Transformer MV earth fault. 1

[squeeky]

Country: Malawi
Soil Type: Sandy.
Earthing network: Tested every year. No problems.
Transformer: Siemens.
Gefol. 2000kVA. 3.3kV to 400v. Dyn11.
MV Cable is PEX. 90 meters.
MV protection: Reyrolle 2DCC IDMTL Fault Level 19.77kA
CT 400:1
θA = 105
θC = 105
Earth = 15
LV MCA breaker Siemens.
Feeding a boiler MCA.
Boiler feed water pump is 250kW, 4 pole. Largest motor in panel. One of two.
Occasionally when the pump #2 is started, as it changes from star to delta, the MV Breaker trips on earth fault.
Motor insulation test shows no faults.
Motor draws normal current.
Star delta starter is wired correctly and has been re-wired.
Earthing has been checked and no errors found. <1 ohm on test to earth mat.
Transformer has been changed.Crack in MV blue phase coil was found.
Protection relay has been changed.
Any ideas?

 

[ScottyUK]

To be clear - MV breaker is the 3300V breaker?

Oil & gas industry seems to have some totally different nomenclature for voltage bands compared to generation / T&D and I keep falling foul of it, so I'm just checking. ;-)

 

[Runsor]

Uneven CT saturation might be the reason. Taka a disturbance recording of the three phase currents when the motor is started and calculate the residual current.

 

[squeeky]

Thanks. I was there in September last year for six weeks and it did not trip the whole time I was there. It tripped twice just before they shut down. I still can't figure out how the LV motor trips the MV breaker. I have a Fluke power quality analyser. I will start the motor and make some recordings. I was thinking about plotting the MV CT curves but I will have to find a way of generating a variable 600A in a country where everything is difficut to find. Thanks again.

 

[PowerElectric]

That is strange...maybe it has to do with the starting current from the motor, of course you know the starting current can be 6 to 8(even more depending on the motor specs) times the normal running current. If that is the problem you can add a backup diesel generator to the motor and have it run when starting the pump.

Yes, please check it out and let us know. For record keeping purposes.

Best Regards

 

[jghrist]

There should be no earth current on the delta side of the transformer. I wouldn't think that starting current would saturate the CTs enough for uneven saturation to trip on earth fault. Check the CT connections. I just had an earth fault trip on an ungrounded capacitor bank during commissioning caused by one phase being connected 600:5 with the others connected 800:5.

 

[dpc]

We just went through a similar trip on a new transformer installation. Turned out to be loose CT terminations at the transformer. The testing outfit had disconnected all of the CT leads in order to test the bushing CTs. They got the leads back under the terminal block screws, but apparently forgot to tighten them.

You could add a brief time delay on the residual ground fault trip to see if that helps. Our local transmission system provider does not use the residual overcurrent protection on delta transformer primaries due to nuisance tripping issues. We generally don't have too much problem with it.

 

[squeeky]

Thanks for the tips. CTs is where I am now going to focus my attention. I'm going off grid into darkest Africa. I will need to find this post when I get back. Two more items of help. If this is a long way down the list, is there a fast what to find it? I keep trying to add attachments and use the up load facility at the bottom of this page but I never see the hyperlink to it when I return to the question. I will take lots of recordings and would like to share the results and possibly the happy ending. Tips required. Thanks again.

 

[slavag]

Sorry.
What is a setting of E/F? 15A with CT400/1A.
What is a time setting?

15A primary is very low setting <<10% of CT primary 400A.

 

[squeeky]

Our Senior Electrical Engineer has software and he sets the levels. He is a young man who comes in with a lot of attitude and wants to rule the roost. What would you suggest? I am traveling again and will be able to look at the laptop data later for the time setting but by memory it is less than one second. P.S. I am the motor guy more on LV than MV but get sent when others are busy.

 

[slavag]

What is motor start current and time of start?
it is your area

in any case, setting must be not less than 40A primary.

 

[waross]

One motor is creating higher transient currents on star/delta transition than the other. What is different? possibly a small timing difference in transition? Is there some reason that one motor accelerates at a slightly different rate than the other? Does one motor have longer feeders than the other motor? Are the star/delta starters identical even to the amount of wear?
An example: If one motor starts unloaded and the other comes in against a head, there may be differences in acceleration time that will cause the transition timing to be less than optimal.
Possibly a timing relay is failing.
Try swapping the transition timing relays WITHOUT changing the settings and see if the issue follows the timing relay.
I have assumed that the star/delta transition is time controlled. If the transition is based on motor current, then check the settings and possibly swap the current sensing components.
And be positive that the other motor has never had an issue. Bad information from the field has done most of us in at one time or another.
An issue in some instances may be the relative phase angles between the star and the delta connection, but I discount this in this instance as such a connection error would likely cause trips much more often.
To recap, three possible issues;
One motor transitioning a little too soon.
Slightly different impedances between the feeders of the respective motors.
Slightly different loads on the motors or some reason that one motor drops a little more speed than the other during transition.
The problem motor may be slightly fast or slightly slow at transition.

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