the height of the straddle cable and eliminates this complication while giving you a similar mechanical advantage to a dual pivot brake. You can learn more about cantilever geometry on Sheldon Brown's The Geometry of Cantilever Brakes article.
The complicated setup of cantilevers results in squeal on a lot of bicycles. It is easy to get the brakes working well enough, except for squeal. Cantilever brakes (especially inexpensive ones) typically have more play in them than caliper brakes and will squeal. Cheap calipers (like the Diacompe FS-E described above) also have lots of play and will often squeal too. For the last 25 years there have been many more cheap cantilevers produced than cheap caliper brakes (due to the mountain bike boom), so most people associate squeal with cantilevers.
It takes time to learn how to setup cantilevers, but with quality brakes you can rid of squeal and easily get a good setup. It doesn't necessarily take a lot of money to get quality brakes. The Shimano cantilevers on my Rivendell Quickbeam retail for about $20 for a bikes worth and don't squeal. They use a link cable so straddle cable setup is easy. They do have the old pad design, so pad setup is still tricky.
Matthew from Kogswell isn't a fan of cantilever brakes because it gives him less flexibility in fork design. With caliper brakes the load is all in the fork crown, so running thin fork blades is not a concern. With cantilever brakes he needs thicker blades (at least down to the cantilever studs) which can resist the forces from the brakes. He thinks that he can design more comfortable forks if he uses thinner blades, but he can't use thin enough blades with cantilevers.
I agree that this can be a concern, but I think that it is a minor one. Almost all of the flexibility in a fork takes place at the bottom of the steerer, not in the fork blades. Some engineers say that the fork blades don't bend at all. I don't know if that is true, but I can say that my bikes built for cantilever brakes don't feel stiffer in the forks than the ones built for caliper brakes.
This brings us to asthetics. Some people find cantilever brakes to be ugly. As functional bike parts I don't really find them to be any uglier than a crank, pedals, or rear derailleur. All are essential parts of the bicycle and fit a specific and useful roll. I don't find caliper brakes to be more pleasing to look at.
Okay, so now we've talked about why people don't like cantilever brakes. Why do I prefer them?
Cantilevers have much better clearance than any caliper. Cantilevers can clear tires up to about 2.5" in diameter, including fenders. I like having the option of running wider tires on my bikes, even if I don't always need to, and would rather have the frame be the limiting factor than the brakes.
Cantilevers have a fixed arm length from the pivot to the pad. This length is pretty short on most designs, with allows them to be stiff without being overbuilt. Cantilevers for 45mm tires and fenders are the same as cantilevers for 25mm tires and fenders. Caliper brakes for 45mm tires and fenders are either going to be much heavier or much more flexible or both compared to ones for 25mm tires and fenders.
Cantilever brakes are adjustable for mechanical advantage. Most of the time this isn't necessary, but sometimes it is useful. On a load carrying bike I can run a higher mechanical advantage for quicker stopping. On a bike with a long run to the rear brake (such as a tandem or recumbent) I can run a lower mechanical advantage to prevent the brakes from bottoming out due to cable stretch.
The cantilever design has been around for a long time and the pivots are still used by V-brakes. Millions of these brakes have been made. In 20 years I'll have no problem finding replacement brakes for a cantilever-equipped bike that I buy today. 57-75mm reach dual pivots have been tried before (Shimano Nexus and Dia-compe FS-E) and abandoned, and now are being tried again. In 20 years I doubt that replacement brakes will be made available. If you buy a frame which requires them today you can stockpile a couple of replacements for future use.
Conclusion
It is interesting to see how excited the community is becoming over a brake that none of us have ever used. It will be interesting to see how things look in 6 to 12 months after riders have put a couple of thousand miles on this new brake. Maybe I'll fall in love with them and take back much of what I'm saying in this article. Maybe others will understand why I'm not excited about the brake.
I hope that Kogswell and other builders consider selling bikes which work with either type of brake, at least until the new brake proves itself in the marketplace.