what makes rectifier diodes in genset keeps burning is improper grounding the cause or other else please help.

Question

our cat 1.4 Mw keep busting its diodes even at 750 kw only of reactive loads. It is in parallel with another set of Cat 1.5 Mw. with the same loads. It has the same engine 3512 and alternator. almost all parts identical even the AVRs and load sharing modules. this is my concern in our power plant and it affected greatly our production (mining company) beside that costly replacements of diodes. Any help will be apreciated. So sad that even CAT Phils can't find the solution to our problem. in fact this is a new engine but since it arrive it gives us a lot of problem. We thought of not anymore purchasing CAT for our new genset requirements for our new Sulphide Plants. Can any CAT people in USA or somebody with the same problem help us.

Answer 1

there are only 3 main reasons to lose diodes, heat, over load, over voltage. is there some kind of large load that drops off line suddenly? if there is and the governer on the genset is slow to respond you may be seeing a high voltage problem that could fry diodes. you might have heatsink issues like not enough silcone heat transfer grease under the diodes, poor cooling air flow, or some kind of heat transfer into the diode such as hot sun or misdirected hot air from the engine or other source. vibration or shock is something else to look at. some reactive loads have problems with harmonics or spikes too. sorry i can't be more specific but maybe i have given you some ideas

Answer 2

i have to ask . is the phasing correct on all 3 legs of new gen set? if one pair is swapped among the 6 available [parallel gen sets] the power of said leg will go to ground during the phase swap overloading the weakest diode. also are any of the transmission lines running through a flooded area of your mine [ possible shorting to ground on the return leg after load devices] i hope if not this, that someone helps

Answer 3

are the same diodes always fried, or they randomly fail? a bad ground is not likely to cause diode probs. more likely a partial short to ground, or high resistance in a circut. does one diode carry higher amperage than the others?