PT voltage class, high resistance grounded system

[JLuc]

Hi all,

I have a simple question concerning PT voltage class signification.

I want to install 3 PTs , wye-wye connected (2400-120V) on a 4160V system.

Since it is a high resistance grounded system, de line-to-ground voltage will rise to 4160V on two phases in case of a fault.

My question is: will my PT tolarate to have 4160 volts on their primary even if it is 2400-120 PTs.

PTs are old westinghouse PTL-5G 2400-120, voltage class = 5kV.

Thank you for your help.

JL

 

[davidbeach]

Unless the VTs have a continuous overvoltage rating of 173%, it would be best not to use them.

 

[JLuc]

Even if the PT ratio is 2400-120, the rating data still says, Voltage class: 5000 volts.

Is it unsafe to use these PTs in my situation?

I know it is not the best option, but I want to know if it is dangerous. The system will be in normal mode (no fault) 90% of the time.

Accuracy is not an issue.

What is the "voltage class" data really means?

Thank you .

JL

 

[stevenal]

The voltage class refers to the insulation. Overvoltage does not just stress the insulation though. From IEEE C37.91:

When an overexcitation condition occurs, saturation of the laminated steel cores of the generator and
transformer can occur. Stray magnetic fields increase in magnitude, particularly at the ends of the cores.
Nonlaminated components at the ends of the cores, which were not designed to carry these higher levels of
flux, begin to heat up because of the higher losses induced in them. This can cause severe localized
overheating in the transformer and generator and eventual breakdown in the core assembly or winding
insulation.

Although the reference is for generators and power transformers, I don't see how voltage transformers would respond differently.

 

[cranky108]

It may not be possible with those PT's (I don't know), but it is possible with two bushing PT's with both having full rating. But the neutrals must be connected to the hot side of the neutral resistor.

 

[scottf]

Jluc-

You will need a 2-bushing VT for this application or a specially designed single-bushing unit with an over-voltage capability of 173% continuous, as has been mentioned above.

The unit you have is a standard 1-bushing unit with a continuous rating of 110% as far as I know. It's a Westinghouse unit that seems to be quite old.

I believe the 'G' in the type stands for 'use on grounded systems' meaning it's a 1-bushing unit for connection from line-to-ground on a solidly grounded system.

 

[stevenal]

I have seen VTs used at the higher rating on resistive grounded systems. For this system, it would be a 4200 / 7270Y:120. Your secondary voltage under normal operation will be about 68.6 volts, so your instruments need to be accurate at that voltage.

 

[scottf]

Stevenal-

That is a common solution as well for protection applications.

However, one has to be mindful that the VT's accuracy will fall off a bit when operated at approx 50% rated voltage. IEEEC57.13 calls for the stated accuracy class to be valid from 90%-110% rated voltage.

If used for indicating metering or protection then it should work fine. It wouldn't be a suitable application/rating for revenue metering though.

 

[stevenal]

scottf,

Interesting. So if you wished to accurately revenue meter this load for the faulted as well as the un-faulted condition, and the remotely located grounding resistor was not to be metered, how would you proceed? Blondel requires three elements.

 

[waross]

I wouldn't try to accurately revenue meter a fault. Your CTs are going to be saturated by a fault. You may try over rated CTs to try to avoid saturation, but then their normal operation will be so far down the curve that you will have introduced more revenue errors than you tried to avoid.
This can be metered for normal operation by either 4160 or 2400
V PTs but they must be two bushing PTs. 5000 volt class regardless. If single bushing PTs are used on a resistance grounded system, the PT cases may raise to close to line voltage during a fault or the PTs on the unfaulted phases may see line to line voltage, depending on which side of the grounding resistors the PT wye point is connected.

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

 

[stevenal]

Waross,

Like the OP, I am considering a high resistance grounded system. The low current SLG fault will not saturate the CTs, and the fault will stay on during operation until it is found and corrected. Wouldn't want the metering to be incorrect during this condition, since timely repair of the condition would not be assured. Since the resistor is remote and the source neutral is not brought out, the VT wye point is on the ground side of the resistor. I know the customer should be responsible for the ground resistor losses caused by their delay, but please consider this scenario for argument.

 

[JLuc]

Hi all,

in my application,PTs are used for indicating only, so accuracy is not an issue.

They are connected line-to-ground, (not line-to-neutral).

The burden will be very low , about 1 VA. I know that by having 173% of the rated voltage on the primary winding, I will lose some accuracy, but as I said this is not an issue for me. I just want to know if this would be a potential danger for the life of the PT.

I joined a catalog page of the PTs I have, mines are #758C841A62.

I know that the PTs I have are not the BEST choice for my application, but I want to know if they can be used before I order the 4200-120.

Thank you for your interest.

JL

 

[waross]

In the event of a ground fault, the other phases will rise to 173% voltage. The transformers will probably saturate. When transformers saturate, the magnetizing current rises disproportionately. Expect either burned out transformers or blown fuses whenever you have a ground fault on the system.

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

 

[JLuc]

Thank you for your comments, I talked with PT manufacturer and he told me the same thing.

Putting 4200 volts on primary of 2400-120 PT = BAD.

I will order 4200-120 PTs.

JL

 

[scottf]

Stevenal-

I think the normal practice is to not try to meter accurately during a fault condition. However, it would seem possible to use 3 VTs connected line-to-line.

 

[jghrist]

Why are Y-Y VTs being used? If they are connected to the ground instead of to the neutral, they will not accurately measure the line-neutral voltage. This is OK if the application requires line-ground voltage measurements. If the VTs are used for metering, I would think delta connection would be more appropriate because on a high resistance grounded system, all loads will be delta connected.

 

[stevenal]

I don't have an answer to the why, since this is partly hypothetical, and distantly related to an old installation where the answers to why are long gone.

I suppose delta wye connected VTs would give voltages in line with the line currents.

Is it possible to specify VTs that would be accurate over the wider range when connected line to ground?

Thanks.

 

[scottf]

stevenal-

It is possible to ask for a voltage range applicable to the accuracy rating which might be outside of the standards.

You just need to tell the supplier at the time of order and understand it will be a special design.

 

[cranky108]

I don't know about metering, but the configuration is still requested by some applications, like mineing and goverment installations. I believe the reason thay find it so attractive is the reduction in ground fault currents.

The problem is it becomes difficult to locate the fault. Is it a bad motor, or bad cable, or a failed capacitor in a capacitor bank?

 

[waross]

Stevenal
Sorry for the delay in answering your query.
I would prefer delta metering. When one phase goes to ground I see a 30 deg. voltage phase shift in two phases as well as a rise in voltage .
To make it more interesting, although the voltage on the PT rises 173%, the in phase component that the KWHr meter responds to will only rise 150%.
When one phase is grounded and the PTs are connected to the ground end of the grounding resistor the connection becomes a two element delta metering connection. The PT on the grounded phase is effectively shorted and the respective element does not function. The meter acts as a two element meter. The connection should measure energy on the delta but will not respond to the energy wasted in the grounding resistor.
With a wye connected metering scheme, the normal metering will be at 69 volts on 120 volt rated equipment. Although the accuracy of the CTs, PTs and revenue meter may be acceptable at this voltage level, it is far from the normal operating range at which the accuracy classes are specified.
I would prefer delta metering where all the equipment normally operates at its rated voltage and there is no voltage rise on the PTs during ground faults.
That said, I have been in situations where the proper equipment was not available and you had to use the components that where on site and "Exercise ingenuity".

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