De-saturating a Protection CT

[burntcoil]

During a techincal discussion, my colleague made a point that if Protection CT saturates, you can not destaurate it to return to its normal state. He said there are two stages of saturation, one is partial saturation (in which case you can desturate it by DC Choking), other is complete saturation where you cannot desaturate it and altimately one have to replace the CTs.

Is it true that we cannot desaturate a Protection CT after complete saturation? or if we can then how?

Thanks in advance for your responses.

 

[electricpete]

I would think that if you can apply alternating polarity excitation of high enough magnitude to force ct far into saturation, then slowly reduce the magnitude toward, you should be able to demagnetize.

=====================================
(2B)+(2B)' ?

 

[scottf]

First, I think you mean that if the CT goes into deep saturation and it it magnitized.

This happens when the CT sees an abrupt current change, e.g. current quickly goes to zero from a higher level. The degree of magnetism is based on the point on wave of when the current goes to zero, e.g. if the current goes to zero at the peak of the wave, then it remains more magnetized than if the current goes to zero near the zero crossing point.

I don't believe it is possible to drive a CT into such deep saturation than the residual magnetism is permanent. Any CT should be capable of being demagnetized.

From a practical perspective, when a CT is saturated and then at some point returns to normal steady-state service, the residual magnetism will work out over time based on the natural primary current level changes over time.

 

[oldfieldguy]

The CT does not "stay saturated", it stays magnetized. This is likely to be more of problem with older equipment where the steel used for CT cores was not quite up to modern chemistry.

A waveform capture of the magnetized CT shows a very distorted output.

As scottf says, this often happens when the CT is seeing aheavy current that stops abruptly.

I have had success by using a variac to drive the CT back into saturation and then slowly dropping the current.

I have also read procedures that use a heavy DC current, successively applied in opposite polarity, but I tend to think that since it is caused by an AC event, AC will fix it.

old field guy

 

[scottf]

oldfieldguy-

How "old" are you referring to? The types of steel used for protection class CTs hasn't changed in a loooong time

I never heard of the hysteresis performance significantly being different with older materials.

 

[waross]

Hysteresis is minimized by a proper composition of the iron and by proper annealing. It has been found that the addition of a small percentage of silicon to iron reduces iron loss and also prevents aging, by which is meant a gradual of the hysteresis loss over a period of time. Aging always occurred in transformers built before the the introduction of silicon steel (about 1910).

From:
Transformer
Principles And Practice
by J. B. Gibbs.
1950 edition.
Originally Copyright 1937

Saturation is the point that further increases in magnetizing force no longer cause an increase in magnetism. I think that the issue here is residual magnetism, not saturation. Power transformers are normally operated much closer to the knee point of the curve than protection CTs. The residual magnetism causes varying ratios of inrush current when the transformer is energized.
Residual magnetism is not caused by saturation. It is caused by an abrupt cessation of exciting current. It may occur at levels far below the saturation region. An alternating exciting current that slowly reduces from a moderate level to zero will remove the residual current.
It is not the fault but the abrupt clearing of the fault that leaves residual magnetism.

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

 

[scottf]

waross-

That post is a bit misleading, in that the degree of magnetism is a function of the level of current when the current abruptly.

The idea is that during a fault, the CT may/will saturate and then when the protection operates, the current abruptly changes towards 0.


A circuit opened during steady-state is likely not going to leave the CT magnetized to the same degree as a circuit opened during a fault. Of course, it varies based on the point-on-wave of when the circuit is opened.

 

[burntcoil]

I need your opinion when the protection does not operates because the CT already saturated. How we can de-saturate it? Can anybody lists the steps to do that? Any real experience to share?

 

[waross]

Scott. It may happen any time a power transformer is de-energized.
Yes it is the abrupt change rather than the level of magnetism. Sorry if I did not make that clear.
Thanks for the clarification.

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

 

[scottf]

burntcoil-

I think you may want to review this thread again. If the protection is not operating because the CT is driven into saturation, that is different than it not operating because the CT has been previously magnetized.

Please clarify so we can offer the best advice.

 

[oldfieldguy]

"How "old" are you referring to? The types of steel used for protection class CTs hasn't changed in a loooong time

I never heard of the hysteresis performance significantly being different with older materials."

I seem to remember that 1940's and older vintage equipment was more susceptible.

old field guy

 

[burntcoil]

Scottf

My primary intent is to know about the CT, Suppose my CT has completely saturated after a severe short circuit, now how I can de-saturate my CT? Can you list the steps for that or give me a specific reference? In what condition we cannot bring the CT back to unsaturated condition.

 

[scottf]

Burnt...

I believe you mean to say your CT saturated and current was shut-off and now it's MAGNETIZED (it's not still saturated).

To demagnetize, use a variac and starting at 0 current, slowly drive the CT up into saturation. Then slowly lower the current down to zero.

This is done from the secondary of he CT with a ammeter connected in series with the secondary wiring and a voltmeter across the secondary terminals. When the voltage starts to increase a lot with small increases in current, then the unit is into saturation.

 

[ScottyUK]

Wouldn't that be when the current starts to rise a lot with small increases in voltage?


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

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

 

[scottf]

yep...I said that backwards

thanks

 

[DTR2011]

burntcoil-

I agree with the variac principle, however you generally want to drive the CT well into saturation, then s l o w l y reduce the voltage.

The one missing component here is how high to raise the voltage. Assuming you are in the US, a standard variac is 120V in - 0/150V out.

If this were anything over a C100 CT, I do not believe you will properly reach the saturation level on the way up / down.

A few options...old school which is to connect the variac to a 5kV VT, connect the ammeter on the neutral (grounded) side. Voltmeter on LV side of VT.

New school is to use any of the commercially available CT test sets, that are designed to test with voltage. I like to see ~1500V on a C800, however the excitation plot will tell you exactly where the CT saturates. The old take 'er to 1 amp never hurts either.

I have some old ComEd literature that discusses how to desaturate when the CT/circuit can not be energized. Roughly it had to do with inserting a variable resistance in series in the circuit. I dont remember all of the details and would not recommend such a process unless there were no other way / primary circuit can not be energied.

My $0.02

 

[scottf]

smallgreek-

That's a good point. De-magnetizing rarely comes up for protection class CTs...it's almost always a discussing limited to metering CTs and demagnetizing prior to performing accuracy tests.

The practical fact is that once a magnetized CT is re-energized, the residual magnetism works itself out over time.

We do have a MV variac to demag CTs with a high knee-point voltage, but we're an instrument transformer manufacturer and most wouldn't have such a device in the field.

 

[waross]

I have seen the higher voltage Variacs used to find the knee point to see if new transformers met specs.
But I think that we are confusing saturation with residual magnetism.
A transformer is saturated when a further increase in the magnetizing current does not cause a further increase in the flux in the magnetic circuit.
Residual magnetism is dependent on the width of the hysteresis curve and is very much less than saturation.
The residual will very quickly be nullified by the rated voltage on a PT, but the residual may cause errors at very low current levels. Taking a transformer to saturation to remove residual magnetism may be safe but it may also be overkill. When you apply a demagnetizing current, scope the current and once the current is symmetrical simply reduce the current to zero.
Yes, I realize that some high ratio CTs may require higher voltages to establish enough flux, but if you just wind 10 or 20 turns of conductor through the window or short the secondary and force current through the primary it should be easy and safer to demagnetize.

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