Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Removing a screw

  1. Apr 29, 2009 #1
    OK, not the usual kind of question posed here, but not entirely off-topic either.

    Last weekend I was trying to remove a couple of socket cap screws (recessed hexagonal drive) that were installed with lok-tite (thanks to the lawyers). After multiple applications of heat from a propane torch, liberal amounts of penetrating oil, and using a hex socket drive on a torque wrench, I managed to remove 1 and strip 2 others, so I'm looking forward to an adventure of drilling, screw extractors, etc.

    During the struggle, at first I was trying to turn the screws immediately after heating. Later, after heating again, I waited until everything had cooled to try again. Now I'm wondering, ignoring the effects of the loktite, is heating more likely to make a screw looser or tighter, or neither? Clearly, the screw must expand. What happens to the hole? Does it get bigger as the drilled part expands, or does the material expand into the hole, making it smaller?
  2. jcsd
  3. Apr 29, 2009 #2
    If you heat quickly, heat the hole part and the screw will loosen. Be ready - you'll have a fairly short time before the screw comes up to temperature. I'm surprised lok-tite, even the permanent stuff, worked that well. Were these stainless screws? Or, small diameters?
  4. Apr 29, 2009 #3
    No, they don't seem to be stainless. They're 5/16, 18tpi, 3/4 in. flathead screws, marked with letters 'Y' 'D' on the head. And the loctite was supposed to be the non-permanent type, but who knows what they used in the factory.

    So, ignoring the screws for the moment, you're saying that the hole gets bigger as the material expands? That's a bit surprising. Clearly, if I draw a circle on a solid piece of material, the circle will grow as the material is heated. That makes sense because the material inside the circle must expand. But it's not so obvious when the circle is actually a hole. Why doesn't the surrounding material expand inward towards the axis of the hole?
  5. Apr 29, 2009 #4


    User Avatar
    Gold Member

    Take a 6 foot length of chain, lie it on the ground in a circle, joining the ends.
    Now try to add one link to the loop. Will the loop increase or decrease in radius? Obviously, it will increase.

    Next, instead of adding one link, replace each individual link with links that are 10% larger. Once you've replaced them all, will the loop have a larger or smaller radius?

    The links, obviously, are your atoms of metal around the hole.

    Finally, add another chain around the first (which will be slightly larger in radius than the first and use a couple more links). Add as many as you want in concentric circles. Now do your swapping-in of larger links again.

    See what happens? Every chain, in order to take up more room, will have to expand outward, not inward.
  6. Apr 30, 2009 #5
    OK, thanks, that's a helpful analogy. Applying simple geometry to it convinces me that even though the individual links expand in all directions, the distances between their centers increases sufficiently so that the inside diameter of the circle expands by the same ratio.

    Then apparently, except for any effects relating to the speed of transmission of heat through the parts, assuming all parts are heated approximately equally and that the coefficients of expansion are approximately equal, the screws will be equally tight whether hot or cold, correct?

    Unfortunately, TVP45's suggestion doesn't help me. These screws are fastening a blade onto a rotor (it's a wood chipper), passing through the blade into threaded holes on the rotor, and I have access only to one side. I can't get to the back of the rotor without taking the whole machine apart, and heat applied to the blade is likely to be transmitted to the screws faster than to the rotor.
  7. Apr 30, 2009 #6


    User Avatar
    Gold Member

    You may have done better with an impact tool.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook