How to Brake on a Bicycle

lift_rear

Whether you ride fast or slow, being able to stop quickly is an important skill. Your ability to stop also depends on how well your brakes perform. In Bicycle Quarterly Vol. 10, No. 1, we tested brake performance to determine:

  • How do you brake most effectively?
  • Do brakes with different stopping power affect braking distances?
  • Do brake shoes and pads make a big difference?

brake_test

For this test, we coasted from a set point down a steep hill (always using the same riding position), and then braked at a predetermined point. Two testers performed the tests, to differentiate the “rider” factor in brake performance.

Why test on a downhill? The stopping distance on a hill is longer, and thus the differences between brakes and techniques are more pronounced. Afterward, we performed a statistical analysis to evaluate whether we were measuring real differences in braking performance and not just seeing noise in the data.

This test confirmed quite a few things about effective braking technique:

Front vs. Rear Brake
On dry pavement, the front brake alone halts the bike over the shortest distance.

Many riders think they need both brakes to stop effectively, if only because most bikes are outfitted with 2 brakes and that implies that one should use both. Here’s the way to think about it: the momentum of your body continues to move forward as your bike is slowing down, so your weight shifts forward. That’s why your rear wheel can come off the ground when braking hard. When your weight comes forward during hard braking, your rear wheel has close to zero traction. If you apply the rear brake under these conditions, the rear wheel will lock up without contributing significantly to the braking effort.

If you can apply the rear brake without locking up the rear wheel, then your weight isn’t shifting forward – a clear sign that you aren’t braking as hard as you should!

We tried braking with both brakes and with the front brake alone, and consistently found that if we focused all our attention on the front brake, we achieved much shorter stopping distances.

rear_brake

When we braked with the rear brake only, the stopping distance was more than three times as long. In fact, Hahn overshot the stop sign and went into the road at the bottom of the hill (above). This was despite Hahn modulating his rear brake carefully to keep it below the lockup point as much as possible. Skidding the wheel would have increased the stopping distance further.

While we couldn’t test this on dry pavement, the rear brake comes in handy when it is so slippery that even moderate braking will lock up your wheels – when you encounter ice, wet leaves, loose gravel, or other very slippery pavement during the first rain after a long dry spell.

Under these conditions, you cannot brake hard, and the forward weight transfer is much less pronounced. In that case, the rear brake provides added friction that will slow you more quickly. Rear brakes also are useful on tandems, where much more weight is on the rear wheel.

brake_line

How Hard Can You Brake?
Very, very hard. We found that to get the shortest stopping distance, we had to pull the front brake lever with all our might. Witness the tester’s bulging muscles on his right arm – which controlled the front brake on this bike!

This is despite using very powerful short-reach Dual-Pivot rim brakes. Some hydraulic disc brakes require less force at the lever, but with rim brakes, you really need to pull very hard on the lever.

When we came to a stop, the smell of burnt brake pads wafted through the air. After 21 full-on emergency braking maneuvers, the Aheadset of the test bike had developed play, because the stem had slipped on the fork steerer. Even the quick release of the front hub had loosened. Braking hard is a very violent affair. Not once did our front wheel lock up.

lift_rear

Going Over the Bars
Many riders are afraid of “going over the handlebars” when braking hard with the front brake. They fear that the braking force will cause the bike to rotate around the front wheel. In practice, we found that even on a steep hill, the rear wheel stayed planted for most of our braking. Wind resistance helps you here: It pushes the rider backward.

Once we had slowed to less than 6 mph, the rear wheel tended to rise. In the photo above, you can see how the front wheel has stopped, while the (unbraked) rear still is spinning. The wheel came up very slowly. This was far from dangerous: The rider simply opened the front brake slightly to make it come back down.

The shortest braking distances were obtained when we slightly decreased our braking power just before we came to a stop, so the rear wheel stayed on the ground.

Since few riders ever brake this hard, how come they still go over the bars? Here is what appears to happen to most riders who go over the bars: If riders don’t brace themselves against the handlebars, their momentum will push them forward over the handlebars as the bike slows. (Imagine being a passenger in a car without a seatbelt as the driver brakes hard.)

To avoid this, Hahn in the photo above braces himself against the handlebars and locks his elbows. He has shifted his weight as far back as possible. You can see his bicycle’s saddle underneath his belly. With this technique, he did not “go over the bars.” And if your bike’s rear wheel does lift, it happens slowly enough that you can counter it by slightly releasing the front brake lever.

Conclusion
We found that we could brake much harder than we thought.

Car companies have found the same thing: Drivers tend to be too timid when braking in emergencies. Many modern cars are equipped with “brake assists” that apply the brakes with full force during emergency stops. (ABS makes sure the wheels don’t lock up.)

On bicycles, the “brake assist” and “ABS” are quite literally in our hands. Fortunately, instead of having to manage four brakes during an emergency stop, we can focus on the front brake alone.

Without electronics to assist us, we can benefit from practicing braking hard. (In fact, the same applies to your car, where practicing braking will make you a safer driver.) Practice on an empty road, preferably on a downhill where you can reach higher speeds. It takes a lot of confidence to pull on the brake levers with all your might, but it can make the difference between stopping safely and running into something in an emergency.

For the other parts of this research (differences between the two tested brakes and between brake pads), check out Bicycle Quarterly Vol. 10, No. 1.

Further reading:

About Jan Heine, Editor, Bicycle Quarterly

I love cycling and bicycles, especially those that take us off the beaten path. I edit Bicycle Quarterly magazine, and occasionally write for other publications. One of our companies, Bicycle Quarterly Press publishes cycling books, while Compass Bicycles Ltd. makes and distributes high-quality bicycle components for real-world riders.
This entry was posted in Bicycle Quarterly Back Issues, Cycling Safety, Testing and Tech. Bookmark the permalink.

84 Responses to How to Brake on a Bicycle

  1. Patrick Moore says:

    Interesting subject. Thank God, I’ve had to panic brake on very few occasions, but, riding fixed so much, I long ago experienced the uselessness of rear brakes in all-out braking, this paradoxically on a bike with just a front brake. This was when I found my fixed rear wheel skidding during hard braking even though I did not deliberately exert backward pressure on the pedals. The natural instinct to stop pedaling when braking suddenly sufficed alone to skid the rear wheel, which was by that time barely touching the pavement, as the “feel” of the skid indicated.

