Straddle Cables Done Right

Straddle Cables Done Right

Straddle cables provide a light and elegant way of transferring the brake force: Every cable-actuated rim brake needs to transmit the force of the single brake cable onto two brake pads that squeeze the rim.
In recent years, straddle cables been replaced by direct-action V-brakes or complex linkages (on modern Shimano sidepull brakes). There are reasons for this: Current practice for straddle cables is less than optimal. However, these flaws can be eliminated with good design, resulting in brakes that are lighter and more powerful than the alternatives.
Above is a typical straddle cable arrangement. The straddle cable is as thick as the brake cable, and that makes it very springy. You can see how it bends around the cable hanger in a gentle arc. When you apply the brake, you first have to straighten the straddle cable. This is lost motion – you pull on the brake lever, but you don’t get any brake power yet. If there is too much “lost motion”, you risk bottoming out your brake levers against the handlebars. To prevent this, you have to set your brake pads very close to the rims. Experienced mechanics “pre-bend” the straddle cable, so it better conforms to the cable hanger, but it’s always going to have some of that springiness.
Why is the straddle cable so thick and springy? It carries only roughly half the load of the brake cable, so it needs to be only half as strong. A thinner cable is less springy and conforms much better to the bend of the cable hanger. The top photo shows our Compass brakes, which use a thin shifter cable as the straddle cable. You can see how straight the cable runs. When you squeeze the brake lever, there is no lost motion.
There is a reason to use a thicker straddle cable: to prevent it from fraying where it clamps to the brake arms. The angle of the cable changes here as you apply the brakes: The cable becomes more vertical as the hanger moves upward and the brake arms move inward.
If the cable is clamped firmly to the brake arms, you bend the cable every time you brake, which eventually may cause it to fray. The angle change is more severe on centerpull brakes than on cantilevers. And when the cable frays, it’s only a matter of time until it breaks, and then you lose all brake power.
There is a better solution to this problem: cable attachments that swivel. Then the changes in angle don’t bend and stress the cable at all. You can use a thin straddle cable, which doesn’t “spring”, and you’ve eliminated all the disadvantages of straddle cables, while keeping their advantages.
With optimized straddle cables, our Compass brakes work effectively the same way as Shimano’s latest racing sidepulls – the two pivots are next to the tire, and the lower arms are short to offer great braking power. Yet the Compass brakes use straddle cables instead of complex linkages, so they are much lighter than Shimano’s racing brakes, and they have less friction. When you use them on the road, you can feel the difference.
Click here to find out more about Compass brakes.
Photo credit (PBP photo): Maindru, used with permission.

Share this post

Comments (76)

  • Christian Bratina

    How does the performance of the Compass Centerpull compare with the Paul Racer Centerpull brake? Our steep dirt road descents at speed require tremendous braking power at times.

    April 4, 2016 at 2:52 am
    • Jan Heine, Editor, Bicycle Quarterly

      The Compass brake is optimized for braking power, so I doubt you’ll ever find it lacking. On dirt, your tire traction limits your brake power, so you don’t need brakes that are super-powerful. More important is good modulation and light action, so your hands don’t fatigue. However, the Compass brakes also excel on dry pavement, where you can get tremendous brake power, since tire traction is almost unlimited.
      The Paul brakes also offer very good braking power. Their disadvantages lie in their greater weight and limited tire clearance, as well as the difficulty to integrate a front rack in an elegant way.

      April 4, 2016 at 9:01 am
      • cbratina

        Sounds like their braking performance is very similar? I have plenty of room for my 35c tires with the Paul Racer, could fit a 40c.

        April 4, 2016 at 9:04 am
        • Jan Heine, Editor, Bicycle Quarterly

          We haven’t done a back-to-back comparison, so it’s hard to say whether the Pauls offer as much stopping power as the Compass brakes… When I tested the Pauls years ago, they seemed fine for stopping power.
          The Paul’s oversized pivots do limit the tire width. Once you get too wide, you cannot remove the tire without deflating it. The Compass brake was designed not just to fit a 43 mm tire with fenders, but also allow wheel removal without having to deflate the tire.

          April 4, 2016 at 9:42 am
    • TroofMan

      When you say “The Compass brake was designed …” do you really mean “The Mafac brake was designed and we just made a copy”?

      April 4, 2016 at 11:16 am
      • Jan Heine, Editor, Bicycle Quarterly

        We didn’t just make a copy. We evaluated each part of the design, did finite element analyses, tested prototypes, and then decided what to make. That we didn’t change much in the end from the original Mafac Raids speaks volumes about the quality of the original design.
        If we “just made a copy” without much analysis and testing, we might have copied an inferior design like the Dia-Compe centerpulls… or we might have messed it up by trying to modernize it in some way that affected its performance.

        April 4, 2016 at 11:22 am
      • TroofMan

        Exactly what did you change from the Mafac (asides from hardware)?

        April 4, 2016 at 3:03 pm
        • Jan Heine, Editor, Bicycle Quarterly

          We cannot go into detail – if we did, we’d give away the results of much research to our competitors. The arm shape is different, and there are a few other tweaks. The Raids also were a budget model, so they used plastic bushings, whereas ours are brass. And of course, our brakes are polished to perfection, whereas the original Mafacs weren’t. That said, if you have a set of Raids, perhaps change the bushings (we sell them separately), and you’ll have a brake that performs similarly to our Compass brakes. If you spend a few hours polishing them, they’ll look as good, too.

          April 4, 2016 at 3:30 pm
  • Ugaitz Etxebarria

    Very interesting post. I’m quite interested in cantilever brakes, they are such a divisive technology. In my experience, properly set up cantis offer truly great modulation and good absolute stopping power.
    I believe that the way you set them up has a great impact on their springyness. For instance, those brakes seem to be installed a bit wider than they should be, effectively lessening the effect of the return springs on the straddle cable.
    Maybe installing them with a lower profile (pads closer to the arm, shortening the straddle cable) would allow the return springs to exert the necessary force on the yoke to keep some tension on the straddle cable. A thin straddle cable would undoubtedly offer greater feel at the levers.
    I used to seek beefier straddle cables, like the ones from Tektro, but after reading this I’m convinced that my brakes will work better with standard ones.

    April 4, 2016 at 5:29 am
    • Jan Heine, Editor, Bicycle Quarterly

      The big issue with cantis is that they twist the fork blades when you brake hard. That changes the pad toe-in and leads to poor modulation. Centerpulls attach close to the fork crown, where blade twist isn’t an issue. That is why they offer much better consistency of the brake action.
      On cantis, trying set up the brakes to preload the springs could be accomplished simply by rotating the spring hole on the canti boss. Many bosses have three holes, so you can choose. However, the effect will be very slight, and won’t solve the problem.

      April 4, 2016 at 9:05 am
      • TroofMan

        Almost all modern cantis have adjustable spring tension though.
        Also the pictured XTR cantis are not set-up with the stock Shimano Y-link straddles which have zero of the issue with cable slack you mention so it’s really not a fair comparison.

        April 4, 2016 at 3:07 pm
        • Jan Heine, Editor, Bicycle Quarterly

          I wasn’t trying to criticize the older-style XTR brakes in the photo, but simply used a photo I had to illustrate the disadvantages of a thick straddle cable. I could have used any other currently-made canti and even some centerpulls.

          April 4, 2016 at 3:31 pm
  • Mark Petry

    that’s a very cool idea, should really reduce “stiction” and improve braking feel. Nice job on the roller straddles too, a very cool RH idea that Toei re-popped, yours look great.

    April 4, 2016 at 5:31 am
    • Jan Heine, Editor, Bicycle Quarterly

      Mark, our René Herse parts are the originals. Compass Bicycles bought the rights to all René Herse designs from Herse’s daughter Lyli, and is reintroducing his best designs.

      April 4, 2016 at 9:06 am
  • Michael

    Is that cable attachment piece welded onto the arm, or is it all one piece?
    How far down the forks do the braze ons have to be?
    These brakes will work on all 559/584/700c? Wheels?
    Thanks for the info.
    Nice looking brakes. I like the straddle cable optimization, including the rolling hanger.

    April 4, 2016 at 5:56 am
    • Jan Heine, Editor, Bicycle Quarterly

      The arms are forged – they are a single piece. Specifications for the braze-on location are available on the Compass web site. The brakes work with any wheel size. They are designed for tires up to 43 mm wide, with fenders (and sufficient clearances).

