high nominal CT burden

[144x]

a CT has been ordered.the nominal burden of the measuring core of the CT is much higher than the real burden.so it does not saturate soon enough to protect the measuring devices.it is suggested to use external resitor.
but there is the possibility of resistor failure and opening the secondary of the ct.any suggestions?

 

[jwerthman]

Having a high CT burden capability compared to the connected burden is generally considered a good thing, at least for relaying applications. Saturation of the CT is undesirable because there may be insufficient current to operate the relays. I've not heard of depending on the CT saturation to protect the measuring devices. Some type of transient voltage suppressor with a breakdown voltage less than the tolerable voltage across the measuring device burden might be a solution, but I have not specific suggestions regarding devices or ratings.

 

[peterb]

What is the accuracy rating of the CT? Was it specified for protective relaying or metering purposes?

Typically, a metering CT will give rated accuracy at currents up to the rated value and will saturate at a low voltage, while relaying CTs will maintain rated accuracy up to the rated overcurrent factor, with saturation ocurring at a reltively high voltage.

A resistor is used in some cases where, for example, you need to increase the effective pickup current of a high impedance differential scheme. This operates by shunting current around the relay coil and can't be used if you need accurate measurement of current.

 

[scottf]

144X-

This is actually a common problem. I assume you are in the IEC marketplace and have purchased a metering core with a security factor rating (Fs)? You can put a resistor in the circuit if you so desire, keeping in mind that the IEC practice of using the security factor to protect the meter is normally overkill. In all reality, the meter core you have will not be able to pass through enough voltage during a fault to damage your meter.

If you are worried about the meter, my recommendation would be to use the resistor to increase the burden to the CT. I would not be that concerned about the resistor developing an open circuit, as the open-circuit voltage of metering cores is normally relatively low.

If it was me, I wouldn't add the resistor. If you're on the selling side of the equation, the added accuracy of the CT being operated at a burden lower than rated will probably pay for many meters over the life of the CT (and I don't think the meter will fail).

 

[Alex68]

Hi 144x
I have some experience of CT because of my job.
In order to help you, I think that we need more information about the problem.

-Is it a measuring or protective CT? or all together?
-What is the nominal and max primary current?
-What are the ratio, the rated burden, the Fs, the accuracy?
-What is the actual burden, considering also the cable?
-Is a P or TPS (TPX, TPY, TPZ) CT?

The saturation of a measuring CT automatically protect the measuring device, because in case of fault the high primary current saturates the core and the secondary current is very low and composed by harmonics.

Please clarify better the problem.

 

[12291973]

Hi all

Let me put my doubt on the same subject. Does the CT saturation depend on the Burden? If yes, Can any one explain why?

AP

 

[144x]

the dynamic current of the measuring device is lower than
the short circuit current at the secondary side of the measuring core.
Fs(actual)=Fs(nominal)*[(Pct +PN)/(Pct+Pp)]
PN=nominal burden
Pp=actual burden
Pct=ct internal resistance burden
Pp for a digital measuring device is very low
so actual Fs where the CT goes to saturation becomes high.
I think we have to use resistor anyway.

 

[Alex68]

Usually we interpose a measure transducer between the CT and the meter. The measure transducer output is 4 to 20mA (properly scaled) and it is self-protected from too high currents. In this way the max current in the meter is 20mA, without risk.

 

[144x]

Alex68
that's something to consider.it may be the solution.

 

[jbartos]

Suggestion: Regarding the extra protection if the secondary CT opens, this needs to be evaluated since the open-circuited CT produces high secondary voltage (it may be around 3000V rms) that can cause sparks, with VA depending on the CT rating. These sparks could damage insulation and ignite combustibles, if any present. Therefore, an extra resistor or overvoltage protection could be viewed prudent to add. However, if the CT secondary circuit installation is somewhat safe then it is not necessary.
Normally, the CT standards include the CT max burdens. It is understood that the circuit will be loaded more toward the high side (close to CT VA rating) than the low side (a small fraction of the CT VA rating). Please, notice that if the circuit is loaded with very small current there are problems arising from contact potentials at various connections that can potentially become somewhat lose.

 

[scottf]

12291973-

The CT satuaration point most definately depends on the connected burden. To put in simple terms, the higher the burden, the higher the "power" that must be delivered to the secondary circuit to maintain the correct current. You have to imagine the CT as a constant current source. For a given core/coil design (i.e. diameter, cross section, number of secondary turns)the higher the connected burden, the lower the point of saturation.

Keep in mind saturation is measured with the relation to excitation current versus secondary terminal voltage.

Alex68 -

If this circuit is for revenue metering purposes, what is the accuracy of the transducer you use. They are generally not very accurate (in relative terms) and not used very often in revenue circuits. If the application is for general indication, then of course it doesn't need to be that accurate.