  2. David Pearce says:

    How interesting. I also like that I am slowly reading your previous blog posts. Your alternative STP route entry from 2010 was so beautiful, and your Carbon Fiber Handlebar 911 from Oct. 5, 2010 was neat and probably meshes very well with this post on braking. I think just as you highlight previous BQ articles, you might like to highlight in a little sidebar, greatest hits from your blog for new users.

  3. David Pearce says:

    Further on your Carbon Fiber 911 of Oct. 5, 2010, I note with interest that you repaired CARBON with METAL and WOOD. I also love the fact that you mention that this should only be a temporary repair! :-) I’m really wary of carbon for this reason, it seems to break with shock. No way I’d trust my fork to it, I’ll tell you.

  4. Andrew Joseph says:

    When I was teaching mtb skills, the first thing I did was show riders how to manage their brakes, getting them to do a ‘stoppie’ really increased their confidence quickly. It also helped them realise how much (or little) they needed to pull on the brakes.

  5. Bubba says:

    Do you have any qualitative or quantitative statements to make about how much worse-off you are if you brake from the tops of the hoods? From experience I’m sure it’s worse, and I imagine it could be a LOT worse. Will your stopping distance be twice as long if you brake from on top of the hoods? three times as long? More? Maybe I will practice both for myself on the hill outside.

    • Much of this depends on the brake levers. I do change my grip to brake from the drops. Not only is the leverage better, but you also can better brace yourself against flying forward. The split-second lost while I change positions is more than made up when I subsequently can brake much harder.

      • tamaso says:

        I would second that, in my experience, braking performance from the hood tops depends a great deal on the hood shape, lever shape, and pivot point. For example, the current (11spd) Campagnolo ergo shape brakes markedly better from the tops than the previous (9/10spd) shape.

  6. charliewhite says:

    Good article, it proves what many of us already knew. Even though I know that I should use the front brake in emergency stopping, I have never been able to keep my hand off the rear brake in a panic stop. I have practiced and practiced, but when a car suddenly pulls out I tend to brake hard with both hands. I also have a hard time getting my butt behind the seat in a panic stop.
    When I have a predictable braking point in front of me, it is easy to remember to move my weight back and brake with the front. I just haven’t mastered it in the panic situation.

    • Maybe practice can help. I used to brake with both brakes like most cyclists do. Now, I don’t touch my rear brake more than a few times every year. As a side effect, my rims last much longer, since the front one has much less grit that wears the rim, and the gritty rear one rarely gets abraded by the brake pads.

      • Edwin Williamson says:

        “Now, I don’t touch my rear brake more than a few times every year. ” After I read the original article in BQ I have noticed how little I use my brakes around town in general, and especially my rear one.
        Great post.

    • Fred Lee says:

      Agreed. Emphasizing the front brake is something I need to practice as well.

      Though I don’t think it’s reasonable to expect someone to drop their belly onto the seat in an emergency. Not to mention, seldom do real emergencies involve just braking, and that position puts you in a bad spot for maneuvering after you’ve shed speed.

      Jan’s point elsewhere about braking in the drops will also serve to lower one’s center of gravity, which helps to accomplish the same thing without sacrificing subsequent maneuverability.

  7. roberthurst says:

    I’ve also done quite a bit of experimentation with max stops. We always found that our shortest stops came when using both brakes. Obviously the vast majority of stopping power comes from the front brake, but the rear brake did seem to provide resistance which lowered stopping distances noticeably over front-brake-only stops from 20 mph. Probably because the rear imparts some stopping power before skidding and during the skid, especially if you can push the wheel slightly to the side as it slides. As you note, in shortest stops the rear comes off the ground toward the end of the stop. Before it comes off the ground, it obviously does some work slowing the bike.

    The wording of your post is interesting. You say when you “focused all your attention” on the front brake you achieved much shorter stopping distances. Of course! But I think you’ll find, as we did, that if you don’t apply the rear brake at all the stopping distance will increase noticeably.

    Robert Hurst

    • It is possible that your front brake didn’t have enough power, and thus the rear brake could contribute to the stopping. With a good brake, there is no doubt – the rear wheel barely touches the road and cannot contribute braking power.

      If you skidded the rear wheel, then you certainly didn’t get any brake power on the rear. Skidding reduces your friction close to zero – that is why all cars now have anti-lock brake systems, because few drivers can be trusted to brake hard without skidding. The optimum traction occurs just before you skid.

      Also, when you skid, the rear wheel no longer tracks, and you tend to slide out of control.

      Did you publish your results? If so, where?

      • ted kelly says:

        ” Skidding reduces your friction close to zero – that is why all cars now have anti-lock brake systems, because few drivers can be trusted to brake hard without skidding. The optimum traction occurs just before you skid.”

        Skidding does not eliminate friction (though completely unloading a wheel does). A car with all four wheels locked up stops in a reasonable distance (I learned this a long time ago, for corroboration a quick google search turns up http://deepblue.lib.umich.edu/bitstream/handle/2027.42/49006/ptv6i1p11.pdf;jsessionid=E83507A12A9691ADF652894723228B6D?sequence=2). But you can’t steer a car with the wheels locked up. Cars have anti-lock brakes because of the control issue, not because a skidding tire has slightly less friction than one thats about to lock up.

      • You are right. I meant to say that the combination of the weight transfer during hard braking and a skidding rear wheel reduces friction on that rear wheel close to zero.

      • davep says:

        JH (below): “You are right. I meant to say that the combination of the weight transfer during hard braking and a skidding rear wheel reduces friction on that rear wheel close to zero.”

        The skidding is a consequence of the transfer of weight off of the rear wheel.

        In braking, something is “skidding”: either the brake pad on the rim or the wheel on the pavement.

        If the wheel (rear) is off of the pavement, then it’s contributing zero to stopping. If it’s still contacting the pavement, it’s contributing something to stopping.

      • Erik says:

        Just a minor correction: the maximal friction is depend of the rubber compound. When ABS was introduced in cars at first stopping distances increased (but the fact that steering remained possible was more important to avoid collision then pure distance), it was only a few years later that tires gave shorter distances with ABS mainly to the evolution of the used compounds. In some circumstances (like on gravel or snow) ABS still increases the stopping distance because there is no material piling up in front of the lockedup wheel.