      April 4, 2016 at 9:08 am
      • Frank

        Hi Jan.
        So just to be sure … your brakes are not compatible with your 48 switchback hills and your smooth honjo fenders (which are 62 wide)?
        Thanks. Frank

        April 4, 2016 at 1:54 pm
        • Jan Heine, Editor, Bicycle Quarterly

          You could probably squeeze in the big tires and fenders, but we do not recommend it. To get sufficient fender clearance, you really need fenders with a larger radius than the Honjo 650B x 60 mm. And the brake isn’t designed for tires that wide, either. We’ll talk about how to make a great Enduro Allroad Bike in the future – there are some challenges, which is why we feel that for Allroad bikes, 650B x 42 mm is the perfect size.

          April 4, 2016 at 2:15 pm
  • ben

    The main brake cable has some leverage over the straddle cable so the straddle cable probably carries more than half the load. If set low enough it is even possible for it to carry much more load than the main cable. I still don’t think you’ll break it but it is probably worth drawing a force diagram or checking out a slackline force calculator.

    April 4, 2016 at 7:45 am
    • Jan Heine, Editor, Bicycle Quarterly

      You are right, with a low straddle cable, you’d get much greater loads. Low straddle cables really aren’t a great idea, they just are a work-around for poorly designed low-profile cantilever brakes.
      One advantage of our brakes is that they are based on an old Mafac design, which has proven itself for many decades and millions of kilometers. Before we introduce our brakes, we rode the Mafacs for years and tested them as hard as possible. So we know the straddle cables don’t break.

      April 4, 2016 at 9:10 am
    • Tony

      Looks to be about .707 of the main cable force in this case.

      April 4, 2016 at 9:45 am
    • F

      Strictly speaking it is not ‘leverage’ as a straddle cable is not a lever. Among the simple machines, I see it closer to the principles of the wedge and the truss. Can anyone give the proper word for this specific form of force-amplification?

      April 8, 2016 at 5:44 am
      • Tony

        It isn’t a force amplification. It’s simple statics and the vertical components of the tension in the two arms of the straddle cable must be equal and opposite to the vertical tension in the brake cable to satisfy the requirements of equilibrium. and because the two arms are not vertical , they also each have an equal and opposite horizontal component. The vector summation of the vertical and horizontal components in each arm is the resulting tension in the arm.

        April 8, 2016 at 5:37 pm
  • RC Johnson

    One thing to consider: it is stated here that the saddle cable “only carries half the load of the brake cable, so it only needs to be half as thick”. That statement does not consider the effect of mechanical advantage and is only true at one specific angle (just under 45º total).
    The force resulting on the end points is relative to the angle of the force applied at the midpoint. At 120º the leverage is 2:1 – meaning the force applied to each leg of the saddle cable is the same as the force pulling on the brake cable. This effect increases exponentially reaching infinity at 180º.
    This graphic illustrates the effect:
    This is good advice for typical installations:
    Angles >120º a saddle cable heavier than the brake cable is advised.
    Angles between 90º and 120º the same size cable would be appropriate.
    Lighter saddle cables would only be appropriate at very low angles.

    April 4, 2016 at 9:50 am
    • Jan Heine, Editor, Bicycle Quarterly

      There are other factors to consider – mostly friction as the main brake cable bends inside the housing. The thin straddle cables have proven themselves for millions of miles… so we know they are fine. Of all the millions of old Mafac brakes in the backs of bike shops, none had broken straddle cables.

      April 4, 2016 at 10:06 am
      • RC Johnson

        Fair enough, I have no problem with promoting lighter cables based on your knowledge and field experience. You no doubt have more historical perspective about the subject than most.
        However, I don’t believe it is responsible to promote any product on inaccurate mathematical principles. It’s a common misunderstanding that a doubling of attachment points always reduces the force acting upon them by half. This inaccurate understanding can have serious consequences. In other applications, such as attaching anchors for climbing or mountaineering, it can even produce deadly results.

        April 4, 2016 at 11:26 am
        • Jan Heine, Editor, Bicycle Quarterly

          Agreed, and I changed the blog text to say “roughly half”. And yes, adding material doesn’t always make the design stronger!

