Harmonics and Nuisance Earth Fault 1

[hossam77]

how do the harmonics lead to an earth fault in motor (more than 75KW) ?

 

[davidbeach]

How is the earth fault being sensed? Some methods of sensing zero-sequence or residual current can be influenced by triplen harmonics (3rd, 6th. 9th, ...) since those harmonics are zero-sequence harmonics. A relay that only looks at the fundamental frequency (60Hz or 50Hz) would not be influenced by the harmonics.

 

[laundry]

hello hossam77
I would like more info how do you know your earth fault is related to harmonic distortion.and have you had your supply checked recently.also do you have alot of inverter and welding machines.

 

[hossam77]

hello laundry...
if we have a motor (400V, 200KW, delta) with high insulation resistance (2G ohm), the earth leakage relay gives a trip signal at the starting of the motor. if i increase the time delay (0.3 sec) to 1 sec the motor start normally and no trip signal appears. so what can make the earth leakage relay feel that there is a zero sequance component in the start up interval?.
i installed a ring CT of a DATA LOGGER around the lines (3 phases) and it recorded a current value in the start up interval and then during runing the current value is zero. ... how can we explain this?

 

[ScottyUK]

Are the three conductors symmetrically disposed within the ring core? The ideal would be for each conductor to pass through the centre of the ring. The easiest place to mount such a CT is often on the inside of the gland plate after the armour has terminated but before the inner sheath opens up. How is your CT mounted?



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

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

 

[Skogsgurra]

Sounds like your starting transient is saturating the CT so that the sum of the three phases temporarily is > zero. It happens quite often and setting a 1 s delay is OK in my eyes. Using a CT with a higher rating could also help.

Gunnar Englund

http://www.gke.org

 

[laundry]

Hello Hossam77
Have read some intresting replies.You have how ever sugested you may have so Harmonic distortion on your mains.if you do then this is most probably the cause of your tripps,as harmonic content, are increased losses in motors transformers and cables, also can cause interference in quit a number of electronic devices.So increasing your trip time to 1sec seems to have cured the problem for now. but I would have my supply looked at by ABB and if you do have a problem then I would but in some filters including high-pass filters.but I am curiuse is this a new installation or have you fitted a new s/b latley ? good luck

 

[RRaghunath]

hossam77,

Is the earth leakage relay connected in the residual circuit of the phase CTs in the subject motor feeder. I have a reason to believe this as your mails do not mention that the relay is connected to ring CT.

It is quite common for the earth leakage relays connected in residual circuit to maloperate during motor starting. A short time delay should take care of this.

The alternative is to go in for ring CT for the earth fault protection.

In addition, having a relay tuned to fundamental frequency also helps avoid maloperation during starting periods.

 

[torslum]

For certain, the magnetizing current of a transformer is containing a large amount of third harmonics (i.e. when starting up the transformer, the inrush current contain a lot of third harmonics), and I guess the same goes for magnetizing a motor.

I would guess, as davidbeach says, if the earth fault monitoring is sensing the zero sequence currents it would be affected by the motor start. As longs as the fault is gone after startup, I don't see any problem with this.

 

[DickDV]

Generally, it is bad practice to measure anything on the motor leads between the inverter and the motor. CT's, clamp-on ammeters, etc. will almost always register large error and the data will not be useful.

Far better is to use the analog outputs of the drive to indicate the desired data.

Many drives have internal ground fault detection. I would trust that more than an independently wired ground fault relay.

 

[Marke]

Hello hossam77

If there is a significant high frequency component in the current being measured, then you can get a capacitive earth leakage due to the capacitance of the motor windings to earth and the cable to earth. This causes problems with VSDs and somtimes with soft starters if the earth leakage protection device is able to respond to transient currents.
Low frequency harmonics are generally less than the fundemental and would not normally be an issue. I would look for fast wavefront waveforms due to SCRs switching or inverters operating.
Best regards,

Mark Empson

http://www.lmphotonics.com

 

[hossam77]

thanks for every ..
i 've got good informations ,,
but i have two motors with the same power and the same source (transformer) and only one of them that has this problem ... the other startup normally with 0.3 sec time delay while the other needs about 1.3 sec, and in the past it was starting up normally at time delay 0.3 sec ... but something changed that leads to increasing the time of the nuisance earth fault for this motor. this is appeared on this motor after spending about 6 months without running.This is what confuses me. the motor has an earth relay (thytronic) with ring CT and without filters for harmonics. The isolation of the motor is high.
what i want to know also .. according to what do the current and the time delay of the earth fault relay adjusted. i found 3 identical motors, 2 of them are adjusted 0.3A, 0.3sec , and the 3rd are 0.6A , o.6 sec.
i work in a refinery all L.V motors (400) without space heater except the M.V motors (6.6KV).

 

[Marke]

Hello hossam77

Some years ago, I came up agains a problem that occured at the first start of the day only and the problem was traced to moisture inside an isolator placed near the load. The fist start would actually clear the moisure and any megger tests done after the first start showed clear.
The client did a test before the first start on a Mondya morning and found that there was indeed leakage.

Best regards,

Mark Empson

http://www.lmphotonics.com