>What is the mechanism that keeper provides? On the face of it, why should >an increase in reluctance lead to the domains re-arranging? It's one of >those things that I know happens, but don't understand why! Peter, A nice man at a magnetics course told me why, but I didn't take much notice because he was talking about Alnico magnets and I was interested in Samarium Cobalt, which isn't nearly so badly affected by removing the keeper. Do you know about B-H curves? They are the diagrams showing applied and "generated" magnetic field, and always show hysteresis between the two. When the magnet is in place with the keepers, you are at the point on the B-H curve where the curve crosses the y axis somewhere above the origin. When you remove the keeper the effect is to push you down the B-H curve. When you replace the keeper you would come back up the B-H curve BUT there because of the hysteresis so you don't end up with as much magnetism as you had before. As to what happens to the domains, I'm not really sure. It's one of those things that you think you understand....until you have to explain it to someone! Regards, JohnP MODEL ENGINEERING DISCUSSION LIST. To UNSUBSCRIBE from this list, send a blank email to, modeleng-request@xxxxxxxxxxxxx with the word "unsubscribe" in the subject line.