          April 4, 2016 at 11:29 am
    • Oreste Drapaca

      With all due respect – to my architect’s eye, the results of the diagram are counter-intuitive: if you need 574% strength at each end of a clothes-line, it is a miracle there are any [or suspension rope bridges]. I think a cable without loops [ie, pulleys] are just vectors for the force applied, without mechanical leverage, which occurs only at solid levers or pulleys.
      Another question: what about the force exerted by the roller on the saddle cable? At the point of contact, it is 100%, so the breaking strength of the saddle cable has to withstand that force, not less, which is at the 2 ends.
      Again, my comment is strictly intuitive [ie, un-informed], so it would be fascinating to read a structural engineer’s analysis.

      April 5, 2016 at 6:02 pm
      • RC Johnson

        I know it seems counter-intuitive to many, that’s why it’s important people understand that it’s not. If you really look closely at suspension bridges they sag with a considerable angle down or the cables run way up high if it’s a long enough span to require towers. If the strength wasn’t a variable factor why waste all the extra (very expensive) cable?
        Another application, picture hangers. Anchored at two endpoints on the frame with the just the weight of the picture and frame suspended at the middle. So why do the hanger kits come with wire instead of a simple piece of string?
        Climbers and mountaineers are taught to never, ever, attach two fall protection anchors together with a sling at an angle greater than 120º. That’s because it actually increases the chance of an anchor failure in the event of a fall.
        Ever been trying to free a vehicle stuck in the mud using a come-a-long? What do you do when you can’t ratchet it with any more force? Go to the midpoint and push sideways. You have to be a bit brave though because you can break the come-a-long.
        Anyway, I think you get the idea. Lot’s of examples of this mechanical principle all around us.

        April 6, 2016 at 2:47 pm
  • Jason Miles

    “Yet the Compass brakes use straddle cables instead of complex linkages, so they are much lighter than Shimano’s racing brakes, and they have less friction.”
    The Shimano front brake BR-9010 is listed at 138g. The compass brakes are listed at 152g. Does this weight include the straddle cable and hanger? Add in the weight of the braze-ons and I am sure the Compass brakes are heavier.

    April 4, 2016 at 9:59 am
    • Jan Heine, Editor, Bicycle Quarterly

      I was comparing to a standard Shimano Dura-Ace brake, which now is listed at 150 g. The BR-9010 brake you mention is a special aerodynamic brake that requires special mounts and isn’t known for its stopping power.
      Also consider that the Compass brake includes the very thick pads. They’ll last twice as long as modern pads, but they weigh more. With thin modern pads, the Compass brake weighs about 120-130 g. However, you are right that this weight is without braze-ons, since their weight depends on how much the framebuilder files them. (We do include the straddle cable hanger.) So complete, the weight is in the same ballpark.
      The Dura-Ace brake is heavier than necessary because it uses the linkage. The Compass brake uses a straddle cable, which allows us to make the brake big enough to wrap around a 43 mm tire and fender, while keeping the weight the same.

      April 4, 2016 at 10:07 am
      • Jason Miles

        Thanks for the info. I specified the weight of the BR-9010 brakes because of your previous sentence. “With optimized straddle cables, our Compass brakes work effectively the same way as Shimano’s latest racing sidepulls – the two pivots are next to the tire, and the lower arms are short to offer great braking power.”
        You are correct that this brake requires a special mount, with two threaded holes next to the tire. Although the compass brake also requires a special mount so I think the comparison is fair.
        I am looking to replace my roller cam brakes and the compass brakes are a serious contender. The smooth post style pads on the compass brakes are the only part that gives me hesitation. Though these pads offer more adjustment I find them harder to setup and harder to adjust trail side.

        April 4, 2016 at 10:53 am
        • Jan Heine, Editor, Bicycle Quarterly

          The smooth posts have the advantage that adjusting for pad wear is easy – you just slide the posts further in. The brake geometry doesn’t change.
          There is another reason for the smooth posts – it allows the arms to be thin and strong. Most Dia-Compe centerpulls us bolt-on pad holders, but that requires a bend in the brake arms, which adds weight and reduces stiffness.

          April 4, 2016 at 11:14 am
  • Matt

    You give an excellent explanationof the advantages of centerpull brakes vs cantilever. However, as most of us ride the bikes we have I would love to see some Compass cantilever brakes. Ones that incorporate all of your research, design skill, and aesthetic.

