I have an interesting problem at hand on a diesel generator onboard a marine platform. The generator is rated at 380V, 50Hz, 3Ph and 1500kW output.


While conducting installation trials of the generator in a mixed load bank at 0.8 power factor, it is observed that when the generator is loaded at 1000kW, the generator runs stable within limits of output frequency and voltage. However, within 3 minutes, after settling on 1000kW load, the output voltage drops to 260V and the generator load decreases to ~640kW. Within 3 seconds the voltage is regained and things are back to normal. At the instant the drop in voltage and load is observed, a rise in frequency to 51.5 Hz is observed. The frequency settles to the rated frequency as the voltage settles. In steady state, this phenomenon continues at intervals of about 200 seconds. Also, interestingly, if you slightly increase the frequency (by 0.6Hz) after generator has settled at 1000 kW, this phenomenon doesn't occur. Additionally, if you tweak the output voltage slightly high at no load (0kW), the 'load shedding' occurs at a lower load and vice versa. Change of AVR and governor has had no effect on this phenomenon.

Voltage and Frequency Drop

Is my generator possessed?

  • 1
    \$\begingroup\$ What is regulating this generator? Is it a generator, or an alternator? Way too many unknowns to venture an answer. Sounds like a synchronization issue, but the controls were not mentioned. \$\endgroup\$
    – rdtsc
    Sep 11, 2020 at 18:07
  • \$\begingroup\$ Its technically an alternator. The prime mover RPM is controlled by a Woodward governor and voltage is regulated by an automatic voltage regulator (AVR). \$\endgroup\$ Sep 11, 2020 at 18:09
  • \$\begingroup\$ @IshanKarve Have you checked the lube oil pressure to the governor while this occurs? \$\endgroup\$
    – jonk
    Sep 11, 2020 at 18:38
  • \$\begingroup\$ @jonk It's an electronic governor. No lube oil in use \$\endgroup\$ Sep 12, 2020 at 3:35
  • 1
    \$\begingroup\$ It sounds like a thermal problem. Do you have boost current sensing (3 current transformers in the output lines) in the feedback loop. If the boost sense transformers get hot it can affect their saturation point. They provide boost to the AVR when you get large load variations and could potentially have this type of response. \$\endgroup\$ Sep 12, 2020 at 4:18

1 Answer 1


@jackcreasey Thanks. Your input on the current transformer was certainly insightful. The defect was identified to a winding in the compounding circuit of the alternator. One of the winding had an in-turn short and other was open circuited. This was due to degradation of the insulating material view condensation and overall compartment temperatures. Post rewinding of the assembly, the problem has since been resolved and great lessons learnt.

  • 3
    \$\begingroup\$ Good work. You can accept your own answer to indicate that the problem is solved. \$\endgroup\$
    – Transistor
    Sep 27, 2020 at 7:59

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.