Overvoltage & 67N Relay Settings in Ungrounded System 2

[nightfox1925]

In an ungrounded system, it is proposed to provide a Wye(neutral grounded)primary-Broken Delta Secondary with a 59G relay connected and a 67N. The 59G will supervise the 67N which in turn will trip the incomer circuit breaker.

I read some articles on this but I would like to request guidance on what settings on both 59G and 67N to based on?

Thanks...

 

[davidbeach]

Where is the 67N applied? If it is in the delta it will respond similarly to the 59G. If in the main circuit it won't do much of anything.

 

[nightfox1925]

The 67N is proposed to be installed in the broken delta winding.

Can we obtain some directionality of current so we can be selective of which breaker to trip during a ground fault?

 

[davidbeach]

A 67N in the delta will only know that there is a ground fault, somewhere. Unless, of course, you set it in the wrong direction and then it will never see anything.

 

[nightfox1925]

I wish to correct my post. We will be using parallel connected phasr CTs as current input to the 67 relay and use the broken connection of the secondary delta as voltage input.

I have two questions:

1. If I properly size the PT, can I connect as much 67 relays to the PT?

2. What would be a typical current and angle setting for this? Can I set the current to 10% of load current? (ungrounded system).

 

[davidbeach]

The maximum current for an "ungrounded" system (they don't really exist) is the charging current of the system. Your transformer provides a (very) high impedance ground reference for the system, perhaps so high that you can't tell the difference.

But assuming an ungrounded system, your 67N relay can not be set sensitively enough to measure the non-existent current associated with a ground fault.

 

[rbulsara]

Finally, the right answer!

Rafiq Bulsara

http://www.srengineersct.com

 

[dier]

For short circuit beetwen phases ( 3phase/2Phasse short circuit) the current is very high so the protection will detect using the information from CT. Based on calculation SC you have to use Iset=0.5xIsc. For ground foult is used zero sequence of voltage 3V0=Vag+Vbg+Vcg. the displacement voltage is used to detect ground fault and to determine the direction.Setting VPHmin must be set les than minimum expexted phase to ground voltage aproximatley 40V. Settings Vphmax must be greater than maximum operated phasse to ground voltage but less than minimum expected operational phase to phase voltage.For nominal voltage 100V aproximtely 75V settings.

 

[burnt2x]

Polarizing voltage can be taken from the broken-delta secondary of the wye(neutral grounded),broken-delta PT bank. But you must ensure you have the capability to sense the capacitive charging current that your system will be capable of delivering during SLG. A zero sequence current transformer ZCT could be used.
If you are allowed, conduct live tests to measure the charging current available in your system(i.e. lines energized; end is grounded safely to ground while current flowing to the ground is measured) or calculate it.
From my own experience, I set the polarizing voltage Vo (could be Vph per dier's post) at 10% of PT rating or 11 V (110V secondary) while the Io setting was about 70% of max ground current available (dier suggested 50%). Please compute the angle between the polarizing voltage and the ground fault current to set the max torque angle within the range where you want the directionality of the relay to pickup. Do not insert any delay at the farthest protection to properly grade your ground fault protection back to the main feeder.
One more word of caution, choose a PT that can carry the burden of all the relays that will take reference from it. I once had a hard coordination problem when the installation I was involved with used two different sets of PT's; one set with 110V secondary and another set with 190V secondary. Hope this helps.

 

[nightfox1925]

Thanks to all your inputs.

Burnt2x, if this is an ungrounded system. Is it possible that the broken delta secondary be able to supply enough current to detect a current magnitude enough to pick up the 67N? I had arguments in the past that there is no enough ground fault current enough to pickup any overcurrent device in an "ungrounded" system.

Can I set the current setting of the 67N to say just above the maximum system charging current?

 

[burnt2x]

What you get from the broken-delta terminals is 3Vo the moment an SLG occurs. You are going to utilize this vector quantity to be able to differentiate cross-country faults from SLG faults occuring in your zone of protection. You will need a ZCT to sense the capacitive charging current 3Io that flows from the system to the ground and back to the system through the capacitive reactance.
The ungrounded system that I have tinkered with had coordination problems because the main feeder protection picked up ahead of the feeder which experienced a SLG. That is proof enough that the ZCT which was set at 70% of the expected charging current sensed the fault current and that the voltage at the broken-delta side of the grounding PT supervised the tripping.
If you will set the current setting just above the maximum system charging current, your protection will miss the SLG. Some people in our trade are used to thinking nothing flows from the line to the ground in ungrounded systems. I am attaching a paper by IPC which will give an idea what expected system charging current values will be at different system voltages.
As I said, it would be best to do a live test to meausure the charging current. I think it was Slava who grilled me on this forum before.

 

[nightfox1925]

Is a 50:5A ZCT (assuming connected burden are ok) sufficient for the application?

I would like to get educated as to how we arrive with the assumption that the SLG current for my application lower than the system charging current in the event of a SLG.

Many thanks...

 

[davidbeach]

Draw a circuit; where is the current going to flow? Remember there has to be a complete loop before current can flow. Were the system truly ungrounded, you would have exactly one point of connection between ground and the rest or your system, the point of the ground fault itself. The faulted phase would be at the same voltage as ground, but there would be no current flow.

Now add in the parasitic capacitances through which charging current flows. Your phase connected to ground can return current to the other phases through that parasitic capacitance; but remember you now have a third of that capacitance shorted out, so you won't even get to full charging current current.

 

[nightfox1925]

Thank you very much davidbeach, Burnt2x and to the rest...this was very informative on my part.

 

[slavag]

Hi nightfox1925 .
50/5A isn't good solution, if you would like use 67N function for the ungrounded system, you need a good accuracy, not only for amplitude of current, for the phase angle too. From our expirience use 100/1A CT and connect it to 0.2A relay input.

 

[nightfox1925]

Hi Slava...I've been informed by Burnt2X that you grilled him once on the same subject .

I couldn't make out your suggestion. 100/1A connected to a 0.2A relay input?

For directionality, what angle set-point would be ideal for this situation or what would be a good basis for me to set the angle?

 

[slavag]

Hi nightfox1925.
Im sorry for delay, I passed your q, sorry :-(.

Yep, Im grilled Burnt2x once on the same subject, before a long time .

My recommendation of course depend on a type of relay.
Usually for this function is choiced SEF-sensetive earth fault input of relay with 1 or 0.2A. Setting is usually 1%
or 0.01In. You can calculated estimated capacitive current ( from cable data) of whole your system or calculated minimal capacitive current ( prefferable). Suggest is 0.3-0.5A primary ( if you have a long MV cables up to few amps).
100/1A is "optimum" for the size/price/data.
It's smaller errors in the amplitude and angle in the small current zone.

Directionality..you need check your relay data, it's must some standard set point angle for the ungrounded systems, I suggest -90deg.
Now, you can set this relay for the 0.3-0.5A primary, 15-20V secondary in the broken delta, -90deg angle. and test it with live single phase SC.

All this from my expirience with 6.9kV ungrounded system at the one site. Im not calculated data, meas on the live SC, but fro my pinion 0.3-0.5A is good for start w/o any calculation, like to system with Petersen coil, where we put 1A primary w/o any calculation for tens years, w/o problem

Hope that help.