    April 4, 2016 at 10:42 am
    • Conrad

      I hear you. I would love a bike with good centerpulls but that is still custom territory. I know some would disagree, but in my experience, V brakes with good salmon pads and the Cane Creek road V brake levers are really powerful and have good modulation. The problem is that they get in the way of most front rack setups. I have tried a lot of different cantilever brakes, wide and narrow profile, and the biggest problem I have is fork shudder in the front if you set them up to have adequate power. The only reliable way to get around it is to use a fork crown mounted cable stop, but once again, it gets in the way of your front rack. I am curious to see what the optimal brake setup winds up being with the allroad enduro bikes/bikes with 50mm+ tires. I am all for discs with suspension but I feel like discs compromise a well designed rigid fork.

      April 4, 2016 at 1:37 pm
      • Jan Heine, Editor, Bicycle Quarterly

        The fork judder on cantilever brakes seems to be mostly an issue when the steerer tube is too flexible in relation to the fork blades. (The steerer flexes under braking, rather than the fork blades.)
        With more flexible fork blades, you rarely get judder, even with a stem-mounted cable hanger.

        April 4, 2016 at 1:51 pm
      • 47hasbegun

        It’s possible to get a fork crown hanger mount to work with a front rack if you get such a rack with a ‘diving board’ mount that goes under the fork crown rather than one that goes into the front of the fork crown. If your frame doesn’t have a threaded hole up into the crown from below, you could use an eyebolt, instead, or possibly bend the ‘diving board’ behind the fork crown and tighten the nut against it, or do both together to prevent rotation and flexing.

        April 6, 2016 at 10:33 am
        • Jan Heine, Editor, Bicycle Quarterly

          It would be easier to incorporate the cable housing stop into the rack. However, I’ve never ridden a bike with centerpulls that had fork judder, so it’s not really necessary.

          April 6, 2016 at 1:33 pm
  • Tobin Henderson

    How is the pad wear on these? Do they wear roughly parallel, or at an angle? My cantis are fine, but I’m tired of always having to adjust the pads as they wear.

    April 4, 2016 at 11:15 am
    • Jan Heine, Editor, Bicycle Quarterly

      You have to adjust the pads on all brakes as they wear, since they don’t touch the rims in the same place any longer. The thicker pads on Compass brakes last a lot longer than modern thin pads, so at least you just adjust them, rather than having to replace them.
      Adjusting for pad wear is easy – simply loosen the nut that holds the pad holder and slide the holder further inward. Make sure the pad hits the rim in the right place, and retighten the nut. Done. There is no fiddly toe-in adjustment, since it’s not needed with good centerpull brakes. (If your posts aren’t brazed on perfectly straight, it’s better to file the washers under the pads or bend the arms slightly, because those adjustments remain even if you change the pads, rather than having to set the toe-in again every time you adjust the pads.)

      April 4, 2016 at 11:19 am
      • Tobin Henderson

        Yes, but I mean if you look at a worn pad from a centerpull brake, is the worn face going to be roughly parallel to the back? On my cantilever brakes, because they move in an arc, they tend to wear at an angle – like, if I were to set them up flush with the rim, the upper edge of the pad will wear rapidly. On my caliper brakes, I only ever have to adjust for cable travel as they wear, because they’re wearing closer to parallel. Of course all (almost – there are some fun examples of solutions to this problem!) brakes have to move in some kind of an arc, but are these centerpull brakes better or worse than cantilever brakes in this regard?

        April 4, 2016 at 11:41 am
        • Jan Heine, Editor, Bicycle Quarterly

          Centerpull brake pads tend to wear parallel. However, the pivots are closer to the pads than they are on true sidepulls (with a single pivot at the top), so the pad angle changes more as the pads wear. So you cannot simply use a barrel adjuster to adjust for pad wear. The plus of the pivot location are shorter lower arms, less flex and more brake power – so it’s worth the small disadvantage of having to adjust the pads as they wear.

