Technical FAQ with Lennard Zinn: A detailed look at canti' geometry

Published October 26, 2010 10:26AM

The review says that the brakes set up in the narrow stance provide around 20 percent more stopping power when set up properly. I thought about this statement because when I was browsing Paul Component’s canti brakes, they say that the wide-profile ones are stronger than their narrow (touring) canti brakes.

This confuses me because I always thought it is the basic design (wide vs. narrow stance) of the brakes that result in the increase or decrease of braking power. In this case, SRAM and Paul seem to have conflicting statements.

Do their statements only refer to their specific brakes? In general, do wide or narrow stance canti brakes have more braking power?
Louis

Dear Louis,
Thanks so much for such a great question! I love these ones that take some thought. I got a bunch of questions about Caley’s article on those brakes, and I am going to combine my answer to yours with the answer to this one as well:

Dear Lennard,
I just read the Wrenched & Ridden article on the new Avid Shorty Ultimates where the author speaks about the power and modulation of different cantilever brakes. It seems to me that the power issue boils down to simple leverage. “X” amount of pull on the brake lever over “Y” amount of movement at the brake pad gives a ratio that represents the mechanical advantage of the system. The higher the number, the more power. Narrow stance brakes give a lower “Y” and higher mechanical advantage ratio. Wide stance brakes give a higher “Y” and a lower mechanical advantage ratio.

Dear John,
What you say is absolutely true, but it does not just happen that way without the person who sets the brakes up having a major impact on the leverage ratio.

And Louis, this is where the answer to part of your question comes in as well. Nobody said it better than the late Sheldon Brown in his eloquent description of the geometry that determines the leverage ratio of cantilever brakes and how to set them up to achieve the best balance of power and ease of use. And John, in it, you will find out why, while you are absolutely right on the leverage ratio (and the fact that you want a spongy feel to the brake, indicating a high leverage ratio), it takes knowing what you are doing to set them up to actually achieve a high leverage ratio.

I think you should read Sheldon’s entire treatise, but to cut to the chase (with some detours to define terms), he says:
“Conventional cantilevers fall into three types, defined by their cantilever angle:

• Wide-profile cantilevers have a cantilever angle much greater than 90 degrees. The best example of this type is the old Mafac cantilevers, in which the anchor arm actually sloped downward from the boss in some installations. This design is now pretty much obsolete. Wide-profile cantilevers have rather low mechanical advantage, and work well only with levers with a high mechanical advantage.

• Medium-profile cantilevers have a cantilever angle of around 90 degrees. Most late-1980’s cantilevers belong to this family. Medium-profile cantilevers are very forgiving and give excellent all-around performance with a wide range of set-ups.

• Low-profile cantilevers have a cantilever angle of less than 90 degrees. The principal advantage of narrow profile cantilevers is that they don’t stick out so far from the frame or fork, very desirable, because protruding cantilever arms can cause a multitude of problems, particularly in the rear, where a rider’s feet may hit them. Narrow-profile cantilevers are also capable of excellent performance, but only if properly set up. A carelessly set-up low-profile cantilever may have very low braking power, even though it feels great on the workstand.”

In the case of both Paul and Avid, we are only talking about the first and last ones in this list – wide-profile and low-profile cantilevers. Avid Product Manager Paul Kantor told me that:
“Avid Shorty Ultimate in the wide stance uses a cantilever angle of 115 degrees.
Avid Shorty Ultimate in the narrow stance uses a cantilever angle of 70 degrees.”

Now that you’ve read Sheldon’s piece, you know that the low-profile brake can be set up to have a higher leverage ratio than a wide-profile brake, and hence more power, and why. You also know that you get a higher leverage ratio by using a lower straddle cable, but if you have it too low, it can clog with mud easily, and it is hard to release the brake since there’s not enough slack in the system to pull the straddle cable out of the end of the brake arm (unless you have Campagnolo or Cane Creek levers with the release button on the lever).

Avid indeed claims more power in the narrow stance, and below are Avid’s actual power numbers, supplied by Kantor, from dyno testing of the brake in the two different setups. He says that, “the cable anchor was placed at 30mm above the tire on the narrow stance, 50mm above the tire on the wide stance.”

Obviously, in this standard straddle-cable setup for each brake, the narrow-stance brake is more powerful. Calculating at either end of the range, you can see that the leverage ratio is also not linear over the range for either brake setup. At low (40N) lever force, the wide-stance brake’s ratio is 122/40 = 3.05/1, while the narrow-stance brake’s ratio is 173/40 = 4.33/1. At high (160N) lever force, the wide-stance brake’s ratio is 471/180 = 2.62/1, while the narrow-stance brake’s ratio is 578/180 = 3.21/1. From Sheldon’s description, I think you can understand why the leverage ratio decreases with more forceful lever pull – because the yoke angle becomes higher.

