VFD Tripping ground fault relay

[Simmi123]

Hi All,

I a new to this motor control with VFD but I know issues related to harmonics on 480 V or 600 V on low voltage systems but to some extent it can be addressed by using inline filters.
This situation is on the secondary of a 13.8kv/600 V side of a undeground transformer which is feeding a CDP panel that has six 400 A breakers connected to it. Each breaker is feeding a 250 HP fan motor connected through VFD. Each of these breakers have a SE-105 startco ground fault relay and as soon as the VFD starts the motor, Ground fault relay trips. I did some research and found some mA current being at DC bus.

Please advise, is it a common occuring phenomena. I would appreciate a word.
Thanks

 

[Simmi123]

This is the first time on my site here. I would appreciate, if any one could drop a word.

Thanks

 

[LionelHutz]

An ungrounded transformer should not be able to support any ground fault current.

But then, there is some capacitance coupling to ground and the Startco could be reading this current.

The Startco could also be having trouble handling the harmonics on the VFD.

This will likely be hard to pin down without further investigation to find the actual issue. Try disconnecting the motor cables and running the VFD without the motor.

 

[burnt2x]

Simmi123,
Please check if you installed the required parallel path isolator (PPI). Go to startco web and download SE-105 manual.

My 2 cents.

 

[Simmi123]

Thanks burntzx. What this isolator would basically do.

I have been doing some research on my own and figured that
basically common mode noise between VFD outputfilter and motor cable is tripping the ground fault relay. This common mode noise being reflected in terms of zero sequence current flowing back to the NGR of the transfomer feeding this breaker through distributed capacitance.

Now, there are two relays actually, NGR has SE-325 and the neutral and SE-105 on the feeder breaker. It looks like that both are seing the ground current, it is just that because SE-105 is coordinate to operate faster so it is tripping first.

Please advise if this makes sense to you.

 

[LionelHutz]

I read your question wrong. It sounds like you have found the problem. Does the output filter have components connected phase to ground?

 

[burnt2x]

PPI is a device by startco that helps ignore ground parallel path from other ground source other than the monitored cable. you can surf the startco website and download the techical information.
Or as Lionelhutz pointed out, pls see attached.

 

[rockman7892]

That is a very interesting technical paper pertaining to this issue. I have a couple of these scenarios in my plant and was never aware of this being a potential problem. Now I know to look for this should I have a similar problem with any of my drives.

Two things I did not understand however.

The Origonal post stated that the transformer was an ungrounded transformer, and then later referenced a NRG (Nuetral Ground Resistor) Is the transformer secondary ungrounded, or is there a NRG?

Secondly I have heard the term distributive capacitance used in the past but never understood where it was stemmed from. My vauge understanding of it is, that all feeders have some sort of capacitance to ground (Between cable and ground) and the longer the run, the larger the capacitance. Can someone explain further?

 

[CoryA]

Hello Simmi123,

The SE-105 Ground-Fault Ground-Check Monitor is an analog relay with a ground-fault circuit low-pass filter having a frequency response of 25 to 400 Hz, and may be prone to harmonic frequencies developed by the VFD. The SE-325 Neutral-Grounding-Resistor Monitor could also be affected; the trip-coordination scenario you described is likely - an additional delay might extend beyond the drive's ramp-up period. Both the SE-105H, and SE-325H options have a frequency response of 25 to 110 Hz for use in such situations. The microprocessor-based SE-134C GF-GC Monitor, and SE-330 NGR Monitor use digital filtering, and are even better suited for harmonic-rich applications. The use of shielded cable might reduce the amount of harmonic ground current by reducing phase-to-ground capacitance unbalance. Are you using shielded cable?

I'll try to answer your question too rockman7892. Consider Simmi123's trailing-cable application. The phase conductors are energized to system voltage, and run parallel to the ground conductor. This, essentially, becomes a capacitor - two plates at differing potentials separated by a dielectric. An increased cable length is analogous to an increase in capacitor-plate surface area; both result in an increase in capacitance. Is this description helpful?

The following paper will be an informative read for the two of you.

Lowering the Limits for Earth-Fault Detection

It discusses topics related to both your questions in greater detail than my short response.

---
Cory Anderson

http://startco.ca/