          April 4, 2016 at 12:58 pm
    • RC Johnson

      The only pads that wear truly parallel are on brakes using a linkage.
      The Sheldon Brown site describes it:
      Parallel-Push Linkage
      Shimano’s XTR and XT V-Brakes feature a special parallelogram linkage. This serves two purposes:
      It causes the brake shoes to remain at the same angle to the rim throughout the stroke, and throughout the service life of the pad.
      It causes the direction of motion of the brake shoes to be close to horizontal, rather than the usual slanted arc centered on the pivot boss. This is a major advantage for those who use very fat tires on narrow rims, because it prevents the shoe from rising up and damaging the sidewall of the tire on release, and also prevents having the brake shoes dive under the rim as they wear down.
      See also: US Patent US5636716 A (Shimano, Inc. 1996)

      April 4, 2016 at 2:00 pm
  • Stephen Bamford

    The author is clearly an expert in the subject. I will have to re-read the article since I, like another reader saw the MAFAC centerpull brake analogy, especially when I double-checked my ‘vintage’ bicycle which has those MAFAC brakes on it. I’m sure the author’s Compass variety feature newer developments, not only in terms of alloys, but also in terms of the pulley type device situated over the right and left sides of the levers. Thanks for sharing on this and other subjects. It is nice to see older technologies that still have merit being credited.

    April 4, 2016 at 1:27 pm
  • Gunther

    Just used these very brakes for the first time for two weeks in the Ligurian alps where there are essentially no flat roads. No complaints about the absolute braking power, although at first sight a good MTB hydraulic disc brake in comparison seemed to be superior in that respect. But I found them to be terribly squealing. I would be really happy if there were a cure.

    April 4, 2016 at 3:24 pm
    • Jan Heine, Editor, Bicycle Quarterly

      Usually, the first rainy ride beds the pads and cures the squeal. The cause are pivots that aren’t perfectly aligned. A permanent cure (which doesn’t require re-bedding the pads every time you replace them) is to increase the toe-in. Either file the washers under the pad holders, or simply bend the arms a little. The advantage of these methods over spherical washers and other adjustments is that you have to do the adjustment only once. After that, you can change the pads, but the toe-in remains correct.

      April 4, 2016 at 3:58 pm
  • Mitch

    The area of a circle varies with the square of its radius, so a cable intended to carry half the load of another cable should not be half as thick. A cable that is half as thick will have one quarter the area, so half the load would result in twice the stress. That said, there’s more to selecting a cable than load carrying capacity.

    April 4, 2016 at 6:09 pm
    • Jan Heine, Editor, Bicycle Quarterly

      Cables don’t break from overloading, but from repeated bending that leads to fraying. The most stressed parts of a brake cable are hidden inside the brake lever and the cable housing. If it frays, the brake will be a bit harder to activate, but since it’s a slow deterioration, you may never notice it, until it fails… So you need a considerable factor of safety.
      The straddle cable is in the open. If you allow the ends to swivel, then the cable isn’t stressed in bending, since it hardly moves. So it’s unlikely to fray, and if it does, it’s easy to see.

      April 4, 2016 at 6:20 pm
  • Philip Williamson

    Thanks for this. Is there a simple way to make up one’s own cantilever straddle cables from shifter cable leftovers?
    On the brakes page, you might consider putting “(per pair)” next to the price, or move the last paragraph in the description to the top. I literally looked at the page and went, “that’s almost $1000 for brakes!”

    April 4, 2016 at 7:13 pm
  • Michael

    In the description it states that one end of the included straddle cable is open so the builder can get the length the way they want it and then braze on the end piece. In this case, so as to get around a seat post light.
    Leads me to two more questions:
    1. Does brake performance vary with wherever the height of the straddle cable meets the roller?
    2. What if I am mounting the Rinkos myself but know nothing about brazing on the end piece to the straddle cable? Id prefer this version but don’t know of any framebuilders in my area. Would a welding sculpter be able to do this for me?
    3. Can I use any other brand of straddle cable for the regular version? Will a brake-cable-width straddle cable fit into the roller’s groove?

    April 5, 2016 at 6:21 am
    • Jan Heine, Editor, Bicycle Quarterly

      1. On centerpulls, the brake performance isn’t affected by straddle cable height, because the two factors of effective brake arm length and straddle cable angle cancel out. That is the same as on wide-profile cantilever brakes.
      2. Brazing the cable end onto the straddle cable is easy. All you need is a small brazing torch. Even a jeweler can do that for you.
      3. You can use any derailleur cable as the straddle cable. The roller groove also fits brake cables. So you have many options.