Any idea how the transducer "self protects" against high dynamic currents? I have never heard of this before.

jbartos-

Most likely, if this CT has a security factor rating per the IEC standard, it will have an open circuit voltage of 1 kV or less (depends on core size and number of turns), since the material used to manufacture these cores has a different magnetic permeability than regular M4 steel used for most applications in the ANSI marketplace.

Just my 2 cent worth.

 

[gvakre]

144x
1)The core for metering porpose should saturate if the fault current exceeds predetermined value(Instrument Security Factor- IEC Standards).This factor may be 5 or 10.
You may use the saturable core across the secondary winding which will draw negligible current at rated overloaded current (normally 120% of nominal current).The core should be thin and the nos of turns should be quite high so that very small Excitation Currents will be drawn.The Voltage at which the Saturable reactor should saturats is =ISF*0.9*(ActualVA/Is+Rct*Is). If you wish I can send you the design if you provide full data/details of CT.

2)If the CT measuring core is used for revenue metering then it will read more energy than the actual one. This is because a CT designed for high Burden when used for much lower burden will produce less errors and as normally the measuring cores are given the turns correction( i.e reduce few turns from secondary to correct the errors-this makes the secondary current more than nominal Esc Current-thus produce positive side errors)the resulting sec current can be more than nominal and the error may be more that the designed class of the CT.If you are user of power you will be paying more on the energy bill.

 

[12291973]

Scottf

From the application guide for the ALSTOM ISTAT transducer, it has t continuous rating of three times the rated current and can withstand 25 times the rated current for 3 seconds and 50 times the rated curent for one second.

Regarding the accuracy, Will that be fine if we use 0.2 class transducer? But Care should be taken to select True RMS transducers. IF we use mean sensing transducers , the reading is going to change if harmonic is present in the I/p quantities. But sometimes it is going to help if you are in the generation side as i think the error is in the positive side ;-)

AP

 

[scottf]

gvarke-

I agree that the CT will generally be more accurate when using less than the rated burden, but I do not believe in true practice, that the RCF (ratio correction factor) will become positive, even if using secondary turns correction.

Also, have you considered the effect on accuracy by placing a saturable reactor across the CT's secondary. I agree it will change the saturation point of the entire circuit, but it will also change the accuracy, i.e. under nominal conditions, the excitation current of the saturable reactor will not make it to the meter, i.e. the current will be lost from the output. When the desired errors are in the range of 0.2%, I do not believe this would be a workable solution.

 

[gvakre]

scottf

If 144x can provide the details of ratio,accuracy, rated and actual burden and STC level,rated voltage etc practical solution can be worked out with saturable reactor.
1)The error of the CT if .2 class can swing betwwen +0.2% to -0.2% at rated current with variation of VA from 0.00 to rated VA.Normally the RCF of 0.15% is given by removing the sec turns so that max +ve error at lowest burden(i.e.25% of rated voltage-IEC) and by designing the core section max -0.15% at full rated VA.Thus even 0.2 class can also read +ve error and if operating voltage is high -66 kV or more the revenue charge will be very much high and will be permanant.
2)For satuarable reactor,the excitation current can be reduced to very low current(of the order of 0.5 milliamps(in case of 1.00 amp rated current) or so)by using Mu-Metal(Nickel Iron) core which will introduce the -ve error of 0.05%.This can be practically worked out with proper details of CT under question.

 

[144x]

CT parameters for its measuring core are as follows:
600-300-150/1 A
0.5Fs10
40VA
Rct 75C <4 ohm on tap 600
system voltage 66 KV
total connected burden of the lead and measuring device
1.2 VA

 

[gvakre]

144x-
It is not possible unfortunately to design the saturable core for very low VA of 1.2.For satuaration factor of even 10 it will not be possible to design the core without exceeding the errors limits.
Hence only alternative is to use the resistor or transducer. You may use few resistors of higher wattage in parallel to avoid the danger of burning the resistore and open circuiting the secondary.

 

[scottf]

144x-

After gvakre has confirmed what I advised earlier...if it were me, I wouldn't do anything to protect the meter, just like in the US IEEE/ANSI markets, where no security factor parameter exists.

 

[gvakre]

Yes thanks scottf.Incidently meters manufactures here in India confirm that the meters are designed to withstand 10 times the rated current for 5 seconds. I do not know about static or microprocessor base meters i.e their capability of withstand high currents.We are trying to educate the utilities hear through Indian Electrical and Electronic manufacturers Association not to specify for ISF as it increases the cost of CTs and some times impossibl.
Can some body furnish the data /experience in this matter?