        (and according to the Belgian textbook on driving, bicycles have a shorter stopping distance than cars due to the very low weight.)

    • I think you’ll find, as we did, that if you don’t apply the rear brake at all the stopping distance will increase noticeably.

      The opposite was the case. When we used both brakes, we got longer stopping distances, simply because you had to manage the lockup of the rear brake and couldn’t focus all your attention on the front brake. Our shortest stopping distances were when braking with the front only, consistently over two riders and many test runs with different brakes. We did a statistical analysis to make sure the differences were real and not just noise in the data.

      • roberthurst says:

        There’s the problem, I guess: thinking at all about “managing the lockup of the rear brake” — I assume that means feathering the brake to keep the wheel from skidding, your apparent assumption being that the rear wheel skid will lengthen the stop. But you have already mentioned why this is a faulty assumption. In a max stop, with properly applied front brake, some rear tire skid is inevitable. You can “manage” the skid by controlling the rear of the bike, but attempting not to skid at all will surely lengthen the stop (in which case, yes, you’d be better off not using the rear brake at all). Here’s the main issue I think: A skidding rear tire ADDS friction compared to a freely rolling tire with no brake applied. The rear brake also slows the bike a bit before the skid starts, so the front brake doesn’t have to work as hard (the bike is going a little slower by the time the front brake is doing all the work). I think these are the reasons that we achieved noticeably shorter stops using both brakes rather than with front brake only.

        I guarantee you that a skidding and/or bouncing rear tire slows the bike more than your freely rolling, completely unbraked rear tire. If it turns a little to the side while skidding, so much the better, learn to deal with that, the stop will be even shorter. In a max stop, the rear wheel eventually comes off the ground anyway, so if you can’t control a skid you certainly won’t be able to control the entire rear end of the bike waving in the air.

        My experiments suggested a range of around .85-.9 g for maximum controllable deceleration on an upright bicycle, based on dozens of runs with several accomplished bike handlers using calibrated speedometers on a slight downhill slope. This is a much shorter stop than cyclists had been led to believe was possible in the past, the old calculations based on faulty assumptions. The results of these experiments were published, more or less, in my book called Art of Urban Cycling, 3rd edition to be published next year under the title Art of Cycling.

        Robert Hurst

      • If you just lock up the rear tire, you might as well not use the rear brake at all. The amount of brake power you get from that is totally negligible. As you mention (and as we found), the rear tire eventually leaves the ground anyhow, showing you how lightly it touches before, when wind resistance still pushes the rider backward. Any added friction certainly would get lost in the noise of the different runs when you do the statistical analysis. If you get statistically significant results that show more brake power with the rear brake, then I suspect something is wrong with your experimental setup or your bike’s brakes.

        If you braked from a very, very high speed (say 100 mph), then it’s conceivable that the rear brake should be applied first, since wind resistance results in a higher loading of the rear tire compared to the front. This was pointed out by Tony Foale in his book on Motorcycle Handling.

        Send us a copy of your book for review when it appears. Bicycle Quarterly’s readers will be interested in learning about it.

      • roberthurst says:

        Jan Heine wrote: “If you just lock up the rear tire, you might as well not use the rear brake at all. The amount of brake power you get from that is totally negligible.”

        This is an assumption that is disproved by physical experiments that you apparently did not perform.

        Using both brakes while using the front brake near its maximum, such that the front tire is as close as possible to skidding without skidding, requires an exaggerated, well-timed body movement and results, inevitably, in a rear tire skid, then a rear tire lift off. But, first of all, the rear tire skid is not instantaneous, and secondly when it occurs it does add some friction to the system, especially if it goes a little askew and the rider can control it. (Hockey players and skiers don’t have brakes at all, but manage to stop quickly anyway — cyclists can do a bit of that too.) Rather than providing negligible braking power the braked rear tire provides enough stopping power for long enough that it results in noticeably shorter stops, again and again in physical experiments, compared to allowing the rear tire to roll unbraked.

        If you are cruising down a hill and slam on the rear brake but not the front, what happens? You will skid, seemingly not slowing at all for a moment, but then you will slow and stop. The hill would have to be durn steep and/or slippery to overcome a locked rear tire. The tire tread would be destroyed by the skid and soon the air would explode violently out of it due to massive heat and wear. The friction of a skidding rear tire is not “totally negligible.” I think we can put that curious notion to rest.

        I initially got interested in this particular subject after reading the claim that the maximum deceleration of an upright bike/rider was .6-.7 g’s. That was widely accepted for decades but seemed obviously incorrect to me and rudimentary experiments proved it was way off. In subsequent, more carefully controlled experiments several riders (admittedly very nifty bike handlers) using carefully calibrated speedometers consistently achieved stops in the neighborhood of .85-.9 g’s, which is right around 15 feet stopping distance from 20 mph (30 feet per second), using both brakes. This is better than many performance automobiles. The experiments obviously have the potential to be ruined by slop but we used a slight downhill slope, and lots of repetition, to help account for that. Nobody has ever shown that a bicycle can stop shorter than .89 g — while the rider stays in control of the machine.

        This information was published in The Art of Urban Cycling in 2004, and The Art of Cycling in 2007, in the chapter called “Panic Stops.”

      • Hockey players and skiers don’t have brakes at all, but manage to stop quickly anyway — cyclists can do a bit of that too.

        Their skates and skis dig into the surface. If you are talking about off-road cycling, then I agree. For the same reason, cars with anti-lock brakes have longer stopping distances on snow, where locking up your wheels provides the shortest stopping distances. On pavement, your tires don’t dig into the surface in a significant way.

        especially if it goes a little askew and the rider can control it

        Whether the rear tire skids straight or sideways makes no difference in the friction on the road. It’s not turning, after all, so you just have an oval contact patch sliding over the road surface.

        If you are cruising down a hill and slam on the rear brake but not the front, what happens? You will skid, seemingly not slowing at all for a moment, but then you will slow and stop.

        Actually, we performed that experiment, and the rider did not stop. He barely slowed, and only stopped after he had reached the flat part at the bottom of the hill. The stopping distance was more than 3.5 times as long as with the front brake. This is without the weight transfer that you get when you use the front brake. All this goes to show that at least with our setup, the front brake was powerful enough that the rear brake could not contribute in a meaningful way.