      April 5, 2016 at 8:09 am
  • Roger

    Jan, I appreciate that each set of mounts must be well aligned by the builder to have a quiet, non squealing, brake set up, wouldn’t be better to include a “filed” washer for each pad so that some adjustment of toe in could be made?
    The saying “pads usually bed in after the first rain ride” isn’t exactly true for every bicycle as I’ve seen on a friend’s MAP bicycle. He’s tried different pads with with some success but the modern style pads are really limited as they do wear quickly. I think it’s somewhat inappropriate to have to “bend” the arms or file the pads on a brake that costs so much per wheel. That said when the time comes for a custom Rando bike it’ll be specked with Compass brakes.

    April 5, 2016 at 8:37 am
    • Jan Heine, Editor, Bicycle Quarterly

      First, on all the bikes I’ve set up with Compass and Mafac centerpulls, I’ve never adjusted the toe-in, yet they all have been silent. Whether it’s my Alex Singer with its Mafac Racers, the camping bike with its 2000s, the René Herse with Raids or the Mule with Compass brakes. Most of my friends run Compass brakes, and their brakes don’t squeal, either.
      So it’s not a big deal, but if you do need to adjust the toe-in, your builder should be able to do it when they make the bike. The Compass brakes aren’t simple retrofits, but intended for the builders of custom bikes. (They require braze-ons that few frames today have.) A framebuilder should have no trouble setting the brakes up correctly. Bending metal is their job – the fork blades we sell also are straight, and the builder has to bend them.
      We looked at all the mechanisms for adjusting toe-in and didn’t find one we liked. We decided that it was better to have the builder set up the brakes once, rather than the rider having to mess with toe-in adjustment every time they touch the brake pads. It’s not like the adjustment for toe-in ever changes on your bike. It’s all about how the brake is aligned, which is determined by the frame. So once the frame is built, the toe-in doesn’t change. Why should the rider adjust it time and again, only to save the builder a little time once?
      That said, on most bikes, the brakes are perfectly quiet after the afore-mentioned first rainy ride – assuming you brake hard a few times. If they squeal even after prolonged use, you probably have a brake pad/rim incompatibility. Some pad materials and some rim materials just don’t seem to get along. The easiest solution is trying different pads. (Different rims are a bit harder to install…)

      April 6, 2016 at 8:06 am
      • nic

        I’ve found empirically that cleanliness can make all the difference with brake squeal. Try rubbing alcohol for your rims (and pads too…), you’ll be amazed how much “black stuff” can come out of them. When perfectly clean, your braking is silent and powerful ( til your next rainy ride…).

        April 9, 2016 at 8:32 am
  • Peter Meilstrup

    Because I ride large frames, braking flexes the steertube, lengthening the distance from hanger to brake, which pulls harder on the straddle, a positive feedback that produces brake shudder. Fork crown cable hangers alleviate this somewhat but can’t be installed under most front racks. V-brakes don’t suffer this problem.

    April 5, 2016 at 11:00 am
    • Jan Heine, Editor, Bicycle Quarterly

      Your frames probably should have stiffer steerer tubes and more flexible fork blades… Flexing the steerer tube is bad for many reasons, not just because it can cause brake judder with cantilevers.

      April 6, 2016 at 8:07 am
      • Jason Miles

        I am interested in better understanding of why flexible blades could have a reduction in judder. It seems counter intuitive to reduce stiffness to reduce judder, but dynamics were never my strong suit. Do you think it is accurate to think of the flexible blades as a suspension to the positive feedback?
        I think a sidepull brake similar the the recently released campy TT brake could be interesting.

