blowin FET in pure sine wave inverter after testing the charging with three phase gen

hi house
I've been designing inverter since 2012 and then it was on modified sine wave. I had an experience in 2014 on site after i designed a 7.5kva/96v inverter and tested that its working fine on getting to site the inverter FET got blown i fixed it the same thing happen again but i observed the effect occurs after testing the charging with a three phase generator i only used the red phase and the neutral to test the charging.
now just last week i'm having the same issue reoccuring now on pure sine wave h-bridge arrangement of FET and 5kva/48v when i finished the installation there was power supply and everything went well and performed as expected but i got a call that the inverter blew up that they on generator the night before and when they tried to use the inverter the next day it blew of without even working i took another inverter to the site this time 3kva/24v and the same happened while i was there when i used the three phase generator to test the charging.
my issue is that what might be causing this effect and how can i avoid such in the inverter.

you have supplied no information about your fets or the ckt of the inverters or generator/charger, are we supposed to be magicians who can see your hardware form 1000's of miles away...?

I'm assuming that you are using a standby generator.

All of the three phase standby generators will only regulate if the load is reasonably balanced among its three phases. Loading them single phase not only may cause excessive vibration, but I don't think it will regulate its voltage properly.