Instantaneous to Ground Trip on Chiller Compressor

[ndwade]

Hello fellow engineers,

I work at a Chiller and Refrigeration Plant where we are experiencing Chiller Tripping issues without being able to find the root of the problem.

Our York chillers seem to 1st experience a "control power failure" during start up then once they reset the control power and go to start the Compressor Motor for the Chiller, it trips at the main gear. I've been told that this mostly happens to just one Compressor, but it has happened to others as well.

The gear trips off on Instantaneous to Ground w/ the following current values across each phase;

I(a) - 1,645A
I(b) - 1,589A
I(c) - 1,135A
I(G) - 1,211A

Here, we see that there is a high ground current and the Chiller trips on Instanteous to Ground with the extremely high current going to ground.

The avg operating current is about 200A for the chillers. The motor is 4160V, approx. 1600 - 1700 HP motor which runs @ 3500RPM. I've been told that the motors have never fully been taken down for maintenance in their 10yr history.

Because of the absence of burning or melting @ the starter, we are under the impression that there is a problem with the motor that is causing the high currents to go to ground.

As I am fairly new here, i'm relying on the word of the people that were here before me who say that the motor for the highly problematic compressor has been hypot-tested and megged out.(the method of the hypot-testing was questionable so i've been told)

In your opinion, what type of problems does this situation point to? Motor, motor starter, switchgear, Power quality?

Any input or help that you can give is much appreciated.

Thanks for you time and I look forward to hearing your views.

Respectfully,
Nadir

 

[rbulsara]

Need more data to validate and some experienced engineer needs to get involved before you do anything.

Having said that, it sounds like the motor is drawing normal starting current. The control power fail may be because of temporary voltage sag during starting and the INST ground fault trips probably is a nuisance trip because of some imbalance or wiring issue.

If every thing all seems OK (no sign of actual ground fault), I would start with putting some delay like 0.3 to 0.5 sec on the ground fault setting and see what happens.

INST ground fault setting is seldom advisable.

You may want to post the folling info to get a better reponse:
What type of starter it is and any settings on it.
The type of relay and schematic of GF protection scheme( or describe it).

Rafiq Bulsara

http://www.srengineersct.com

 

[oldfieldguy]

I'm with Rafiq.

Our ground trips have a small delay. The first example I pulled up from a file has 30 milliseconds.



old field guy

 

[waross]

Possibly a bad connection in the CT secondary wiring.

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

 

[dpc]

Has this been going on for 10 years or just start recently? Any changes to the system, equipment or settings recently?

David Castor

http://www.cvoes.com

 

[ndwade]

Thanks to all for your replies to date. I am attempting to get more information on the equipment that way I can narrow down the subject a little further.

This chiller tripping issue has been going on for the last 4yrs. The records for these trip events are non-existent. Most of the info is provided from the head of operations who has witnessed this occurence over the last 4yrs.

Ii do not know of any changes to the settings but it is something we are looking into. The problem is intermittent and our chiller rotation varies so its been difficult to see the frequency of the tripping.

I've spoken to the manufacturers of the switchgear (Eaton) and they seem to think the ground current is significantly high and may point to some wiring problem or external problem other then the gear.

We are still investigating and thinking about bringing in someone with more experience on this specific equipment.

Thanks and I'll keep looking more info.
Please feel free to throw out any ideas which come to mind.

Regards

 

[Zogzog]

You need to do some real testing of the breaker, cables, and motor. If you lack confidence in your in house people ability or test equipment I suggest hiring a testing firm to look at it for you.

 

[electricpete]

It was a good clarification this is a longstanding problem. That means you have also had many successful starts. Now (since you asked), we can say the probability of an actual electrical fault is very small, particularly in the motor.

At our plant also, our protection is 51G, not 50G.

Many good suggestions and scnearios above.

One thought would be that you have brief saturation of CT's due to the combination of high starting current along with decaying dc component. This may add to the problem (along with relay setup). If you provide CT type, we may be able to get a feel for how close to saturation.



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

 

[electricpete]

Ideally, one would need to know length and size of cable and list of everything connected in the circuit to judge saturation.

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

 

[electricpete]

...I'm talking about CT secondary cable.

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

 

[rcw retired EE]

What type of starter? Across the line, reduced voltage, wye-delta? When switching from start to run mode, a high current can occur on some systems, saturating the CT's and giving a false Ground Fault trip.

The fact that the motor stays running indicates a low probalbility of an actual ground fault.

