On 12/14/05, Morgan Olsson wrote: > > I think this could be made a easier. > > 1) Why the heatsink at all? During theese microsecond pulses there is no chance at all that the heat from die propagate even to the surface of th eSCR package. Heatsink is only relevant if diong discharges short after each other in time. And even then not very hot; take average current (easiest to measute as capacitor bank charge current) times SCR conducting voltage (max 2V?) Not many watts. The capacitor bank voltage is 900V. The peak wattage is over 22MW. During the "instantaneous" transition between non-conducting and conducting, a _huge_ amount of heat is generated. The heat is generated inside the package, and takes awhile to get to the exterior, but it's important to have heatsinks so that the heat is carried away as quickly as possible after the event. > 2) Why one expensive SCR? Much easier to get parts if desinged using many parallel circuits of {discharge cap, SCR, coil winding}. So that the coil is made up of twenty intertwined conductors, each connected to in one end to an own SCR; in the eother to an own capacitor. Just make sure everythng is parallel/symmetrical and fire all triac firmly at the same time by another firing SCR. Getting them to fire simultaneously is the tricky part. The _easier_ method is to use one hefty SCR. The cheaper method _may_ be using several smaller SCRs. If the hefty SCR is free or cheap, then the cheaper and easier method is using the hefty SCR. -Adam -- http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist