UPS Inverter and Mains Not Synch

[gpi2019]

We have 2 generator rated at 282kVA & 939kVA both runs automatically during power failure. But the 282kVA will be the first to supply the UPS and will be transferred to the 939kVA genset after 10-15 minutes but during the operation using the 282kVA genset an alarm appears on the UPS controller "Inverter & Mains Not Synch" and then again "Inverter & Mains Synch" several times every 1-3 minutes but it ceases when the load transferred to the bigger capacity genset. The 282kVA genset is a mechanical governor while 939kVA is electronic governor genset with ECM. We already tried to adjust the stability of the lower capacity genset (frequency varies from 59.9 to 60.3 hZ). We also adjusted the voltage same with the bigger capacity genset and also replace AVR and speed controller of the 282kVA but still couldn't cease the alarm.

Is it possible that the problem was on the UPS and not the genset? Or is the genset is to small to carry the load of the UPS? The UPS is rated at 160kVA.

 

[FreddyNurk]

Its entirely possible that the UPS is set such that it expects to see a frequency deviation that may be normal when supplied with grid power, but flags issues due to the frequency deviations that are often normal for a generator. This is likely to be adjustable given the size of the UPS.

Its also likely that the larger set has better frequency regulation, and is less affected by smaller load changes than the little one, hence less issues with the big set.

 

[waross]

The frequency is set by the governor, not the AVR.
The common governor setting for those size generators is 3% droop.
The no-load speed will be 61.8 Hz dropping to 60 Hz at full load.
An electronic governor typically responds to load changes faster than a mechanical (hydraulic?) governor.
As the load varies on a diesel generator, he speed varies. The governor detects the speed change and makes a correction.
From the information provided it is probable that the frequency is changing due to large load changes.
For a given load change, the larger set will have only about 1/3 the change in frequency as the smaller set.
This is a rough approximation. The actual ratio will be affected by the inertia of the sets, the governor action and the actual power of the prime movers. More accurate then 1/3 may be "more than 1/4 and less than 1/2".
Three solutions:
1. Get on the larger set sooner.
2. Find a UPS that is more tolerant of frequency swings.
3. Live with it.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[ScottyUK]

Depending on the UPS, it may not just be the absolute frequency values but the rate at which it changes that is causing the problem. The inverter can only track frequency changes on its bypass up to a certain slew rate, and with a small machine with a large load the rate of change may well be exceeded.

 

[gpi2019]

Thank you all for your comments.
I will try to make the management understand the problem we are experiencing on the back up power. Maybe i'll suggest to replace either the generator set or the UPS.
Just another question. Can hiring a Power Quality Engineer help me pin point the cause of the alarm on the UPS?

 

[Skogsgurra]

Not likely. A PQ guy usually doesn't care much about other things than harmonics, transients and voltage levels. This is beyond most such guys abilities. Get yourself a decent recorder that can handle voltage and current wave-forms with good resolution. The Picoscopes are very good and there are units that can handle your voltage directly. Or use voltage dividers for their eight-channel units. Fused test leads, of course. And current clamps with voltage outputs. Like the Chauvin-Arnoux clamps will help a lot. Rogowskis have their limitations but are sometimes easier to apply. I prefer DC clamps to be able to see sub-mains frequency events. Such an "investment" will cost you less than USD 3000 and that is probably less than a PQ guy will cost you. Doing this will give you a much better understanding what is going on and probably enable you to find the solution as well.

Over-priced PQ analysers hide most of what is happening and concentrate on power quality as such. They are blind to most other phenomena.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.