VT specification for very low burdens.

[Distribution73]

Hello,
I am trying to prepare a VT specification and have come across what I believe is a pretty standard issue:

IEC 60044-2 sets the accuracy clases for burdens ranging from 25% to 100% of the VT rated burden. However the burden of new protection and metering devices that have to be connected is so small that it does not even reach 25 % of 10 VA, which (I believe) is the lowest of the standard VT rated burdens.
For these cases, what is your view on the most suitable way to specify the VT? I thought of the following posibilities:

a) Add dummy load in paralell to the instrumentation to raise the burden and specify VTs with low standard rated burden (10VA).
b) Specify non standard VA rating (as low as 0.5VA) and contact manufacturers. However this will probably be very costly.
c) Specify the VTs using C57.13, which seems to cater for 0VA burdens. However this wouldnt be in line with the rests of the substation specs which is defined in IEC terms.


Your view is much appreciated.
I suppose that this must be a fairly common problem, however I haven´t managed to find any thread advising on this issue. If this is the case, my sincere apologies!

Fernando

 

[Skogsgurra]

Never thought of it that way.

I have always assumed (sorry, Jeff) that the ratio is specified at zero load and that the accuracy is maintained up to rated burden. I often use VT:s for recording purposes and the recorders usually have a 10 or 1 Mohm input resistance, or very close to zero load.
I have never had any problems with measurement discrepancies when compared with data from other equipment in the system.

I have thought like this: If you have .5% at 10 VA, then your accuracy at zero load is either close to 0% or, if the loading effect is compensated for, a .25% too high value at zero load, which implies 0% at 5 VA and -.25% at 10 VA. In either case, that error is too small for me to worry.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[Distribution73]

Hi Skogsgurra,
Your feedback is much appreciated. Surely your take is absolutely fine for the application you mention.
However, one of the VT secondaries i am trying to specify is used for Tariff metering and surely any deviation from the rated VT accuracy will raise concerns. I believe that the VT manufacturers compensate voltage drops in the VT by slightly modifying the VT turns ratio ( I might be wrong though), so if the accuracy is (say) 0.2, at nearly zero burden might be higher...

 

[Marmite]

If your application is for tariff metering then you need to add burden resistors so that the burden comes within the range 25-100% specified in the standard.
Regards
Marmite

 

[Distribution73]

Thank you ver much for the feedback.

 

[Skogsgurra]

That's definitely something to remember, Marmite.
I never do tariff things. Only "difficult and impossible" problems.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[LSpark]

I look after high accuracy tariff metering systems so come across this problem regularly these days. I add a dummy burden in parallel to take the winding up to 25% burden. Where there are two secondaries I burden both up to 25% even when only one of them is used for tariff. Then you can use the errors straight from the test certificate without doing any complex calculations.

Don't do this with tariff CTs though!

 

[Distribution73]

LSpark,
That´s very interesting issue you highlight: "The interdependence of the VT secondaries". From reading IEC 60044-2, I had understood that the "accuracy vs burden" of one secondary should be achieved with the rest of the secondaries loaded from 0 to 100% of their rated burden. If that was the case you wouldn´t need to load secondaries other than the tariff metering ones to achieve the accuracy stated by the manufacturer. Isn´t this actually the case with commercial VTs?
Thank you!

 

[LSpark]

I've just had a look at one of our test certificates (400kV, class 0.2, 20VA unit) and there is a slight variation in accuracy (0.02% and 0.8min) when the burden of the other secondary is varied. That compares with a variation of 0.03% and 1 min between 25% and 100% burden for this VT. So the effect on accuracy when the 'other' secondary burden is varied is similar to that of the winding you are using in this case.

I'm lazy, I prefer to use a known test point (or near enough as it depends on what resistance value I need and how close one of the standard ones is).

 

[Distribution73]

THank you very much LSpark.