CT ratios

[pwrtran]

A CT nameplate is labeled as such:
1200/800/600-5
X1-X4: 1200-5, 2.5L800
X1-X3: 800-5, 2.5L400
X2-X3: 600-5, 5L400
It seems the CT has three ratios on the given tap connections. However, if connect the taps as below you may get:
X3-X4: 400-5
X1-X2: 200-5
The questions are:
1) is this a common practice to get more CT ratios other than just go by the tap connections shown on the nameplate?
2) having above given information how do you think the connections in 1) would affect the accuracy and the knee point?

Thanks!

 

[randyman]

If turns are fully distributed between taps, then you can obtain other ratios. And you can estimate the terminal voltage by V/T method. A simple way is to start with the full winding (800V/240T) then use that multiplier for all other turns combinations. To verify if you stay in the 2.5L class or what the knee point voltage will be will require the secondary excitation curve. Using the same V/T concept, the knee point voltage (if known for any tap) can be used to calculate the others.

As for obtaining other ratios, in the IEEE world you can pretty much use any combination of turns. In the Canadian world you can do it but the taps as you have shown have their guaranteed ratings. If you look closer at the V/T you may notice something else. Some of the taps will yield a higher terminal voltage than the class stated. In application that is OK as intermediate relay classes are not defined by the standards, but can be derived and extrapolated in the same manner. For example:

Your coil turns are x1-40T-x2-1120T-x3-80T-x4
taps turns ratio term V Class
x1-x2 40T 200:5 133V (2.5L OR 10L)100
x3-x4 80T 400:5 266V (2.5L OR 10L)200
x2-x3 120T 600:5 400V 2.5L400
x1-x3 160T 800:5 533V 2.5L400
x2-x4 200T 1000:5 667V (2.5L OR 10L)400
x1-x4 240T 1200:5 800V 2.5L800

Hope that helps!!

 

[pwrtran]

randyman
Thank you very much! I never heard of this V/T (voltage/turn ?) method before. Is the full tap 204T a industry standard or a good assumption? Is there a IEEE standard talking about this method?

 

[davidbeach]

240, not 204, and yes it is an industry standard number of turns for a 1200:5 CT.

 

[capuchi]

david beach,

if i may ask, what would the standard V/T be for a 3000:5? sorry i dont have IEEE book nor mfr's data. would appreciate your advise.

capuchi

 

[davidbeach]

The 800 came from your full winding rating, the 240 comes from the turns ratio. No idea what your 3000:5 is rated voltage wise, but like all 3000:5 CTs will have 600 turns.

 

[scottf]

Not to split hairs here, but the discussion on number of turns is assuming a window-type CT with a single-turn primary.

The same principle holds on the derating of lower taps on a multi-ratio CT however.

In other words, a 1200:5A CT doesn't necessarily have 240 secondary turns, but rather 240 times the number of primary turns.

 

[davidbeach]

scottf, true. It is too easy to fix on a mental image of a CT as something with a hole in the middle for the one primary turn. But there are other types too.

 

[randyman]

I assumed it was a window type by the class designation alone, 2.5L, which indicates low reactance design with distributed turns. Usually, but not always, are wound types not given that designation.

 

[scottf]

randyman-

Having a wound primary makes no difference normally to the secondary winding design, in other words, multi-turn primaries are often found used on CTs with equally distributed secondary windings. Multi-turn primary CTs can and do often have the same ratings as window-type CTs.

Also, 2.5L does not, in and of itself, mean the CT is a low-reactance design. The low-reactance design aspect really only comes into play in the ability to calculate the protection rating at other taps, versus having to actually test it. That's why there is a difference in the IEEE protection ratings between a "C" class (calculated) and a "T" class (tested).

CSA's use of the "L" class doesn't differentiate between equally distributed secondary windings and non-equally distributed windings.

In my opinion, the wording of section 4.4 of CSA CAN3-C13 is pretty confusing (and frankly incorrect) in the use of the term "wound-type" in reference to the secondary winding distribution. I believe what they were after were CT designs that use segmented secondary windings utilizing foil windings.

 

[pwrtran]

Use V/T method the knee point voltage can be calculated. However, how to estimate the accuracy for the ratio listed other than that shown on the nameplate? Appreciated!

 

[scottf]

pwrtran-

In practical terms, accuracy limiting voltage and the accuracy rating are the same thing for protection applications.

The 2.5 in 2.5L or the 10 in 10L normally doesn't make much of a difference in application.

 

[randyman]

Hey scottf,

Couldn't agree more with you on that section 4.4, and unfortunately the IEEE section isn't any better!! Not working much with the CSA standard I hadn't realized they have no designation in place that separates the two. So then how does a user know whether it is a low reactance or high reactance design requiring additional test verification??

There was a time when both bodies used the reference "H" for high reactance designs, and that was most always associated to wound-types. The "L" was always used for low reactance designs which is where I was coming from. One could argue that if a toroidal core is used and the windings are equally distributed, then YES you could calculate performance in the same manner outlined for window-type.

 

[scottf]

Randy-

IEEE uses C and T, with T standing for a non-equally distributed winding design.

I don't really like the term "high-reactance"...I'm not really sure that term makes sense, since the reactance of a non-equally distributed winding isn't really any difference than for an equally distributed winding. What is actually "high" is the leakage reactance. I suppose that's splitting hairs a bit too

When you use the term "wound design" what are you meaning? Do you mean a multi-turn primary?

 

[randyman]

scottf,

When I refer to wound-type I am refering to multiple-turn primary winding. I agree with you too, "high reactance" is a vague term. And you are right about little difference, probably less than the uncertainty limits of measurement. I think in the old days when CTs were stacked "EI"s or cut "C"-cores, then the "leakage" reactance was a bit more significant, and our CT standards aren't quick to catch up wih the times. But we are beating this to death and getting way off the original topic.

From one CT guy to another .. catch you next time!!!