# Why is retorquing sometimes necessary and other times a mistake?

Bike mechanics will implore amateurs not to retorque a bolt. One must first undo the bolt and then retorque.

Yet car wheels specifically require retorquing on a different day soon after installing the wheels.

Where is the logic? Why is retorquing a bad idea in one case (the torque can become excessive and the bolt may snap; but why?), and a requirement in the other?

One (perhaps critical) difference is that bicycle bolts almost always need to be greased before tightening, whereas car wheels need dry torquing. If this is the reason for the different treatment, can you elaborate on the reasons?

• some bolts are single use only Commented Mar 20 at 21:40
• Check here - mechanics.stackexchange.com/search?q=retorque Commented Mar 21 at 13:34
• Personal experience: Both loosening and sloppy work definitely happen. Also, if an accident results, higher potential liability to the shop doing the work. If they tell you to everyone to come back in 50 miles, but few actually do, that's an out for them. Commented Mar 21 at 13:37
• @PeteW Lol, I wasn't entirely sure whether it's a Properties of Materials question or a simple exercise in Elementary Mechanics. Now you're suggesting that the answer is purely legal. Commented Mar 21 at 14:41
• and that's not necessarily a bad thing either, as long as the check is actually done. "bad work can happen here, because this task (lift the car, wheels on/off) is given to the most junior/ inexperienced employees" is a borderline useless diagnosis, from systems reliability point of view. "needs to be checked" can lead to a fix. The twist is that people tend to skip doing the check. Commented Mar 21 at 15:03

This is not something engineers learn, it's something mechanics/millwrights learn.

Because static friction is higher than dynamic friction, bolts should be torqued once and while the bolt/nut is still moving. To re-torque a bolt/nut it should be backed off so that it is moving when the limit is reached.

Note that there are some bolts that should never be re-torqued. Head cylinder bolts for engines sometimes are designed to stretch and cannot be re-used nor re-torqued.

• Re: "This is not something engineers learn, it's something mechanics/millwrights learn." I'm not sure what your point here is, but this is besides the point. I'm seeking a physical reasoning why it should or shouldn't be done. Should I have asked at physics.SE? You're most likely on the right track distinguishing static and dynamic friction. Care to elaborate ever so briefly, while perhaps explaining the apparent contradiction: grease would ensure absence (or at least reduction) of static friction — and hence retorquing in the presence of grease ought to be harmless. Commented Mar 21 at 0:48
• @sam, just pointing out that engineers in general are not necessarily the best people to ask about using tools. A lubricated bolt has a greater difference in static and dynamic friction. Commented Mar 21 at 6:44

I don't see how re-torquing with a torque wrench could be bad.

It's probably more re-torquing with a normal wrench when no re-torquing was necessary. In which case, that makes sense because most people would probably apply enough force to move the bolt or screw every time you did.

The advice to undo the bolt before re-torquing supports this.

So let me ask you this:

Bike mechanics will implore amateurs not to retorque a bolt. One must first undo the bolt and then retorque.

How often is a torque wrench used for this? I wouldn't be surprised to hear 0% of the time.

Yet a car wheels specifically require retorquing on a different day soon after installing wheels.

How often is a torque wrench used for this? This answer should be 100%.

On a bike, you usually have just one bolt doing a job, not a bolt pattern. The general scale of things lets you go to a much higher cost fastener on a bike. And most bolted fasteners on a bike don't have tons of safety margin. With a car wheel, the fastener cost is quite low, and you have a bolt pattern to contend with. You also have a filthy environment with the potential for rust, metal burrs, flaking paint and junk on the hubs, etc. So making sure the nuts or studs have seated properly after the cyclic load has been applied for a bit is a smart idea. Also note that the wheel lugs on a car transmit wheel torque when driving and braking. On most bikes, they don't.

Agree with Phil Sweet on this one.
Also the bolts on car wheels are chamfered as they have to hold the wheel in position as well as hold it on. Those bolts are the only things holding the wheels in the right place. (Bike wheels are usually sitting in a slot.) They not only have to be torqued correctly, they also have to be torqued in the correct sequence to get the wheel centered where it ought to be.