>The AVR simply isn't fast enough to do that, grant. Can the AVR >intelligibly respond to clock transitions every 1 microsecond? Nope, >not unless it was 50MHz. The only way would be to slow the driver >down by a LOT. And since you posted the source, tahts possible. :) So mel, when is the EIII you posted in October going to be done? I'm realy looking forward to the Lead Acid bearing device... :)