Maybe not really the answer on your question but:
many cases of equipment failure may be determined by inspection of diode or thyristor silicon tablets. Adhering to manufacturer's specs for torque or pre-load pressures, surface finish of heatsinks and properly calibrated clamps could save expencive, if not catastrophic failurers in high-power equipment. Simple good houseceeping practices (periodically checks for loose wiring, dust, carbon tracking, discoloration of aux. components and pressure settings of clamps) could result in many years of fault-free operation.
Some guidelines and possible causes:
1.Severe burning and silicon damage on surface.
Misaligned heatsinks, incorrect clamping, high current surge due to external short-circuit. [/red]
2.Servere burning and silicon damage at gate area.
Insufficient inductance in circuit or high surge current (di/dt). [/red]
Insufficient gate drives, faulty or incorrect RC network, gate/aux. cathode cables reversed.[/red]
3.Burning and carbon tracking at edge of silicon.
Voltage break-over due to excessive transients, blocking voltage degraded as a result of device running too hot, faulty snubber or transient suppressor components.[/red]
4.Silicon damage with "hot-spot" on surface.
Transient or dV/dt failure, snubber or transient suppressor components faulty.[/red]
5.Device short circuit (not blocking).
Blocking Voltage breakdown. If no visible damage to silicon, the element is most likely cracked as a result of severe over-clamping, oadly misaligned heatsinks or locating pin, or mechanical shock. It is unlikely that a fracture will be seen due to the tablet being sandwitched between the discs.[/red]
6.Device short circuit with exposed silicon.
Over-clamping or possibly misaligned heatsinks.[/red]
RCC