        .85-.9 g’s, which is right around 15 feet stopping distance from 20 mph (30 feet per second), using both brakes. This is better than many performance automobiles.

        I googled a bit and found that the Porsche 911 GT3 RS 4.0 recently set a record, braking from over 200 mph to a stop in 6.5 seconds. This equates to 1.4 g. Car and Driver states that standard sedans can brake at about 1 g. Of course, once the owners put cheap Chinese tires on their cars after the first set wears out, braking distances are bound to increase significantly.

        However, it’s safe to assume that a car has a shorter stopping distance, on the same surface, than even a good bike with a skilled rider. Cars simply have more rubber on the road, and they are long enough that the rear won’t lift off the ground. Therefore, cyclists should be careful when following cars.

        I look forward to seeing your data. If it shows a statistically significant improvement from having a skidding rear tire, then we will try to replicate that.

        As a last thought, skidding your tire will ruin it, if you run a supple high-performance tire. This would make your emergency stop a very expensive one.

  8. Champs says:

    This confirms what we all knew, and what Campy used to reinforce by only offering rear brakes with a single-pivot design.

    You can also put this together as the mnemonic that rear brakes are commas, and the front brake a period. If you mean to carry on. Don’t use a period. Nor should you stop on a comma,

    • I was disappointed when Campy went away from their logical ways and now uses dual-pivot brakes front and rear. The lighter single-pivots had plenty of power to lock up the rear wheel… but I guess customers didn’t like the idea of a less-powerful rear brake.

      • Larry T. says:

        I’m the opposite – finding Campagnolo’s weaker, single pivot “skeleton” brake to be harder to modulate, resulting in MORE lockup in the rear. I used this setup for a couple of seasons in Italy where the descents on our Legendary Climbs tours are my favorite parts, but switched back to the more powerful, dual pivot, which I find easier to modulate. And before someone posts “you don’t know how to brake!” I’ll point out that back-in-the-day when I was a pro Superbike roadracer, the rules required a working rear brake, which I NEVER used. We purposely left enough air in the hydraulic lines that during tech inspection it would stop the wheel turned with the inspector’s hand while pushing the brake pedal, but did essentially nothing otherwise. Anyone who watches MOTOGP knows the rear brake is mostly useless since you can often see daylight under the rear tire as they brake at the end of the high-speed straightaways, the reason we used the tiniest rotor and caliper we could find back there. But strangely, during high-speed descending on a bicycle (not panic stops) I find overall better slowing for the turns using BOTH brakes, though of course the rear lever is squeezed with a bit less force. Max braking force is achieved at the point just before lock-up, making modulation key. Good points here on hand positions, on the drops is better. I run my brakes a bit slack so my small hands can both grip the bar and leave the index finger to pull on the end of the lever while descending – “one finger braking” is plenty if your finger is down where it can develop the max leverage on the lever. Some fear the lever contacting the bar and are scared of running their brakes with any more than a hair between the pads and rims but I feel there is more than enough travel there, as once you lock things up, you’re in big trouble, so why have this happen with all that space between the lever and bar?

      • during high-speed descending on a bicycle (not panic stops) I find overall better slowing for the turns using BOTH brakes

        It seems that different riders use different techniques when descending. I tend to brake hard, at the last moment, and then open the brake progressively as I turn in. Braking deep into the turn means there is more weight on the front wheel, which improves traction. Of course, in that situation, braking on the rear would be very dangerous, as you have little weight and traction on the rear, and locking up the rear wheel would make you crash immediately.

  9. So do you suggest routing the front brake to the right lever for right-handed riders, a la Sheldon Brown and the rider in the photo above? It would seem one would pull hardest with the dominant hand in a split second emergency. Also, what about setting up the rear to grab less?

    Also, a related question: I’m setting up braze-on centerpulls with aero levers and I’m so far having a bit of trouble with the front being mushy. I’m wondering if the angle of the cable from under the bar tape to the brake cable hanger is too severe. Any suggestions?

    • I usually set up my brakes with the left lever controlling the front brake. Yes, I need to pull hard on the lever, but I have enough power with my left hand.

      Regarding the mushiness of your brake, a severe angle of the brake cable cannot help. Why not route the cable over the stem and then into the hanger from the other side, so that the curve is less severe? I did that on the last bike I had with aero brake levers, and it worked great.

    • Fred Lee says:

      In my opinion, Sheldon Brown’s efforts on this front were very misguided.

      Unless one has very weak hands, one can lock up most brakes with either hand. Dominant or not makes very little difference. But the risk of having your brakes opposite of everyone else’s is that whenever you’re riding someone else’s bike, or someone else is riding yours, they will lose the majority of their stopping power when they instinctively use the wrong lever.

      • William says:

        I have all of my bikes set up with the right lever controlling the front brake. My rationale relies on the fact that I signal with my left hand (in the US.) This allows me to better modulate and control my braking going into turns and allows me to signal adequately. When I lend a bicycle to somebody, I make sure that the other person knows what is what before they need to ask. When borrowing bikes, I don’t find adapting a nuisance.

      • antimony says:

        Maybe I do have hands that weak, but I’ve never managed to lock either wheel, on any bike I’ve ever ridden. And I’ve done some panic stops, although I’ve never practiced them, which perhaps I should. I am sort of curious what the optimal braking pattern is if locking a wheel is not going to happen (and thus no thought is needed); I tend to use the rear to modulate speed because I find steering on a twisty downhill more intuitive with rear braking (it might not be optimal, but it would take a lot of convincing to try retraining my brain) and then go to the front when I really need to stop NOW. Also, I rear-brake a lot while signaling, but am always going a speed where that’s reasonable.

      • Alex Merz says:

        I set up right-front so that my bikes match my motorcycle, and so that (like William) I can have my right hand on the front brake lever while I signal with my left hand.

      • Larry T. says:

        Right hand, left hand..I find it doesn’t matter performance-wise, it’s what you’re used to. My wife still prefers “Italian style” with the right lever working the front brake, same as on motos. Bikes in OZ are still sold this way according to our Aussie clients while European bikes have switched. One would expect as a (very) old moto guy I would prefer this but I learned early on the difference and it’s never been an issue.

    • ted kelly says:

      Regarding your cable routing question, you might try using a noodle (like on V brakes). When your cable route causes a tight bend they can work better than regular housing. They are available with various angles, and can be fine tuned tighter by bending around a large socket or other strong cylinder.

