Posted: Tue Dec 21, 2010 6:02 pm
now the current direction in the face cusp disruptors. here is my logic and my conclusion:
opposite coils should have the fields face opposite directions. i.e. should be clockwise or counter-clockwise, not e.g. both north-up or both north-down. that is obvious enough.
so that makes 3 pairs of coils, and it's now a matter of how the orientation of those pairs are related.
each pair can be either clockwise or counterclockwise, which we can represent as 0 or 1. and there are three pairs, so that's three bits. 3 bits is 8 combinations. but 000 is topologically the same as 111. i.e. you're just flipping the direction of current, not moving any field lines. likewise 001=110, etc. so now that's only 4 unique combinations:
000
001
010
011
but 001 is the same as 010, you've just changed the order in which you enumerated the pairs. it's also the same as 100, which by the first rule is topologically equivalent to 011.
so now you see we have only two topologically unique orientations:
000
001
and going from one to the other is simply a matter of flipping a coil pair.
point being, just make opposite coils opposite polarity (e.g. if one is north out, the other is north down), and you're all good.
i can flip the sign of the current easily, so i'll just model one config and then the other. we don't neeed to know which is which a priori.
opposite coils should have the fields face opposite directions. i.e. should be clockwise or counter-clockwise, not e.g. both north-up or both north-down. that is obvious enough.
so that makes 3 pairs of coils, and it's now a matter of how the orientation of those pairs are related.
each pair can be either clockwise or counterclockwise, which we can represent as 0 or 1. and there are three pairs, so that's three bits. 3 bits is 8 combinations. but 000 is topologically the same as 111. i.e. you're just flipping the direction of current, not moving any field lines. likewise 001=110, etc. so now that's only 4 unique combinations:
000
001
010
011
but 001 is the same as 010, you've just changed the order in which you enumerated the pairs. it's also the same as 100, which by the first rule is topologically equivalent to 011.
so now you see we have only two topologically unique orientations:
000
001
and going from one to the other is simply a matter of flipping a coil pair.
point being, just make opposite coils opposite polarity (e.g. if one is north out, the other is north down), and you're all good.
i can flip the sign of the current easily, so i'll just model one config and then the other. we don't neeed to know which is which a priori.