In general, will screws tightened at room temperature become tighter or looser at very low temperature (5 K)?

On the one hand, the screw should loosen because it will contract radially, while the surrounding material will contract slightly less (or even expand, in the limit of a very large screw radius and a very thin ring of surrounding material).

On the other hand, the screw should tighten because the screw should contract linearly along its axis, making the threads tighter, while the surrounding material should contract slightly less because it has other boundary conditions.

Is there a general answer, or does it depend on material, screw size, and/or the exact geometry?

  • $\begingroup$ At least part of your guesses are incorrect: consider the classic case of a sheet of material with a round hole punched out. When the sheet warms, does the hole get bigger or smaller? (answer found at your choice of reputable physics tutorial web pages) $\endgroup$ Commented Jun 10, 2016 at 12:45
  • $\begingroup$ @CarlWitthoft Sometimes it is necessary to think beyond the classic case. $\endgroup$ Commented Jun 10, 2016 at 21:10
  • $\begingroup$ You definitely can't answer this definitively without knowing the materials of the two components. The coefficient of thermal expansion is a materially-dependent property. A tungsten screw in a ceramic plate probably wouldn't exhibit a noticeable change in tightness, for example. $\endgroup$
    – sintax
    Commented Jun 14, 2016 at 17:04
  • $\begingroup$ Assuming the same material, of course. $\endgroup$ Commented Jun 14, 2016 at 20:12

1 Answer 1


The bolt tension could increase, decrease, or stay the same.

It depends on the thermal stress distribution in the whole structure, which is likely to be eventually supported by something that remains at (approximately) room temperature. This is essentially the same situation as any bolted joint that will carry a load. The "pre-stress" put into the bolts by tightening them when the structure is assembled must be sufficient to accommodate the "working stress" in the joint, whatever causes it.

Of course there are ways to join structures that allow for thermal expansion without creating thermal stresses. They would involve sliding joints, and/or connecting links that can pivot to accommodate the expansion/contraction, etc. A simple example is the tail-pipe of a car, which hangs "freely" from suspension points on links that can pivot as it expands, rather than being rigidly bolted to the car at both ends.

If the bolts are a different material from the parts they are connecting, that is an additional complication. What happens depends on the relative sizes of the coefficients of thermal expansion for the materials.

  • $\begingroup$ This is very clear. So, what about a single screw in a single piece of material at uniform temperature, isolated from the environment (perhaps hanging by a thread)? $\endgroup$ Commented Jun 10, 2016 at 21:12

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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