  10. ianbrettcooper says:

    Speaking solely as a bike commuter who pretty much scorns speed and sport cycling, I think that anyone who has to apply brakes suddenly and forcefully is probably traveling way too fast for a situation he should have been able to anticipate and prevent long before it came to the point of having to apply brakes hard. I prefer to assess risk, improve lane position and modify speed somewhat overcautiously so that any braking I might be forced to do is gentle. I see no reason to run at potentially dangerous situations at 30mph (or even 10mph). Slowing to 5mph doesn’t make any cyclist a wussy – it just ensures a bit more safety. No appointment is so important that it can’t wait a few more seconds.

  11. Andy says:

    Jan, have you considered using two front brakes? Or, is there any system in which two brake levers can control a single front brake? It seems that if you don’t use the rear brake anyway, some system which equalizes the squeezing power from one hand to be split over two while offering the same amount of braking may be preferred.

    • Interesting thought… One could even consider putting a brake on both sides of the fork crown, and use a separate lever for each. Maybe then one could actually flip over the bars, or break the traction of the front tire. ;-)

      For now, a single (powerful) front brake does all I need it to do.

  12. Patrick Moore says:

    The old-fashioned British Racing Tricycle has two front brakes, often a caliper or centerpull paired with a cantilever or V-brake, given that adding rear brakes to a “Delta” tricycle is fraught with mechanical hassle and performance ambiguities. This is more for redundancy than stopping power. My late ’80s Ken Rogers trike had an Altenberger dual pivot paired with old, high-profile Shimano cantis and either brakeset was sufficient to stop a speeding, grocery-loaded trike in mid stride.

    I must say that, on my two fixed gears with front brakes only, I don’t miss the rear brake.

  13. Bill Gibson says:

    My finely honed hard braking skills are why I can’t ride a fixed gear in traffic. I go into a fully braced, far back and low position, which quickly ejects me from a fixed gear bicycle! Never mind, I’ll stick to my instincts: I’ve never been in the cool crowd anyway…

  14. roberthurst says:

    “As a last thought, skidding your tire will ruin it, if you run a supple high-performance tire. This would make your emergency stop a very expensive one.”

    Indeed it will. We ruined several rear tires while performing this experiment.

    Ask yourself — how could this be possible if a skidding rear tire contributes no stopping power? Obviously it does. Quite a bit less than the front but certainly not negligible.

  15. ted kelly says:

    If you will allow a small reprise to your previous post, I wonder what sort of shoes and cleats you use, and how you choose them. In the first photo of this post it looks like the rider has typical “road racing” shoes which some folks describe as hard to walk in. Am I right about that? Given your embrace of the occasional walking of the bike, do you consider walkability when selecting shoes and cleats?

    • Hahn uses road racing shoes. He says he can walk fine in them. It may be worth a try for me, too, since I really don’t like the interface of SPD pedals. Cleats and retention mechanism on the pedal wear out pretty quickly. It’s a system designed for mud clearance, which is not what I need when randonneuring. Unfortunately, “real-world” riding still is not on the map for the big component makers, and so it’s unlikely that we will get a walkable pedal system for commuters, bicycle tourists, century riders and randonneurs anytime soon, even though that population is much larger than that of road and mountain bike racers.

      • ted kelly says:

        I have found that cleat covers make walking on speedplay cleats much nicer. Probably not as big an improvement on plastic bottomed cleats but on smooth surfaces (e.g. tile) the covers eliminate that skating syndrome. Still a truly good solution without extra bits like covers would be really nice.

      • Gert says:

        I use SPD-SL pedals and the cleats are OK to walk on also on brevets to and from controls and for short distances up hill if necessary. Wet floors can though be a little tricky.
        When I used to commute to and from work. The short distance from the bike parking to the showers did reduce the lifetime of the cleats significantly. But they are a cheap part in an expensive hobby.
        By the way tried braking front only today. Worked fine, but if I do not concentrate I use the rear as well.

      • Bill Gobie says:

        I found the most walkable and durable road pedal system to be Time’s. Unfortunately the pedals were fragile. Now I use Shimano’s road system when I need to be attached to the bike, although after five years I have learned the plastic parts on Ultegra pedals are not replaceable. Maybe Dura-Ace is next for me, or 105 and consider them expendable.

        Cleat covers: What the heck do you do with them after visiting an unsanitary restroom?

  16. Gino says:

    I think that using all the available force capable by a strong rider on a road bicycle caliper brake setup could result in fork, frame or brake damages. I think that should be cautioned as-well. There is such a thing as too hard braking on a road bike configuration or any light weight bicycle configuration.

  17. Erik says:

    From multiple experiences I can assure that magura HSxx brakes are perfectly capable of locking up the front wheel. (Yet, even taking in account the added distance of my lying body, stopping distances are very short :-)).

  18. Stig V. says:

    Where I live, having two independent braking systems on your bicycle, is a legal requirement. Besides being very good common sense. Even if the front brake alone may be able to stop you just as fast as when using two brakes, it is nice to have a secondary system, should the primary fail. And I’d guess most of us have had a brake cable break.

    Also, the weak secondary system is useful for adjusting speed when riding in a tight group.

    • Also, the weak secondary system is useful for adjusting speed when riding in a tight group.

      You are right that modulation is as important as overall stopping power. A good brake is both powerful and has excellent modulation. That means you can use it both for minor speed adjustments and hard braking. In a smooth group, just letting up on the pedals is enough to adjust for minor speed variations. If you actually have to brake in your group, it’s probably not safe to ride with them…

      • Erik says:

        Sometimes true, but just to give an example: we sometimes train on an oval open-air track. (300 meter) When riding in a group the first rider catches the wind, every 180° turn changes his wind-resistance drastically. Letting up on the pedals is not enough to compensate this fast slow down.

