About bicycle brakes

Analysis of our brake failure, various information found about this topic and upgrade to motorcycle brake

Until recently I haven’t been thinking that the bicycle brakes (actually car and motorcycle brakes neither) can fail because of overheating. Partially because we don’t have such long and steep hills here but mainly because current brakes are reliable and designed to not overheat under normal conditions.

In summer 2016 we pedaled up to the top of the Grossglockner High Alpine Road and after 7 hours of labor we were looking forward for the downhill. We started the descent (approx. 12% grade) and we almost didn’t stop before the first twist. I was used that we just let the bike go downhill and a little slow down before turns. We went down all the hills before like this without problems and relatively fast. We trusted that our hydraulic brakes with 200mm rotors can manage everything. This time we speeded up to 40km/h, I was braking a little to keep the speed and after 1,5 minutes I wanted to stop on a pull-off. I pressed the brake levers, we slowed down to 20km/h and then the levers went down to handlebars and the bicycle was not decelerating more. Surprise, stress, panic… I put my feet on the ground and tried to brake with the shoe sole, jumping on the asphalt. We managed to stop.

We realized that the descnet will be difficult. From that time we were breaking and keeping speed of approx 20km/h and stopped every 500-1000m to cool down the brakes… and bitching that we allowed to persuade us that two brakes with big rotor are enough. As a bonus we were walking next to the bike for the last 6km because there was almost nothing left on the rear brake pads and in case of failure we couldn’t stop the bike with just one brake!

Some facts and analysis of this braking maneuver:

  • our total weight was approx 270kg and altitude loss in this part was 70m
  • if we don’t count the cooling with air we generated so much energy that the brake calipers and rotors should theoretically warm up about 500°C (see the note 1 on the bottom). Brake fluids boil at less than 300°C
  • If we brake continuously the total brake power would be 2000W. If we count just the second part of this descent when we were actually braking and not just gaining speed it would be 3200W (for comparison – the norm 4210-4 tests the heat resistance of brakes under load of 300W for 15minutes). (see the note 2 below)

These are simplified calculations because I couldn’t find or devise anything more complex. Generally the information about brake power and heat resistance is not available. On the other hand, the brake failures are not common.

Some interesting information I found related to brakes and braking:

  • actual heat resistance test of brakes – Shimano ICE-tech brakes were tested and withstanded 1050W for 30minutes
  • brake failure on a road downhill – this guy was riding a road bike with hydraulic disc brakes and after descending 150m in 3minutes going approx 50km/h the brakes failed and stopped working, he hit the side of the road and fortunately broke just couple of ribs. The discussion with brake manufacturers is also included. Generally the topic of hydraulic disc brakes for road bikes is often discussed and this one is one of few examples of failure (the disc brakes are still not allowed on races – officially because of possible injuries after crashes in peloton)
  • discussion with brake manufacturers about brake fluids for road bikes – 3 things happen when the brakes heat up too much: 1) coefficient of friction rises with the temperature at the beginning but later with higher temperature it drops off (this is called fading) 2) when the temperature continues rising, glazing can happen – the brake pad material melts and covers the rotor which further reduces the coefficient of friction 3) finally the brake fluids heats up to the boiling point and the brake fails completely. The solution is not a brake fluid with higher boiling point but we must think about how to keep heat out of the dangerous places, direct it to safe places, and then get rid of as much of it as possible.
  • DOT or mineral oil – which is better? – DOT brake fluid absorbs water and after some time the boiling point drops because of the water contents. Mineral oil doesn’t absorb water but if some water enters the system, the boiling point reduces to 100°C.
  • brake failure because of a cracked piston – even the high-tech brakes with the best heat management can fail when the piston cracks and the brake fluid starts to leak
  • in the Bicycling Science book is calculated that on a common bicycle it’s possible to reach 0.5g deceleration on flat road without the risk of going over handlebars. In real tests the reached decelerations of 0,44g with braking with both brakes, 0.36g with just the front brake and 0.29g with the rear brake. For comparison the car decelerated with 0.73g (because of this it’s not safe to tailgate motor vehicles). Cars (and theoretically also tandem or recumbet bicycles) are limited by the road-tire friction coefficient. Common bicycle has 40% of weight on front wheel and 60% on rear (on flat terrain) but the center of gravity is relatively high (cca 110cm) which notably influences the  brake force distribution during heavy braking and also during braking downhill (much more weight is on the front wheel)
  • The ideal brake force distribution between the front and rear tire depends on the height of the center of gravity and the wheelbase. During light braking the brake forces front and rear are more or less equal but during higher deceleration is the force on front brake much higher and the force on rear brake drops (the article shows also curves for car, motorbike and different types of bicycles)
  • According to the norm EN 4210-2 the bicycle should be able to stop on 7 meters in dry conditions from initial speed 25km/h. United States federal regulations  (16 C.F.R. Section 1512.5d) requires to stop the bicycle from 24km/h in 4,57meters.
  • On some roads (like the  Zirlerbergstrasse in Austria) bicycles are prohibited to go downhill because of the steep grade (16%)
  • Fading and brake fluid boiling is critical for disc brakes but the heat generated on rim brakes after long heavy braking increases the pressure in tires and softens the bead of the tire which can cause a sudden blow-off which is a critical situation in high speed and even more dangerous when it happens on the front tire (source)
  • No brake on tandem bicycle will endure a mile of 18% descent if your speed is constrained to 15 mph. Interesting braking tip (using the rear disc as a drag brake): continously brake with the rear disc brake until you start to feel the fading, stop the bicycle using the front rim brake and let it cool down for 5 minutes (source)
  • tandem bicycle should have 3 brakes – 2 of them on the rear wheel. One is either a backup rim brake or before a drum drag brake was used for constant braking downhill.

After this experience with brake failure we decided to upgrade the brakes on our tandem:

  • front brake from a motorcycle with 200mm disc. Assuming the motorbike with rider weights 200kg and should be able to safely stop from 100km/h, the brake should have enough power and mass of material for the heat dissipation to not fail even on long descent. Also the brake pads should last longer. This adds approx. 1,5kg compared with the bicycle brake but we think it’s worth it.
  • rear break – 203mm disc + backup rim brake (controlled by the front rider – also for his better feel that he can control the speed because he is the one who senses it most)

Now I start to realize that maybe we were lucky on our honeymoon tour in Americas (2013-2014) – during the descent to lake Atitlan and the speed record in Atacama. And that all our “guarding angels” were really busy during the descent from the Alpine pass. Most cyclist don’t reach the limits of their brakes, but it’s good to think if the normal brakes are enough when touring unusually (stuff+tandem+child trailer). Your life may be at risk!


1. The potential energy Ep=m*g*h (in our case 270kg*9,81*70m=185409J) and the temperature change after energy absorbtion t=Q/(m*c) (in our case 185409/(0,6*600)=515°C). I was counting with 2x300g of material and because the rotor is  probably made of stainless steel and the rest from various materials with lower thermal capacity than aluminium I counted with 600J*kg-1*K-1. Of course it’s different in reality – ignificat amount of energy is absorbed by air drag during the speed of 40km/h and the air is also cooling the brakes. On the other hand, not the whole rotor is heated up evenly – the center attachment parts and the caliper are heated less than the working parts (outside part of the rotor and the brake pads) and also the front brake  is loaded and therefor heated up much more than the rear brake.

2. Power is the energy over time – in our case 185409J/96s=1931W and 143030J/45s=3178W (54m of altitude difference over 45seconds) and in reality we cannot divide this power evenly between the front and rear brake because they are not evenly loaded.

One Reply to “About bicycle brakes”

Leave a Reply

Your email address will not be published. Required fields are marked *