My singlespeed bike emits a creaking noise(coming from somewhere in the drivetrain) while I'm riding on it, however it doesn't when pedaling on the repair stand.

This question is not about this problem but it made me wonder how do mechanics test to find the problem and verify that they fixed it?

Do they check every pieces of the drivetrain to find a potential problem needing to be fixed? Not knowing whether it is this problem causing the noise.

Do they go out and ride it to try to reproduce the noise?

Do they have some way to apply power on the transmission while on the repair stand? (With a bike trainer or similar)

  • 2
    Mechanics are very heavy-handed. Nov 15 '17 at 18:30
  • 1
    I remember a bike for a 350lb guy that was having brake issues. The normal mechanic test ride was not adequate. The mechanic working the bike ended up test riding it with someone on his back (piggy back, if you will). The brakes were aligned properly until you loaded the bike with 275lbs+ of weight. Some adjustments are just plain difficult. Nov 16 '17 at 13:02

With most noise-diagnosing, my usual protocol is to test ride thoroughly, including on an uphill, as the first step, to try and get the most accurate impression possible of what the customer is experiencing. You kind of need to do that to know, or at least have a pretty good guess, that you've taken care of the issue at the end. A caveat here is for my own safety (people ride some sketchy stuff) I may torque crank and stem bolts first, which can in theory throw off the before/after picture, but oh well.

I think this is the best practice, but it's also not altogether unreasonable on a lot of creak-fest modern road and mountain bikes (integrated headset and unthreaded BB) to just do all your disassembly/parts replacement/grease-slathering up front and then see what you get afterward. If a customer tells me there are multiple creaks on such a bike or if I can pick up on it from just handling the bike, I might go that route.

There aren't really any ways of simulating load in the stand. There are a few tricks useful for isolating noises in the headset and bottom bracket areas involving sideloading or selectively loading certain areas with the bike on the ground, but they don't simulate full riding loads so they're not a substitute for test riding.

  • 1
    Concur - you can stop pedalling while test riding, change gear, and see if the noise corresponds to pedalling cadence, or to wheel rotations/road speed, and whether it changes frequency with gear changes.
    – Criggie
    Nov 15 '17 at 19:34
  • You missed the most important bits that I lack - experience and wisdom to know which approach to apply first :)
    – mattnz
    Nov 15 '17 at 22:18

Not a mechanic answer, but I once diagnosed a creaky quill stem/steerer interface by riding on a set of rollers while someone else had a good listen.

In hindsight and after-testing, the cause was evident by pulling/twisting the bars while stationery, but only in a certain way.

Not sure how common this is amongst mechanics, riding on rollers is not an immediate easy skill and rollers aren't overly common.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.