  19. John McNamara says:

    This article and discussion couldn’t be more timely in my case. Recently, I’d twice skid/slid sideways and gone down during slight descents on wet roads. After 10 yrs of daily commuting, this was a learning experience. What I discovered: 1. there were a few dents in my rear wheel rim, one was more pronounced (there was significant loss of tread opposite of this dent) and was causing the brake to lock up the rear wheel. And 2. my rear wheel’s rim was severely worn in, more than when last I checked, while the front rim much less so. Jan’s comment about barely using the rear brake made me think that I need to give attention to what my hands are doing: most of my friends say they use both brakes, as I thought I did…until now. If I am going to the rear brake more, it may be because of a poorly performing front brake combined with an innate fear of going over the bars, both feeding more impulse to grab the rear brake more or faster.
    It may be that I’ll need to back my rear pads out 1 or 2 mm and dial in my front brakes more finely?
    Finally, I wonder if the cantilever brakes I’ve been using almost exclusively are suboptimal to other kinds of brakes?

  20. kurtsperry says:

    I think in the real world for panic braking, the real wheel brake is best forgotten. Sure in theory there is another small increment of braking force using the rear but in the real world the rear wheel unweights a fraction of a second into the braking and trying to do both at once is more likely to result in user error–and a longer stop. Sport motorcyclists have naturally argued about this for years and there are top world champions who have both used the rear brake racing and others who never or almost never used it.

    Also, if you have strong forearms and hands, it is entirely possible to reach the traction limit of the front tire braking at speed even using an old sidepull-type brake if it is properly set up and with good pads and clean, well bedded in braking surfaces on the rim. Center pulls or dual pivots offer more mechanical leverage at the caliper and are better suited to those with weaker hands and forearms, but a higher total leverage ratio (the sum of the mechanical leverages of the levers and calipers) always means having to set up the pads closer to the rims for a similar lever travel, which means more scraping when wheels come slightly out of true or pick up crud from a wet road.

    • You are right about dual pivots – their main advantage comes being able to set the pads close to the rim, as the brake returns to the same location after every braking maneuver.

      However, the advantage of centerpulls does not come from the higher mechanical advantage, but from the reduced flex. The pads can be quite far from the rim without risking to bottom out the lever.

      • kurtsperry says:

        The mechanical leverage of a centerpull brake caliper is typically considerably higher than a sidepull, especially at shorter reaches. This is almost built in to the respective designs. All one need to do to verify this is is to measure the lever arms on each side of the pivot(s). The ratio between the leverage distances on each side of the pivot or pivots will geometrically set the ratio. Short reach brakes or pads run at the top of the adjusting slot will obviously, all else being equal, increase the leverage ratios as well. Shorter reach brakes or brakes with shorter distance between the pivots and pads will also be less prone to flex and chatter but may compromise tire/fender clearances. You want the brake reaches to be as short as will accommodate the required clearances.

        This obviously ignores the leverage ratio built into the lever end which is equally important but generally varies considerably less here than the differences between sidepull and centerpull calipers, at least among the ones I’ve used.

  21. I have managed to lock up the rear wheel in typical New York moments, when the cab in front stopped on a dime and I squeezed abruptly both levers. The rear always skidded, which surprised me [on dry pavement, mind]; I counter-steered and diminished the pressure on both brakes instinctively, which seemed to bring me closer to the cab. In a study, one is prepared to brake fully, but not so in urban traffic, when one is easily distracted… This was on a Brompton with well tuned brakes, front on the dominant hand and on new tyres.
    I appreciate the importance of breaking with the front mainly, but I tend to agree with Robert Hurst, in that a skidding rear on the ground adds to friction.
    I use the rear exclusively when I turn and need a slightly lower speed in a hurry [again, in urban conditions].
    In terms of ABS, skidding & cars, things are a bit more complex: a F.1 pilot was/is able to modulate and stop shorter [ABS is banned in F.1], but for normal drivers… The accepted wisdom for normal cars is that skidding increases the stopping distance, in normal and mainly in wet conditions.

    • I think we all agree on the basics. It doesn’t really matter whether the front brake provides 97% or 99% of the stopping power during hard braking. For the average user, focusing on the front brake will obtain the shortest stopping distance. It’s of secondary importance whether or not a skidding rear tire contributes marginally to the braking effort or whether it causes a loss of control that will distract from the braking effort.

  22. Steve Palincsar says:

    The loss of control from a skidding locked rear wheel is hardly of “secondary importance” if it causes the rider to crash. Of course it shouldn’t, but all too often that’s exactly what happens.

  23. HillDancer says:

    There was an attempt to address the first bullet point, but not the other two. My experience says disc brakes whether mechanical or hydraulic, with optimal pad & rotor pairing, can decrease panic stop distance, and improve control when slowing. Swiss Stop BHP (blue) rim brake pads will shorten panic stop distance compared to time hardened pads, and improve control when slowing.

    Too bad the examination didn’t included Hetre tires on common width rims with rim brakes compared to Hetre tires mounted on wide rims with disc brakes. I suspect new negative g records could be set with the latter.

    • Based on our experience, at least the common Avid mechanical “road” disc brakes (BB7 “road” model) have much less stopping power than a good rim brake. We have tested a number of bikes with different rotors and pads, even had a local expert fiddle with one of the test bikes, but there was little difference. The brakes feel great at first, but as you pull harder on the levers, you only get flex in the brake mechanism, but no more braking power.

      Bill Davidson has a nice little demonstration. He takes a bike with those brakes, pulls the lever for the front brake to the bars, and then pushes the bike across the shop floor. The wheel continues to turn. He then takes a bike with a rim brake, and the wheel won’t turn.

      For the next issue of Bicycle Quarterly, we are testing some of the new hydraulic “road” disc brakes. Based on our experience with hydraulic mountain bike disc brakes, these should work better.

      • Patrick Moore says:

        My experience with discs, solely with the Avid BB7s, is that they are quite as powerful as any other brake I’ve used, and certainly more powerful than any centerpull I’ve used (Mafacs of various kinds, Shimano Tourneys and Dura Ace — probably rebadged Tourneys, Weinmanns, and probably others that I can’t remember). My experience with the centerpulls may be vitiated by undue flex — the cable housing hangar introduces a possible liability here — and all of mine were mounted on the bolt-on pivots, but I’ve not gotten better results than with single or dual pivot calipers, and occasionally worse, with the centerpull giving a linear incease in retardation with lever pull until, at a certain point, pulling harder didn’t do anything more. This was on a fixed gear that carried heavy loads, and I noticed the defect when bombing down steep hills under load — scary. I replaced that one — a Dura Ace, IIRC — with a Shimano dual pivot, which worked considerably better. All my rim brakes are set up with Salmon pads and I let them break in to the rims.

