Welding Machine on DG Set

I don't know if I'll be able to answer your question directly, but I can give you a couple of tips on welding near brushless alternators. Working on the theory that nothing is a perfect insulator you may have produced circulating currents in the alternator, or injected a spike. You also didn't mention if the alternator was on load and failed while welding.

Anyway, when welding on any machine the earth/ground connection of the welder should be as close to the weld as possible, even if the machine has earth bonding use an extra earth cable at least equal to the cross-section of the welding cable. As I mentioned nothing is a perfect insulator, therefore you may have introduced a spike through your stator onto the rotor, then to the diode outputs of the exciter. Feeding back in from the stator to the rotor can be likened to transformer action of an unknown quantity.

Next, I wouldn't weld on a rotating machine anyway as you have to be aware of micro welding of the bearings, and that's another subject, but certainly not on an on- load alternator. I would however disconnect the rotating diodes, earth the stator connections, and to be sure - earth the exciter diode connections. You may also want to earth or disconnect your AVR cables, you don't want a back fed spike going there either. If you do earth your exciter there's 99.9% chance of losing your residual magnetism, so when you start the machine again you will need to flash the DC field.

As for your second problem, it seems that your load differential relays are doing their job, try and resolve your load imbalance problem. Start off on no-load and build up the load, watch the current on all three phases as the current builds up.

Hope this has been useful.

Thanks but I feel you misunderstood or I have not given it correctly. I have not done any welding on DG or the alternator. It is regular maintenance jobs done for pipe welding done when there was power failure and DG set was running.

Regarding second problem, I must confess, I could not understand exactly, what you have suggested. Please forgive, as I am not an Electrical Engineer.

Hi Uday

Your second question may be the answer to your first problem. In your generator protection system you will have current transformers on your bus-bars that supply information to your load sharing and monitoring system, part of that system will be that you can't generate power on a vastly unbalanced load. So either your welding machine is taking the one phase beyond it's trpping point, or you really do have an earth/no-earth problem. Without knowing exactly what your configuration is it's difficult to be precise.

Try to replicate the same situation and see if you get the same result, it could have been a coincidence the DG stutdown when you struck a welding arc, then there could be a different problem to look at.

Diodes would fail in the event of unbalanced load on three phases of generator. If you have used a welding machine with single phase 'or' even two phase, that will create the problem. As you indicated that 4 out of 6 diodes were faulty, probably there was overloading of two phases.

To me it sounds like whomever rated the breakers on the generator didnt do it properly.

May i ask where did you purchase this from?

<...DG set is having capacity of 484 Kw but it is tripping after 275 kw...load is not distributed equally on all the phases. Can it be a cause of tripping the DG at lower value than rated?...>

Of course it is! The power output of the DG in this particular case is determined by the current limit in the most heavily-loaded phase. Once the current limiting device has operated, that's it.