In a message dated 3/8/04 7:42:52 PM, Dennis writes:
>I stuck the nozzle of a small portable torch into the shell and hit the cup
>with a good ten second blast of heat <clipped> Then I
>did likewise on the outside face of the cup, carefully and not long enough
>to do any damage to the paint on the shell. Whether the heat allowed the
>oil to further penetrate the threads, or the cup tweaked, or what I'm not
>sure, but after it cooled down, the cup screwed out with relatively little
>effort.
The torch and flaming penetrating oil is my favorite last-ditch removal method. But you'll probably get the best results if you apply the heat to just one part, not both. I believe the reason this method works so well is that heat travels faster and more efficiently by conduction than it does by convection or radiation (if my physics is in error here, I welcome any elaborations or corrections). When you heat one of two parts (especially the outside one), it absorbs energy and heats up quickly (by conduction)... when the heat reaches the threads (or whatever other spot is stuck), the heat takes longer to bridge that gap (because some of it must travel by radiation across the gap) and thus the heated part is rapidly expanding while the unheated part is still trying to absorb heat. The difference in expansion rates often "pops" the joint, even if it is rusted or seized. You get the best results when you have a highly conductive metal outside of a metal of lower conductance (like an aluminum crank arm on a steel spindle), but even if it is the other way around (a stem inside a steering tube) or you have to heat the inside part, the torch method still works great!
Bob Hovey
Columbus, GA