Torque—the rotational equivalent of linear force—enhances the integrity of mechanical assemblies when applied correctly. Torque values are integral to engineering design as they apply to fasteners at joints which can be a structure’s weak points. Knowing the intended torque makes us smarter, safer mechancis.
Veteran mechanics have calibrated their hands from experience but even they can be off. In the PBMA booth at a recent trade show we challenged passersby to try and match a torque number by feel alone. The most experienced hands were better but many regularly over tightened compared to spec. This makes sense, as practical wrenching seeks maximum stability (against uncertainty), short of damage.
In theory we want the least torque that delivers 100% security. Yet without meters, we are safer modestly over-tightening rather than trying to stay near a minimum. Hence, torque wrenches!
There’s more to bicycle torque than for other vehicles. There are:
- Primary structure torques for fixed joints (hub-frame, crankset-BB, saddle-post). It’s best to follow manufacturer’s recommended torques.
- Primary structure torques for clamps (seatpost-frame, bar-stem, stem-fork). These structural clamping situations are rare in other vehicles.
- Secondary joints that are not structural (accessories, cable routing)—some clamped and some fixed.
For cars and planes there are few, if any, of category 2. All their primary structure joints are screwed, riveted, or welded. Clamping friction is not enough in their world. Cars have clamps for mufflers and hoses but neither are primary structure joints. Managing vital primary joints with clamping is unique to bicycles.
Structural clamping must be 100% secure but extreme force (beyond actual riding) can still dislodge. We’re familiar with crashed bikes and their turned saddles and crooked handlebars. This is normal.
Physically Test the Joint
For clamping torques, less is better and we depend on assembly paste and a physical test of the joint. As many clamp interfaces are between components of vastly different character (forged aluminum stem to carbon handlebar), less torque is safer (provided the joint is tested).
Assembly paste and published torques are new participants in bicycle work. Many structural clamping joints require a fraction (1/3 – 1/2) of the torque when using assembly paste—a boon for composite and thin wall metal components. Such practice is good and here to stay.
Other screws on the bike are from the last category, accessories. These include caps, covers, fenders, bells, bags, lights, and cable routing. Their attachments are important but not primary structure. Mechanics need to use experienced feel to avoid damage or loosening.
Recently, I received a comment about a Wera torque wrench:
Too bad their torque wrench ranges from 2.5 – 25Nm. I own a Syntace torque wrench from 1 – 20Nm and often use for example 1.8Nm for tightening brake levers or shifter clamps on my carbon handlebar.
I disagree! Personal style is bound to vary but for pro mechanics torques below 2.5Nm are best navigated with assembly paste, feel, and testing. A torque wrench in these situations is, IMHO, out of place (nothing against Syntace tools or that particular range).
Know your torques and, BTW, question the accuracy of your tools. That topic—calibration—is upcoming!