2000 HP Chiller

[mpparent]

Gents,

I have an issue w/ a 2000HP chiller, and wondered if I could get some insight. Originally, the chiller was fed by a 2000kVA, 4160V substation, that was about 30 yrs. old. We then replaced the substation transformer with a new one (I think the only change may have been the new one had a slightly larger Z). The chiller has a primary-reactor starter (fed directly from the sub via bus) which has a GE 369 motor management relay. Now, before the sub xfmr was replaced, we had no issues. What is happening now is the chiller sometimes will drop off (the 369 takes it out on UV from voltage dips on our grid) if running, and sometimes if it is not running and the operators want to start it, it will not start (once again, they are telling me the 369 takes it out on startup on a UV condition). An operator called me yesterday, and said they couldn't get this chiller to start. He said it had been sitting for 3 or 4 days. I asked him just for grins, to try starting it again (I'm not aware of how long it was between starts). The chiller started with no problems. Any ideas or insight is appreciated.

Mike

 

[jraef]

Well for one, it seems to me that 2000HP from a 2000kVA transformer is cutting it close to begin with, so a "slightly larger Z" as you put it (assuming you mean higher impedance) could make all the difference. Coupled with a slight voltage dip from the utility to begin with and you can easily be in trouble. This is one of the best arguments for using a solid state starter vs electro-mechanical. When conditions change, it is often merely a mater of tweaking the adjustments. In this case, you will not be able to do anything short of changing your reactor or transformer.

"Venditori de oleum-vipera non vigere excordis populi"

 

[rbulsara]

while jraef's comments are valid to most extent, I would think the larger Z really makes difference in starting. If it is starting all right (which it appears to be, at least on most occasions) and not tripping on Overload, I would first look in to adjusting the settings of the UV.

What is the UV setting, btw? A larger Z could cause some greater voltage drop during a momentory dip, but it the motor is not tripping on overload, UV trip is most likeky a nuisance trip..

 

[jraef]

Rbulsara is right. I was stuck on explaining why it worked before and why it trips now. What you really need to do is determine if the UV trip settings are correct for your application. You may also consider increasing the trip delay time so that you can tolerate a transient UV condition long enough to start successfully. UV trips are there to protect against continuous low voltage conditions. If, as rbulsara says, the motor does not trip on overload there is nothing inherently wrong with a voltage drop during starting as far as that motor is concerned.

"Venditori de oleum-vipera non vigere excordis populi"

 

[busbar]

Aside — The relay might also be tested to assure its time-voltage characteristic has not drifted.

 

[electricpete]

In addition to change in transformer Z, also look for any change in transformation ratio.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[mpparent]

e-pete,

Definitely change in Z, but the xfmr ratio is the same as before.

Mike

 

[aolalde]

The secondary voltage drop of your transformer is directly proportional to the transformer impedance.impedance or current increase the voltage drop.

V2=aV1-I2/Zt

a= transformer ratio
V1= phase primary voltage
I2= secondary load current
Zt= transformer impedance

You need to match the UV setup to the new expected voltage drop.

 

[mpparent]

Thanks for all of your responses,

I have some new data now that suggests the grid is "sagging" quite a bit longer than the delay we have programmed in, and the RUNNING UV setting that we currently have. I'm going to change the setting from 8 to 10% and see how that works. I'm also going to try to contact the chiller manufacturer to see if the chiller can tolerate that low of a voltage for the low voltage period.

Mike

 

[mc5w]

You may also have to change the primary taps of your substation transformer. Your old transformer might have been set to produce 4,320 volts or even 4,400 volts at no load. With load the voltage would drop down to 4,160 volts.

Mike Cole, mc5w@earthlink.net

 

[RRaghunath]

UV trip is always time delayed. What is the delay set in the relay in the present case. May you can think of increasing the same without change in the UV setting, if you like it that way. It should work.

 

[ScottyUK]

Mike,

When you say 'the grid' is sagging, do you mean the HV side of the transformer is seeing a sag, or the LV side? If your motor is causing a noticeable sag on the HV side, then expect the utility to come calling once they realise where the sag originates, because it will quite possibly affect other customers fed from the same HV source.




----------------------------------

If we learn from our mistakes,
I'm getting a great education!

 

[mpparent]

All,

The time delay I believe is set for 4 or 5 seconds. The UV relay is seeing the sag of the LV side. The odd thing is that I would expect the same response from our other chillers as well, but they remained in operation during the event. It's quite strange to me,...maybe something else is going on.

Mike

 

[mc5w]

What I meant was that the taps of the old transformer may have been set to compensate for the impedance of both the transformer and the supply system. A 4,160 volt system that powers only 1 load can have a no load voltage as high as 4,400 volts without harming anything. When the load is running the voltage drops down to 4,160 volts.

The same concept is used in small control transformers that are say 130 volts open circuit to compensate for voltage drop during the inrush period of contactor and relays coils.

Mike Cole, mc5w@earthlink.net