Multi Ratio CT Testing Question

[DTR2011]

Greetings-

We have a customer that insists that all ratios of a MR CT be tested. We normally test all of the main taps, ie x1-x5, x2-x5,x3-x5,x4-x5. We are being asked to measure the inter tap ratios (x2-x3, x1-x2, etc). It is my understanding that the inter tap ratios are obtained via calculation.

I can not find any information in the IEEE C57.13.90 (Field Guide for CT Testing), that states if the inter tap ratios are to be calculated or not.

I am wondering if any CT OEM's could chime in, or provide some useful insight on this.

Thanks

 

[scottf]

What tests are you performing?

 

[52ali52]

Hi, if all taps are checked in ratio test, it is impossible to find a fault in inter taps ratio's.
x1-x5:1000/1 (suppose 1000 turn and check it)
X2-x5:800/1 (suppose 800 turn and check it )
x1-x2:200/1 ((1000-800) nothing new to check)

 

[DTR2011]

Hi-

We perform ratio / polarity on main taps and excitation on the full tap. Insulation resistance is performed on the full tap. Burden tests are not in the specification. Winding resistance is not in the specification.

I agree that nothing new is found in the inter-turn taps. We are testing with the voltage method, as the CT's are internal to the power transformer.

My question relates to why these tests are requested (inter turn taps), with the only response being that "that's the way we have always done it". As the utility client is paying the bill, I am looking for some justification one way or the other, as the additional tests add a significant amount of time. We have about 400 CT's per substation and 2 more substations to commission.

Thanks

 

[hir]

Dear friend,
You can get voltage value to be injected for between whatever turn we want to test.

Atually this voltage testing methode is adopted in turret CT which are inside the Bushing of the tranformers.

During field testing there is no need to perform interturn - ratio test but when you are facing any problem with regular turn ratio like s1-s2, s1-s3, s1-s4 and if you have any problem with s1-s3 then it will help you to make easy identification of faulty winding section.

Hope this will help you.

 

[davidbeach]

If they're paying for the testing and they want the testing done, why are you arguing with them? You want them to find a different testing company in the future?

 

[stevenal]

So the manufacturer missed a turn at the polarity end of a 600:5 CT with a one turn primary. All your ratios measured to X5 should be right on, except for X1, which will 119:1 instead of 120. This is an error of of less than 1%. If you ratio X1 to X2, however, you will see 19:1 instead of 20, a 5% error.

The question regarding calculation is confusing, since the manufacturer information should show all ten possible ratios. If you don't have the information for some reason, the intended ratios are calculable.

 

[DTR2011]

Indeed they are paying for testing. That does not necessarily mean it makes sense. We test the excitation up to a voltage that would prove that eg 1500 V for a C800 Ct. The spec calls for applying 120V(a single point) and measuring the current.

My understanding is that the requirement has more to do with labor rules than an actual technical requirement. For the additional time it takes, I'd rather perform a DC winding resistance test and a burden test and provide the customer with more "bang for the buck"

Additionally, my concerns is that the ratios deviate by more than 5% in the inter tap tests, yet come out right on the money when calculated.

I am interested in what CT manufacturers have to say about how they are tested in the factory.

Thanks

 

[oldfieldguy]

Let's assume for your CT you can apply 100 volts to X1 and X5. Okay, then while that voltage is applied, use your hand-held meter and measure voltage X1-2,2-3,3-4, and 4-5.

Add them all up, and they will equal X1 to X5, and the proportions will be correct for the reduced winding ratios.

For other combinations, you simply add the voltage measurements, i.e., X2-X4 equals the sum of X2-3 and 3-4. You're treating the secondary winding as an autotransformer.

For that matter, you can apply voltage to any two terminals and read proportional voltages on the other terminals. Using full winding, though, guarantees you don't apply 100 volts on two terminals and end up with a thousand volts across the full winding.

But we're engineers and we're smarter than that, right?

It adds a few minutes to the tests.

It's a good test, too. I've found more than one CT installation where the secondary wiring was incorrectly terminated on the terminal block, and a couple where the internal connections were incorrect.

old field guy

 

[52ali52]

hi
I attended to routine test of current transformer for once. And they tested ct according to nameplate (only highest tap for each core). However if you use inter tap in your scheme you somehow have to test it for your client.

 

[scottf]

Factory accuracy tests for multi-ratio CTs with only a protection accuracy consist of a ratio check/polarity check for each winding segment and confirmation of the protection rating on the full winding (e.g. for a C800 rating, developing 800V on secondary without exceeding 10A excitation current).

For HV CTs, an inter-core coupling test is also done to check to make sure there isn't a shorting path for the primary winding. This is done by applying a voltage to one secondary winding and checking the induced voltage on a different secondary winding.

Of course, all of the other routing tests called for in the standard are also performed (e.g. PFWV, induced test, etc...).

I'm not sure about your reference to "calculating" the taps. Normally when one uses the term "calculated" in regards to protection CTs with multi-ratio windings, it's in regards to the protection performance at tap. For instance, a 2000:5A MR CT with a C800 rating can be calculated to have a C400 performance at the 1000:5A tap. The "C" in the C800 rating stands for "calculated", which means that the secondary winding is applied in such a way that performance at the lower taps can be calculated based on the ratio proportion. The other type is "T" as in "T800", which stands for "tested" and means that performance at lower taps must be tested to confirm the rating. CTs with a "T" class generally indicate that each winding segment is not distributed about the core, which introduces significant leakage reactance.