You can compare the yoke angle before the brake is applied in Caley’s photos of the Avid in wide and narrow set-ups.


Lever force (N)	40	60	80	100	120	140	160	180
Braking force (N)- Dry- Wide Stance	122	168	208	259	312	386	425	471
Lever force (N)	40	60	80	100	120	140	160	180
Braking force (N)- Dry -Narrow Stance	173	232	279	336	408	476	534	578

Now, to answer the other part of Louis’s question. The Paul Components Neo-Retro brake is akin to the Avid in the wide stance, and the Paul Touring Canti is akin to the Avid in the narrow stance. The Avid has exactly the same “anchor arm” length (PA) and “shoe arm” length (PS) in both configurations. I am not at all sure from looking at photos that the two different Paul designs share those common dimensions, and I don’t have any I can measure. It is possible that the Neo-Retro could have a longer anchor arm length than the Touring Canti, which could give the leverage nod to the wide stance Neo-Retro brake, but, if anything, it looks the opposite to me in the photos. And the shoe arm length looks similar in both models. So I would guess that the narrow-stance Touring Canti, if set up with a low straddle cable as described by Sheldon Brown (or using the two straddle-cable setups described by Kantor for Avid’s dyno testing), would actually be more powerful than the wide-stance Neo-Retro – the opposite of what the Paul site says.

However, there is one thing I want to point out that I think might clarify a lot. The Paul site describes how the wide-stance Neo-Retros flex flimsy seatstays out more than the narrow-stance Touring Cantis do. It attributes this to higher power in the Neo-Retro, but I actually think it is due to something else. Look at the attached beautiful diagram of the two configurations of the Avid Ultimate atop one another that the other Paul – Paul Kantor of Avid – sent to me. For the sake of this discussion, that diagram will also work when describing the two Paul Component brakes. The thing I want to point out is that, when the pad contacts the rim and is being pushed hard against it, the fulcrum effectively is no longer at the brake pivot (the cantilever boss) – rather, it is at the face of the pad where it is touching the rim. As the cable pulls harder, the pad face can’t move any more because the rim is there. But, like at the other end of a teeter-totter pivoting at the pad face, it is pushing the cantilever pivot away from the wheel. Since the pad face is now the fulcrum, the effective lever arm length is from the pad face to the anchor point of the straddle cable at the end of the arm. But notice the distance from pad face to cable-anchor point is considerably longer on the wide-stance brake. I think that this is the reason that wide-stance Neo-Retro brakes flex the seatstays out more – because the lever arm pushing the cantilever pivot (and hence the seatstay) away from the wheel is longer.

I had a 2007 Van Dessel Gin & Trombones. It came with a carbon straight-blade fork with carbon steerer. I tried Cane Creek short cantis, IRD high cantis, and finally, my current setup was Paul Neo-Retros with Salsa cable hangers (50mm wide) set 3 inches above the tire front and rear with long Kool-Stop Thinline pads. These changes all did quite a bit of good, but the chatter remained.

So, I called Van Dessel and they sent me an OEM Ritchey fork (Comp Carbon Cross but without eyelets). This fixed everything. I have had no chatter since, and I’ve even carried the fork and setup over to my Lynskey Pro+Cross. I would assume the bladed arcs of the fork legs help prevent the flex that leads to chatter.
Jordan

Dear Jordan,
Actually, I’m quite certain that it is the stiffer steering tube on the Ritchey fork that fixed the problem, not the curved fork legs, since what happens below the cantilever posts is largely irrelevant to brake chatter.

It’s possible that a stiffer crown area on the Ritchey fork contributed as well, but it’s unlikely. That’s because of two reasons:

(1) Most forks are quite stiff from the crown to the cantilever bosses (due to the pair of blades that are deep front-to back), and,

(2) the distance between the lower headset race and the cantilever posts is short. Because of these two reasons, using a cable hanger that bolts onto the fork crown (thus eliminating the steering tube from the equation) usually prevents brake chatter.

Since my original post on this was long, you may have missed the essential point, namely that it is the flex over the total distance from the cantilever posts to the attachment point of the cable hanger above the headset that creates the problem. If this length increases in front due to the steering tube, crown, and upper part of the fork legs bowing as the fork flexes backward due to application of the brake, the front brake cable will become tighter. This will squeeze the pads harder against the rim, even if you are trying to modulate the pressure with your hand on the lever.

If the traction is good enough that the tire does not slip on the ground, and if your weight is far enough back that the extra braking doesn’t cause you to flip over the handlebar, then the wheel will be forced to rotate, momentarily breaking free of the pads, followed by being grabbed again by the pads, then grabbed harder by the pads due to the bowing in the upper part of the fork tightening the cable, etc. The cycle repeats until you release the brake lever (or the fork breaks).
Lennard
Readers can send brief technical questions directly to Zinn.