        April 6, 2016 at 12:34 pm
        • Jan Heine, Editor, Bicycle Quarterly

          When you brake, the fork will flex as the bike continues to push forward, and the front wheel is slowed. Where that flex occurs is important.
          If the flex is mostly below the brake attachment, then it won’t cause any judder. Judder occurs when the part between the brake and the cable hanger flex. It’s like a bow, with the fork being the bow, the brake cable the string. The flex effectively shortens and lengthens the bow part, which pulsates the brake.
          So if your fork blades are stiff, but your steerer tube is flexible, you’ll get all your flex where it causes judder. Cantis are worse, because the distance between the brake and the fork crown (where that flex occurs) is greater, providing a longer lever (and thus more flex). Stiffening the steerer and fork crown makes the “bow” less flexible, thus it will deflect less under a given force. And since it’s a dynamic (pulsating) load, if you provide a spring in the lower fork blades, they will take care of the forces without them ever reaching the fork crown.
          This matches my experience. I have ridden dozens of bikes with centerpulls, and not one has had brake judder. Classic bikes with flexible fork blades and cantilevers also don’t have brake judder. For example, my Urban Bike doesn’t have it. However, I’ve experienced judder on cyclocross bikes. My Alan (stout aluminum fork blades, probably not-very-stiff aluminum fork crown) has it, but fortunately only at low speeds. Bicycle Quarterly’s Ritchey Swiss-Cross test bike (carbon fork with very stiff blades, carbon steerer that probably wasn’t very stiff) exhibited dreadful judder at all speeds, until we switched out the brake to one that had less slop, which partially cured it. THe latter experience shows that there are other factors that can compound the problem.

          April 6, 2016 at 1:45 pm
  • Tran

    Jan, I own a Ritchey Swiss-Cross and wholeheartedly concur that it shudders badly. The only time that the brakes don’t shudder is when it rains – leading me to dread descending Pyrenees passes on this bike in dry conditions… What do you mean by brake “slop” and which brakes did you swap in ?

    April 7, 2016 at 2:02 am
  • marmotte27

    Anybody know of a brake cable hanger that attaches at the bottom of the steerertube, right above the bottom head set cup?

    April 7, 2016 at 6:26 am
    • Jason Miles

      I have never seen what you are describing but this patent is close:

      April 7, 2016 at 4:30 pm
    • Jason Miles

      I don’t think you would want to have the hangar fixed to your head tube because you would brake every time you turned your HBs.

      April 7, 2016 at 4:32 pm
    • Michael April 7, 2016 at 9:56 pm
    • Gary Jacobson

      I may have something that would work. Campy made a hanger (Euclid perhaps?)designed to cover the top headset cups. Could be used on bottom I suppose.

      April 7, 2016 at 10:12 pm
      • marmotte27

        Thanks, I’ll look into these.

        April 9, 2016 at 8:13 am
  • Chuck

    Are your thinner straddle cables available for use with my existing cantilever brakes?

    April 7, 2016 at 8:54 am
    • Jan Heine, Editor, Bicycle Quarterly

      We do sell the straddle cables separately, but they are basically the same as a derailleur cable. That makes it easy to replace them, if that ever should become necessary. Proprietary straddle cable can be difficult to find years down the road.

      April 7, 2016 at 4:09 pm
  • Tom

    I’m curious how wide the Compass calipers open when the straddle cable is unhooked. For example, with a 42mm wide tire, would it be necessary to deflate the tire to remove the wheel from the frame or fork?

    April 7, 2016 at 12:23 pm
    • Jan Heine, Editor, Bicycle Quarterly

      There is ample clearance for a 42 mm tire to go through the brakes, without needing to deflate the tire. That was one of the requirements when we designed the brakes.

      April 7, 2016 at 4:10 pm
  • Gunther

    Are there more photos of the blue bike with the Compass brakes? This seems to be a very nice color.

    April 8, 2016 at 1:51 pm
  • thebvo

    Could it be possible to make “yuge” centerpull brakes for an all road enduro?

    April 8, 2016 at 2:58 pm
    • Jan Heine, Editor, Bicycle Quarterly

      It’s totally possible, but they’d be large and heavy. And on gravel, you cannot brake all that hard, so the fork blade twist you experience with cantis may be less of an issue. Enduro Allroad Bikes (road bikes with tires wider than 45 mm) are still very new. With more experience, we’ll know better what works and what doesn’t.

      April 8, 2016 at 5:04 pm
  • rollingthroughnw

    For what it’s worth, the Avid Shorty Ultimate brakes use a derailleur cable as a straddle cable.

    April 10, 2016 at 4:55 pm

Comments are closed.

Are you on our list?

Every week, we bring you stories of great rides, new products, and fascinating tech. Sign up and enjoy the ride!

* indicates required