Since I wasn't clear on some of this, I found this most interesting. The following clear description is taken from "How to Dunk A Doughnut: The Science of Everyday Life," by Dr. Len Fisher. Chapter 3, "The Tao of Tools," is full of interesting information. (Dr. Fisher, originally from Australia, is a physicist with a fine sense of humor.)
Undoing a bolt is a different story. It takes more force to undo even a clean, well-oiled bolt than it does to do one up, because the initial force to overcome friction and get the mating surfaces sliding (head and nut surfaces against the work, and two threads against each other) is greater than that required to keep them sliding, so wrenches will need to be held a little further out. When it comes to undoing a rusted bolt, the problem is not that the two threads are "stuck together" by the rust. The real problem is that, as iron turns to rust (a complex reaction product of iron, oxygen, and water) it expands, generating enormous pressures that increase the frictional forces between the threads. One can see how high the pressure can be by looking at a stone into which iron spikes have been driven. As the iron rusts, it is not uncommon for the pressures generated to be so high that the stone is split. Oil is of little use in reducing the frictional forces caused by rust. A better trick (if you have the time) is to use a weak acid such as vinegar to gradually penetrate and dissolve the rust. Alternatively, if the joint is accessible to heat, application of a propane torch will expand the bolt, the nut, and the gap in between to relieve some of the pressure.
Tightening bolts:
Stuart Burgess has made some interesting further points, which I quote here with permission:
1. A neat way to tighten a bolt is to heat it up first, and then do it up to hand tightness. When it cools down, it shortens, and becomes pretensioned.
2. Many screws are made of soft metals, and are easily damaged by hardened screwdrivers, and still more by hardened misusers of screwdrivers.
3. Spring washers are good at indicating the correct preload in screws and bolts.
Undoing a bolt is a different story. It takes more force to undo even a clean, well-oiled bolt than it does to do one up, because the initial force to overcome friction and get the mating surfaces sliding (head and nut surfaces against the work, and two threads against each other) is greater than that required to keep them sliding, so wrenches will need to be held a little further out. When it comes to undoing a rusted bolt, the problem is not that the two threads are "stuck together" by the rust. The real problem is that, as iron turns to rust (a complex reaction product of iron, oxygen, and water) it expands, generating enormous pressures that increase the frictional forces between the threads. One can see how high the pressure can be by looking at a stone into which iron spikes have been driven. As the iron rusts, it is not uncommon for the pressures generated to be so high that the stone is split. Oil is of little use in reducing the frictional forces caused by rust. A better trick (if you have the time) is to use a weak acid such as vinegar to gradually penetrate and dissolve the rust. Alternatively, if the joint is accessible to heat, application of a propane torch will expand the bolt, the nut, and the gap in between to relieve some of the pressure.
Tightening bolts:
Stuart Burgess has made some interesting further points, which I quote here with permission:
1. A neat way to tighten a bolt is to heat it up first, and then do it up to hand tightness. When it cools down, it shortens, and becomes pretensioned.
2. Many screws are made of soft metals, and are easily damaged by hardened screwdrivers, and still more by hardened misusers of screwdrivers.
3. Spring washers are good at indicating the correct preload in screws and bolts.
