Metering summation CT connections 1

[ScottyUK]

I'm in the process of modifying an existing MV switchboard to convert a motor starter into a transformer feeder. I have a question regarding the metering scheme.

The original metering scheme is an electro-mechanical design and uses a number of summation CTs with multiple 5A primaries and a single 5A secondary. The new transformer feeder will have independent metering so I need to remove this feeder from the summation scheme. I had intended to short this 'spare' winding, but one of my colleagues has put enough doubt in my mind that by shorting the primary I will effectively add another low-burden 'secondary' which may affect the accuracy of the metering. There's obviously a path for an ampere-turn balance to be achieved through the meter regardless of what happens to this 'spare' winding so I'm not worried about the possibility of open-circuiting the CT but clearly I don't want to degrade the metering accuracy.

Summation CTs aren't something I'm overly familiar with so I would appreciate any thoughts from people who are more familiar with them than I am.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[mgtrp]

I would agree with your colleague. You only need one low impedance path per CT. This is why you do not short unused portions of multi-ratio CT.

 

[ScottyUK]

Yes, it is similar in principle to the test windings sometimes found on big bushing CTs, which are usually left open. I understand that, I'm just not certain what the detrimental effect of shorting the winding would be, other than adding a slight additional burden perhaps due to the shorting link and wiring.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[scottf]

If I follow your question correctly, you should not short the primary input where you motor feeder would have previously been connected. You would just leave it open. That would electrically be the same as there being 0A current in that feeder.

 

[ScottyUK]

Thanks Scott.

Just for future reference, and because I'm gonna get asked, what would the effect have been if we had shorted it?


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[scottf]

I'm not really sure off the top of my head.

Actually, as I think about it, it would depend on the construction of the summation CT. If it is a design where all of the primary windings are wound around the same core and the secondary is also wound around that core, then it actually makes more sense to short the unused primary.

However, there are other type designs that are constructed differently.

I change my previous answer to "ask the manufacturer".

 

[stevenal]

I'll disagree with scottf again. Assuming a single core, he had it right the first time along with the colleague. The windings don't know if they are primary or secondary. Since the unused winding has no source, it is more convenient to consider it a second secondary winding. Sum the primary amp-turns to see what the two secondary windings produce in total. To see how the currents divide, you will need to know the minute differences in burden. No need, though, since you can see that any current through the shorted winding subtracts from that through the meter and your meter will under-read.

 

[scottf]

Stevenal-

If it is a single-core design, you're correct that the winding doesn't know if it's a primary or secondary winding. Therefore, if the winding is left open-circuit, it effectively presents infinite burden and will drive the flux density of the core towards saturation.

If you had a conventional CT with 1 primary winding and 2 secondary windings, you would obviously short an unused secondary winding, right?

However, the impact still depends on the construction of the unit. If it is a high-ratio (meaning 1 or 2 primary turns, it may be better to leave the winding open. If it is a wound primary with many primary turns (like a 50:5 or 100:5), then it might be better to short the winding.



Bottom line...he needs to ask the manufacturer.

 

[stevenal]

"Therefore, if the winding is left open-circuit, it effectively presents infinite burden and will drive the flux density of the core towards saturation."

Not if there is another non-infinite burden winding the primary can link to.

"If you had a conventional CT with 1 primary winding and 2 secondary windings, you would obviously short an unused secondary winding, right?"

If single core? No, it will divert ampere-turns from the intended path through the instrument. This would be like shorting the unused taps of a multiratio CT.

Actually tried putting an extra shorted conductor through the window of a CT once unintentionally in the form of a ground loop. Bad things happen. I think I wrote about it here. Excitation test found no knee, no saturation.

 

[ScottyUK]

Thanks for the discussion guys. The manufacturer isn't there to ask anymore, and the company which absorbed the last remains of the OEM can't maintain records of their own stuff, let alone legacy products they inherit.

It's a low ratio type, wound primaries, wound secondary. Nameplate says 5+5+5+5+5+5 / 5A

Physically it looks rather like a conventional toroidal CT with a lot of terminals.


I can follow the logic of Stevenal's suggestion to leave it O/C to prevent current in the 'spare' winding contributing ampere-turns and robbing them from the secondary. I am not certain that an additional secondary is always shorted - I am pretty sure that the few large bushing CTs I've seen with test windings had the test winding left open circuit. Maybe I'm mistaken.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[rasevskii]

Well, suppose it had one less 5A primary winding wound on the core, IOW a 5+5+5+5+5 instead of a 6 x5A. The unused winding would be the same as if that winding did not even exist, in other words open circuited or as if it was wound with twine instead of wire. That means that you have to leave that input open not shorted. If you short it, the output winding will have a reduced output, depending on the impedance connected to it compared to the short circuit impedance.

I remember these summation CTs were a common item in the UK where multiple feeders had to be summated. All the primary CTs had to have the same ratio of course, for this to work.

rasevskii

 

[scottf]

I'll state again, it depends on the construction. There are multiple ways to make a summation CT. They aren't always made with simply multiple winding around the same core.

 

[davidbeach]

Seems to me that if the primary circuit can be open (breaker open) now without fouling up the summation, then you ought to be able to leave the primary open once it is no longer needed.

Just like when you parallel CTs together for bus differential, any combination of primaries can be open at any time without needing to do anything special.

 

[ScottyUK]

rasevskii,

I agree with your reasoning; on reflection my colleague is probably right. Also in agreement with David and stevenal.


scottf,

I don't think this CT is anything special in terms of design, it's roughly 40 years old and the design might well be a lot older than that. I really appreciate the input from all of you. I'll leave it open circuit and report back on any adverse effects we encounter, equally if it behaves itself I'll drop a comment in.

Thanks all.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[DiscoP]

We leave the spare inputs open on the summation cts that we use for metering. We have an often retold story about the time a guy who thought he did the right thing by shorting an input to the summation ct (before my time mind you). It was discovered many thousands of dollars later!

 

[scottf]

ScottyUK-

Measurement the voltage across the open primary. If it's low or close to 0, then you're ok and it was the type construction to leave open-circuit.

Also, maybe there is a schematic showing the construction.

We make some summation CTs in Europe that don't work the same way as we're used to in the US market.

 

[stevenal]

Europe or USA, why use a summation CT at all? If your ratio is 5,5,.../5, why not just parallel the feeder CTs to get the same result? I suppose the summation CT could be used to isolate an unconventional grounding of one feeder CT, but this would not be a common occurrence. Why use a CT if a terminal block will work?

 

[rasevskii]

@stevenel:

I think the reason for a summation CT is that if you paralleled CT secondaries, each with 5A then the total would be for 6 x 5A inputs= 30A output. The summation CT output winding is not the same ratio as the input windings, that is it is an interposing CT with a ratio between output and input.

Example: say that there are 6 inputs of 5A and the primary CTs are 100/5 for 6 parallel feeders from the same busbar. Say that each is carrying 100A out from the busbar. Therefore the total is 600A, and the summation CT output winding would have 5A if all the inputs are 5A, in other words it has a ratio as seen by any input to the output of 6 to 1. (or 1 to 6 if you prefer)

It is a single phase device and so three of them are required.

I think that is how it works IIRC...

rasevskii

 

[stevenal]

Thanks. Missed the + in the spec. It makes more sense now, and I even have a possible application in mind.