I have a 44KV/4.16KV/15MVA/ delta/wye-secondary neutral 10amp resistor ground, main transfo with station arrestors ( l.a.), 39kv-mcov on primary, and intermediates 5.1kv-mcov on secondary. The current impulse used is a standard 8/20 microsec x10KA . BIL of primary transfo is 250kv - I calculated the discharge voltage at primary l.a. = 117kv ; the ground grid volt rise = 5kv; so subtotal surge voltage at primary of tranxfo = 122kv; I then added the voltage rise due to 50ft of overhead conductor between l.a. and transfo prim bushing = dv/dt = (2x122kv/1.2microsec)x L , L= travel time = (50ft/1000ft/microsec ) = .05microsec; therefore total surge voltage at transfo prim = 122kv + 10kv = 132kv !

Now, bil of transfo sec = 75kv and bil of secondary swgr ( 4.16kv) = 60kv ; the sec swgr feeds 4kv/1000hp motors 70ft away, which have a withstand of = 18kv ( from Manuf). The discharge voltage of the sec l.a. = 16.2kv ( for 10KA ) WE DON'T KNOW HOW MUCH ATTENUATION THE TRANSFORMER OFFERS AND SO WILL ASSUME NONE FOR NOW ! Using similar calcs, I arrive at a total voltage surge , at the motor terminals = 24.2kv . Obviously, the motors are not protected. I now add arrestors and capacitors at the motor disconnect, next to motor and am able to reduce the surge voltage to = 20kv . ----- The question is how much attenuation does a typical power transfo provide ( I know it does, but I'm guessing 50 to 75%, i.e. from 10KA prim to 2.5 to 5KA on sec ) This would certainly reduce the actual stress on the motors but it would be nice to know for sure ? This is an SOS for anyone's help/comments etc.