High voltages seen on CT's 2

[bspace123]

Hello

We have a 6.6kV switchboard drawing ~100A as per this single line diagram: Link

The switchboard has two sets of current transformers connected opposite in parallel, acting as a differential scheme. When current flows into and out of the board, the currents cancel and the protection relay see's no current. When there's a fault, obviously the protection relay see's the total fault current.

We observe a crackling/arcing noise at the back of the protection relay when the switchboard is online (relay is seeing very little current as there is no fault on the switchboard. This little current is expected and mainly due to CT/secondary impedance missmatch). The noise is sounding like significant voltages are being induced onto the relay CT input terminals. All connections have been checked for tightness which leads me to believe we are getting some capactive coupling through the primary/secondary of the CT's. We tried to measure the voltage with a 1000V rated multimeter. The voltage unfortunately cooked the meter, so we are unsure of what it was but it must've been >1000V. The CT STAR point is grounded at one end.

The noise dissappears when we short out either set of CT's (i.e. only one set connected to the relay).

Questions:
1. My guess is that the capacitive coupling is discharging from one set of CT's to the other. If we take one set out of the cicuit by shorting the secondaries, the capacitance has no where to discharge and hence remains in a charged state. Is this correct?

2. Is grounding both ends of the CT STAR points, the only way to remove this capacitive voltage from the relay? This is not a good idea as we may see ground loops causing maloperation of the protection relay. Can anyone suggest any other methods?

 

[Compositepro]

A current transformer's output is current. The voltage will be whatever is required to drive that current up to a limit (very high voltage). You cannot have current flow in both directions and say they cancel. Each transformer is trying to make the current flow in opposite directions but that cannot happen. So your CT's are behaving the same as if they were open circuited.

 

[davidbeach]

Properly paralleled CTs would have the difference current going through the relay.

The star point should be grounded at exactly one point. That will keep the conductors in the CT circuit at a voltage near ground.

 

[bspace123]

Thank you for the responses.

I thought the same initially, there may be an open circuit somewhere in the CT circuit, but there is not, as when you short either CT out, the high voltages dissappear off the relay. We have also measured continuity throughout the circuit and all is fine. If there were an open circuit, we'd see the high voltages from at least one set of CT's (even when the other is shorted).

We ground the secondary STAR point to avoid high voltages due to capactive coupling from the primary. My understanding of this that when we switch off the primary, there is no risk to personnel being shocked as the CT circuit is bonded to earth. If they were not bonded to earth and someone were to come into contact with the CT circuit, with a stored potential up to the primary voltage, then they would potentially create the discharge path (if one were to exist) for the capacitance and be at risk of electric shock.

Understand the protection risks of bonding the CT's are both ends. But if this understanding is correct, then how do we prevent these capactive voltages from being present on the secondary? The discharge path for the capacitance appears to be between the CT's?

 

[JG2828]

It sounds like there is a loose connection inside the relay. You may want to open up the relay and look for signs of arcing and check for loose connections in the current path.

 

[Palletjack]

If I understand correctly, you have two sets of opposing polarity CTs connected in parallel on a terminal board, before the connections get to the relay?
If so, your answer is above by compositepro.

 

[davidbeach]

Should not develop any voltage if properly connected and if there’s no resistor. Current in and out sums to zero. Everything else takes a low impedance path through the relay. As a partial diff there will always be some spill current through the relay.

A problem in the relay is a possibility. Measure the voltage across the current inputs of the relay, should be essentially zero. Internal to the relay there should just be a short loop of wire through an aux CT.

 

[scottf]

When you say you tried to check the voltage and the fried the meter, were you measuring voltage between the input terminals or from one terminal to ground?

Capacitive coupling causes the potential of the CT's secondary winding, as a whole, to rise...not the voltage between the secondary terminals.

Do you have the ability to view the current Wave shape with the relay or a scope?

 

[waross]

One ground, properly applied will avoid capacitive coupling.
The first sign of an open CT secondary is often a fried meter.

We observe a crackling/arcing noise at the back of the protection relay when the switchboard is online

That is another common indication of an open circuit on a CT secondary.

when you short either CT out, the high voltages dissappear off the relay.

Another common sign of an open CT secondary.
Fix the d___ relay before someone gets injured.

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

 

[newelecteng]

hi bspace123

is this problem appear after maintenance or from day 1?
did you try to replace the relay?
can you give us clear photos of protection scheme drawing

 

[bspace123]

Thank you for all the responses! Turns out it was a poor earthing connection on the STAR point causing capacitive voltages to build up and discharge.

 

[LionelHutz]

I'm curious why certain members think that a differential scheme where the differencing is done via the secondary wiring of 2 CT's can't work, especially considering how many drawings and working examples of such a CT wiring scheme can be found.