I recently had an experience where we were running high level arc flash calculations assuming a change in the size or impedance of a 138kV/13.8kV transformer. By decreasing the impedance or increasing the size, I would expect to see increases in incident energy on the 13.8kV system immediately downstream. We saw increases in incident downstream to lower voltage levels (4160V, 480V, etc.). Can anyone explain this to me? I would not think that this increases the variables that make up the calculation of incident energy (time to clear, current, distance away from arc)