        IME, the problem with the BB7s is not power but modulation: they tend to be somewhat “on and off”, even more so than V brakes. Note that these are the mtb version and pulled by Tektro or Cane Creek V-brake drop bar levers. I once tried the road version and those were entirely useless despite a 180 front rotor and the best housing money could buy (this was BMX housing constructed like a very beefed up Shimano indexing housing, but designed to be non-compressible for brakes). Note also that IME pads and discs have to wear in together, a process that takes many more miles than with caliper pads and rims..

  24. Patrick Moore says:

    I should have specified at the top, and not at the bottom, that the mtb BB7s are a different animal from the road version.

  25. HillDancer says:

    I can stop my bike with road BB7s from over 40mph with one finger, with one brake, front or rear, without the lever traveling to the bar. I will be glad to help you get it right.

    • I am glad those brakes work for you. Our problem was not that they would not stop the bike, but that they would not stop the bike fast enough. We measured stopping distances, and found that the BB7 Roads took significantly longer to come to a stop on our test hill (see the Summer 2013 issue of Bicycle Quarterly). Before somebody says our brakes weren’t set up correctly, we experienced this with test bikes from Civia, Volagi, Tournesol and several others. I don’t think all these makers don’t know how to set up the brakes, but that these brakes simply don’t have the power of a good rim brake.

      • Correction: We tested the stopping distance of Shimano’s new CX75 mechanical disc brake, not the Avid BB7 Road, in the last issue. On the road, the CX75 and Avid’s BB7 Road felt almost indistinguishable.

        This means that not only the mechanical “road” disc brakes from Avid, but also those from Shimano, suffer from insufficient brake power. It’s not a problem with all disc brakes: For the upcoming Autumn issue, we test two disc brakes that offered plenty of power.

  26. HillDancer says:

    FYI, Shimano has recalled CX75s manufactured after February 2012.

    There is a flaw in your evaluation, something about the mechanical disc brake dynamic you fail to recognize and the mistake is repeated. The BB7 caliper does not deserve the “insufficient brake power” generalization.

    • something about the mechanical disc brake dynamic you fail to recognize

      I pull on the lever, and I expect the bike to slow down quickly. I am not the only one who experiences this, and Bill Davidson’s demonstration (see above) is pretty convincing.

      It is possible, of course, that all these manufacturers and bike shops cannot set up the brakes correctly, but if it is that hard, then Avid and Shimano should provide better instructions.

      I suspect what we are seeing instead are different expectations of what constitutes “adequate” brake power.

      • Larry T. says:

        Back-in-the-day some cheap Honda motorcycles had cable-operated disc brakes. While they looked great on the showroom floor compared to old-timey drum brakes, they were terrible when it can to stopping. I think a disc brake must be hydraulic to really work well, whether it’s on a bicycle or moto. Based on a couple of hours riding a Shimano hydraulic disc braked MTB, I think their road brakes will work fine – but how much brake does one really need given the tiny contact patch of a bicycle tire? THAT is what really limits braking ability.

      • I agree that Shimano’s hydraulic disc brakes are very powerful. We had a test bike with them, and I had to recalibrate my hands, as I pulled on the lever too hard. I can see this being useful for mountain bikes, where the vibrations can cause severe fatigue of the arms. On a road bike, a little more effort at the brake lever can be designed in, to make the brake effort more consistent with that of rim brakes.

  27. HillDancer says:

    I hooked a rubber band around my brake lever & handle bar and wasn’t able to push the bike across the floor, I repeated the same outcome using the tip of a little finger. The anecdotal example you keep referring to is not representative of a properly installed and properly bedded mechanical disc brake, and absolutely not a BB7.

  28. Paul says:

    In the book “Bicycle Science” there is a drawing of a front brake attached to a spring on the rear brake and as soon as the rear wheel left the ground this spring would activate and disengage the front calliper. Lost my copy, perhaps you have seen it.

  29. I saw at the Denver Bike Show in Feb 2013 a disk brake with cable connection from brake lever to a small lever actuating a piston – a hybrid. Has anyone used that ?

    • We are testing one for the Autumn issue of Bicycle Quarterly.

    • Gert says:

      I read a review of one in german called “doppelmoppel”. The testers were very enthusiastic. But it is priced at 289 euros, so…
      Anyway Shimano and Sram now make hydralic levers for some of their groups. Disc brakes for roadbikes is for whatever reason the new big thing at the Eurobike show

  30. olybikes says:

    We like to recommend that any mechanical disc brake be set up with compressionless housing, which I have yet to see come stock on ANY mech disc brake bike. To me, this is a horrible mistake since such brakes often have long runs of housing, which is “lossy”. If you run mech-discs, try it. I felt a favorable difference on my MTB BB7s with drop levers designed for such cable pull.

    Aside, regarding disc brakes, proper run-in/bed-in/burn-in is important. If not done properly, it can cause a noisy brake. Also, maximum brake effectiveness relies on proper run-in. I do not know, but wonder whether improper run-in prevents _eventual_ maximal brake effectiveness. That is, does an improperly run-in disc brake prevent later correction to run-in or does “the system need to be reset” where the rotor is cleaned of all previous (incomplete) run-in material and then the run-in process started all over again?

    Jan, you mention vibration causing severe fatigue for off-road riders. I’m not discounting that, but for me, extended braking on steep technical descents are highly fatiguing (whether cable or hydraulic, rim or disc). Are you saying the that braking during high vibration is synergistically fatiguing? It is common to run levers closer to the bars so that biomechanically, riders are at the stronger portion of their fingers’ flexion. Makes a big difference for me. Maybe extra important since many lever blades are designed for one finger braking — better give those pointer-fingers all the help they can get, because they’re alone on the job.

    • roper run-in/bed-in/burn-in is important

      How do you “properly” run-in/bed-in/burn-in a disc brake. I did notice that the hydraulic brakes on a recent test bike became much better after a few long braking maneuvers, but how do you do this “properly”?

      • Patrick Moore says:

        I agree that proper bedding in is even more important for disk pads and rotors than for caliper pads and rims. I’ve been told — and it seems to work — to do a few repeated fast downhills with hard braking; in fact, someone just the other day said that carefully heating the pads with a gas jet to burn out impurities, then lightly sanding with 120 grit, works even better. But IME mere use breaks them in eventually.

        The same seems to be true for Shimano mechanical disk brakes. I’ve only setup and used these on one bike, but after a few miles they worked fine, and right off the showroom floor they worked as well as any caliper I’ve used. You certainly could not lock the brake levers and push the bike across the floor.

        Regarding cable housing: I disagree that mountain BB7s require special housing. The full-length, generic housing on my Fargo doesn’t keep the rear from being very powerful. On the other hand, with ROAD BB7s, nothing you can do, at least IME, makes them work well.

      • a few repeated fast downhills with hard braking

        Well, then we did bed them correctly – that is what we do anyhow. We let the bike roll and brake at the last moment approaching the corner. On rim brakes with salmon-colored Kool-Stop pads, riders who follow behind often smell the burnt brake pads. Surprisingly, this does not affect brake performance, and during the brake-free straights, the rims and pads have ample time to cool down. (I have stopped in mid-descent numerous times to check rim temperature, and the only time it wasn’t fine was on a Bike Friday with 20″ wheels.)

      • We had one test bike reworked by a shop priding themselves in their expertise with these brakes. They installed compressionless housing. The only way we could make the brakes work without having the levers bottom out against the bars was by setting the pads so close they rubbed the disc at all times. I guess we pull harder on those brake levers than most riders…

      • Patrick Moore says:

        To this and your earlier reply regarding bedding in by fast downhills stops: I wonder if you are talking about the Mountain BB7s? As I said earlier, the Road BB7 setup I used was useless no matter what I did, and I did a lot including installing a 185 mm rotor in front, compressionless housing all round, “organic” pads, furious attempts to bed in. But the Mountain BB7s, at least mine, are very strong, even with full-length, cheap or generic housing, and 160 mm rotors. They take, say, 50 miles or so of use to bed in, but after that they, or at least mine, are very strong — they certainly lock or jam to a stop 60 mm knobbies on pavement, front and rear, without bottoming the levers.

        They don’t modulate as well as some other brakes I’ve used, though.

        The one set of Shimano cable disks that I set up for a friend and tested briefly (~20 miles) were also quite strong even without much bedding in. Note that these were pulled by mountain-type or flat bar type levers, not the V-brake drop levers I use on my Fargo with the Mountain BB7s.

        Disks can bend and do need to be close to the pads, but not excruciatingly close. This can be a pain, but one learns not, for example, to roughly shove wheels into the trunk of one’s car — ie, to be careful with the rotors when the wheels are off the bike.

  31. HillDancer says:

    Optimal Disc brake friction is generated by microscopic bonds being formed and broken, the bond is between pad material transferred to the rotor and the pad. Each pad type (Metallic, Semi-Metallic, Organic, multi constituent compound) has its own optimal bed-in procedure, research and follow manufacturer directions. FYI, organics encompass a wide range in which resins and ceramics are included. Some pads materials are so narrowly focused they suck at anything but being quiet. As a general bed-in rule, a few slow downs from a running pace to bring the rotor to a moderate temperature, followed by 10-12 repeated firm braking events at speed (~10 mph) to a near stop, not lock-up. Allow the rotors to cool completely after initial bed-in. More than one bed-in session may be required. Hard braking down hill is not the best way to transfer material and temper a steel rotor. Often this will promote pad glazing, a condition where binders used in pad processing form a glaze preventing pad transfer and/or friction material contact. A glazed pad can be renewed by sanding with 600 grit or finer abrasive. The pad must be cleaned thoroughly after using a de-glazing abrasive. If a pad’s surface is touched with unprotected fingers, it is contaminated, contamination will delay proper bed-in. Touching the rotor will contaminate its surface, cleaning with denatured alcohol or Acetone is the best course of action. When installing a pad with a different pad material than used originally, clean the rotor well.

    Linear housing like Jagwire Ripcord or SRAM Slickwire for BB7s provide the best performance currently. In addition to the linear wire construction, these two examples have an outer sheath of woven Kevlar, which reduces mechanical lose due to housing deformation. L3 housing is not a suitable substitute. Excessive cable length and numerous cable fittings introduce flex in the system. Plastic ferrules are too flexy, avoid if possible. The POP fitting on the business end of the cable is important, do not omit this ferrule. Inline brake cable adjusters from Avid and Shimano are good, others brands are not. A cable tension adjuster is very important for the front brake.

    The road model BB7 caliper is meant to be used with short travel high mechanical advantage levers, not levers designed for V-brakes, Cantis, center pulls, drums, etc., and not levers with the new pull ratio from Shimano.

    For users living in flat lands, and disc brake use on loose unpaved surfaces, a rinky-dink 160mm rotor is adequate, a 140mm rotor is always sub-optimal for road use. The laws of physics can not be denied. Greater work requires greater leverage, and prolonged & frequent braking creates heat (see pad glazing above). Hydraulic actuation employs a boost mechanism, mechanical actuation relies on a longer lever arm for its work boost, in other words a large rotor. A large steel rotor (>160mm) has increased heat capacity as well increased thermal transfer ability, conductive transfer from the pad, and from rotor to air via convection. In addition to adequate diameter, a steel rotor must have sufficient swept surface area. A wispy rotor with lots of empty space may look cool, but it will perform poorly. The minimum steel rotor size for the front of a road bike, and a 160lb + rider should be 180mm. There is promise of a rotor & pad material on the horizon that should alter some mechanical disc brake limitations.

    Adjusting brake pad gap and caliper position is explained in written documentation supplied with new BB7 calipers, and is available in online tech docs for second hand & built bike purchasers.

    The entire issue of evaluating a brake through stopping distance alone is missing the point of how we use our brakes 99.9% of the time. A stick in the spokes can lock a wheel, so what.

    • The road model BB7 caliper is meant to be used with short travel high mechanical advantage levers, not levers designed for V-brakes, Cantis, center pulls, drums, etc.

      The BB7 Road calipers actually are poorly matched to current road brake levers. A lever that pulls more cable (like old non-aero levers or those for cantilever brakes) would make the brake a little harder to activate, but it might help with the problem of the lever bottoming out against the bars. We haven’t tried this, though, because we felt that generally, that wasn’t anything to be gained from disc brakes compared to a good rim brake. After all, a rim brake is a disc brake with a very large rotor…

Comments are closed.