 

[rhatcher]

I would be surprised if you have a CT on the ground conductor that is actually measuring ground fault current. Instead, I guess that your metering is measuring ground current based on a residual ground scheme. Specifically, a large CT has the three phase conductors pass through it and the balance of the current is measured. Theoretically, the balance of the current in a three phase circuit should sum to zero. Any imbalance is assumed to be flowing through the ground circuit.

The phase currents you give show an imbalance with the 'C' phase being low by about 30%. When adding the phase currents with 'A' = 1645A>0, 'B' = 1589A>120, and 'C' = 1135A>240, I get an imbalance of 484A>54.

Coincidentally, the ratio between the metered ground fault current of 1211A and the residual ground fault current of 484A is almost exactly 2.5.

I would suggest that you have a problem with the metering. You should check to be sure that the proper CT ratio is applied to your ground fault relay.

As others have said before, if you actually had a ground fault of 1211A, you would know it due to the resulting smoke, carbon, and melted metals that would result.

I also suggest that, if possible, you should set your ground fault relay to a time delay of 0.10-0.30 seconds as this allows for imbalances that can occur during motor inrush currents. Settings that are instantaneous or are in the millisecond range (practically instantaneous) can be problematic.

 

[HamidEle]

Starting motors can definitely cause nuisance trips in grounding circuit.We expereinced High unbalance current flowing through Grounding circuit, which actually tripped upstream relay. We inreased the setting on both phase and ground. In the mean time, we set the grounding relay delay at 3s. Since then, we have seen any trips at all.

 

[electricpete]

I would be surprised if you have a CT on the ground conductor that is actually measuring ground fault current. Instead, I guess that your metering is measuring ground current based on a residual ground scheme. Specifically, a large CT has the three phase conductors pass through it and the balance of the current is measured.

You are correct my comments do not apply to all ground detection configuration. They would apply to residual connection on CT secondary, but not to core balance CT.

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

 

[PHovnanian]

As others have surmised, the motors are probably OK, based upon their having started and run successfully for the past 4 years.

I'd chase down the 'control power failure". Its possible that something is amiss here. A reduced voltage starting scheme, compressor unloading equipment or even the motor contactor might not be working properly, leading to starting problems.

After 10 years, the motors and control gear might deserve a bit of manintenance attention.

 

[ndwade]

Thanks all for your comments on this. Since my last post, I have been collecting some specifics behind the protection.

The ground protection is of the residual type with the following settings;

50N/51N
CT - 400:5

Phase Protection Settings
LDPU - 0.65 (260A)
LDT - 20sec
Curve I2T
SDPU - 7 (2800A)
SDT - 0.1sec
Instantaneous PU - 10 (4000A)

Ground Protection Settings
LDPU - 0.1 (40A)
LDT - 0.3 sec
SDPU - N/A
SDT- 1 sec
Instantaneous PU - 2.25 (900A)

It seems that 4 of our compressor motors which are on our Northbus trip while 2 compressor motors on the Southbus do not trip. All protection settings are the same for all 6.

I have been trending all of them recently and found that the 2 on the South bus have even higher starting currents yet no tripping occurs.

I believe the tripping is occuring due to some imbalance. But still haven't found the source of this imbalance which is causing the ground fault trips for the 4 Northbus compressors.

More comments and troubleshooting techniques are welcome as I've checked all settings.

The differences in the tripping occurences in the North vs South bus is puzzling me.

Thanks for your time

 

[electricpete]

Since you mentioned residual connection, the CT saturation scenario is still in play.

Why differences in busses? Here are some thoughts just thinking out loud....
Is it possible that different types CT’s are used on the 2 busses? Or different configuration of relays fed by the CT’s. Or different distance for CT secondary wiring. Less likely differences in source impedance angle that might have a small effect on the L/R decay of the current (depends mostly on motor, source impedance should be a small effect I think).

Ideally if you can record the three CT secondary currrents during a north bus motor start, you might get some clues. You can sum the currents together to see if they in fact reach the setpoint. You can probably tell if CT saturation errors are present from the waveforms. If for some reason the answer isn’t obvious, you might record south bus start for comparison.

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

 

[ndwade]

electricpete,

Soory for the confusion, they aren't actually two different bus but rather two different sides of the bus connected by a tie breaker.

I am in the process of monitoring and recording the start-up currents for all chillers.

I am yet to see a trip condition where a ground fault is registering but as the temperature increases and we'll be turning the chillers on and off, one is bound to rare its head.

I'll continue to investigate some of the issues you have pointed out.

Thanks for your help.

Other comments are welcome as well.

 

[waross]

How are the contact faces in the tie breaker?

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