CT secondary load and tolerance 3

[IgorMakarov]

I have a problem. I'm working on a design an AMR (Automatic Meters Reading) system. Our local standard for current transformers (wich are very close to IEC and IEEE standards) says that secondary load of a transformer must be between 25 to 100 %. If not, the transformer doesn't work in raited accuracy class.


Then, I have transformers with 10 VA secondary burden, and all that I need to connect -- one microproccesor meter with consumption 0.1 VA (current circuit) only. The wiring is very short and can't give any sufficient load.

What should I do? I can't change the CTs, and this way is not good in this case -- the lowest burden is 2,5 VA, it's too much, too.

I had thought about adding resistors in the current circuit -- but I found a recomendation do not do it becouse angle error increasing.

I heard about special devices for additional loading for the secondary side of CTs. Dear colleagues, can any one tell me aboyt those stuffs?

 

[jbartos]

Question: Which industry standards do you adhere to? IEEE or IEC?

 

[IgorMakarov]

I live in Russia and we have own standards, but they were close to IEC. To bemore precisious, our standard for CTs based on IEC 44-1:1996

I think it not very important. There are similar statements in IEEE standards, too.

 

[Alex68]

Don't worry. You can connect the instrument without any problem.
The ideal configuration of a CT is with the secondary winding short circuited. So the less is the load the better is the ct performance.
Actually the less is the load, the lower is the voltage drop on the secondary circuit and so the lower is the emf required. In this way your ct works far from the knee point of its saturation curve and the accuracy of the measure is greater.

 

[scottf]

Well...a few points on the above:

There is a big difference in burden ratings between IEC and IEEE. IN the IEC standard, it is true that the accuracy is to be from 25% to 100% rated burden. In the IEEE standard, the accuracy is maintained from 0 to full rated burden, i.e. B0.9, B1.8, etc...

Alex68 is correct to a point on the relationship between CT accuracy and burden. The correct way to state it is that CT error becomes less negative (or more positive) as the connected burden is reduced. It really depends on the CT design, as to whether the CT will be in the accuracy class at, or close to 0 burden. The IEC rates the accuracy to be from 25% to 100% burden for a few reasons, mainly that it allows the CT manufacturer to use secondary turns correction or parallel turns to adjust the CTs load line, in an effort to make the CT more cost effective. What this means is that the CT could indeed be out of accuracy below 25% rated burden if turns correction is employed in the design. My company does not typically use turns correction and tests down to 0 burden for IEC or IEEE CTs, but I know others do use it for IEC designs regularly.

If you have a copy of the test report for these CTs you can tell what the accuracy will be at 0 VA at a particular current level. Plot the load line out and then extend it linearly down to 0 VA. Or you can contact the manufacturer and get their input.

 

[IgorMakarov]

Thanks.

I can take the test reports from manufacturer if I buy new transformers. Actually, I have several projects, some of them involve old transformers and test reports inaviable.

Let assume that a transformer has very good accuracy near zero load. But it is not a decision of my problem/

I must make design according to the standard whethertheless good the particular transformer or not. So, if the standard says that load must be between 25...100%, I have to do this.

So, I ask again: Heard some one about dedicated devices for additional loading of CTs?

 

[vic3fan]

You can make the feeding path longer, adding a few meters (feet) of conductor between the CT leads and the instrument. It can be easily made in the panel, and you save some money; in this case, make sure you wind the conductor in such a manner that you will not create a supplementary inductance.
Also, think about the possibility to connect the instrument through an auxiliary CT.

 

[IgorMakarov]

I thout about additional wiring -- but it should have lenght hundreds meters.

To use aux CTs is strictly prohibited by our norms for revenue meterind.

 

[vic3fan]

You can also search for instrument manufacturers who still produce additional resistors used for extending the voltmeter ranges. These resistors should be more reliable than others and it can be easily added in the panel.

 

[scottf]

I do not see any problems with using resistors to load the CT into the 25% or higher burden range. Referencing your first post, I don't see any impact on the phase angle error outside of the general impact of increased burden. Keep in mind that modern electronic meters are almost purely resistive, as is the normal cabling run.

 

[jbartos]

Comment: CT VA rating has to be properly selected with respect to the electronic meter low current draw (high input impedance). Then, the IEC CT should have enough burdens created to achieve 25% minimum burden.

 

[bigmantony]

The standard does state the CT should maintain accuracy with 25% to 100% rated current.
Engineers usually allow greater loadings than required when sizing the CT.

CT manufacturers compensate the turns of the CT to meet the current error but if they compensate by more than the CT accuracy class then the CT may not meet the current error at lower rated burdens hence the requirement for 25%.

You mentioned the problem of the phase difference being affected by adding resistance into the circuit to bring the load to 2.5VA.
You are correct, a resistive load will improve the current error and increase the phase difference, but the test on the CT at 25% rated burden, in your case 2.5VA, should be carried out at unity power factor and therefore the CT should comply.
I am not aware of any special type of device for additional loading but you can add a 0.1 W resistor and if you do I suggest you consider a rating greater than 2.5 watts to avoid any variation in the ohmic values.

Generally the pilot leads for connection from CT’s to the meters are 2.5mm² but if you were to use a pilot loop of 9M of 1.5mm² cable this would bring the loading up to at least 2.5VA.

An auxiliary CT ( interposing CT, IPCT ) was mentioned. This could be a solution and would add somewhere between 2.5VA and 5VA into the circuit but you then would have to allow for the errors of the IPCT. It is also an expensive solution due to the cost of the IPCT and the installation and fitting.

 

[WAT71]

Your question is one I get often

The most common rule of thumb being a current transformer works best at 20-80% of rating so if you have a ct that is 100:5 it is best suited to work with in 20 and 80 amps
The burden on the output on Electromechanical meters is about 10VA however with new meters the burden has dropped lower than 1 va
IEC has similar but not as tight of tolerances as ANSI has

If you are submetering most of the rules go out the window the accuracys and quolity of some of these submeters will make you surprised.

 

[scottf]

WAT71-

I;m not sure I agree with your comments. When you state "works best", I assume you mean in terms of accuracy? If that is the case, CT's "work best" when operated between the rated current and the rating factor current (i.e. maximum continuous current rating). When using CTs for metering, to the greatest extent possible, the CT's ratio should be sized as low as possible, allowing the rating factor to cover the maximum current expected for the installation.

Note that the original question is in regards to the connected burden (i.e. what's connected to the CT's secondary) and not the percentage of rated current.