Parallellin CT's 1

[chadwiseman]

I have an application where I need to use a breaker that has four sets of 300/5 CT's in its bushings. I need a much lower CT ratio (around 75/5 would be ok). Is there any reason why we cannot connect the CT's in parallel to acheive an effective ratio of 75/5? Is this something that anyone has done before, or seen done?

 

[dpc]

If the CTs are identical it can be done. Blackburn has some discussion on this in his Protective Relaying book.

 

[cranky108]

With such a low ratio, is the secondary current under a fault going to be a problem?

Is this for metering?

 

[chadwiseman]

This is for protection. Unfortunately we have a 500 kVA xfmr that is being protected on 46 kV bus. The full load is only 6 A, so we need a low CT ratio to be able to provide overload protection for the transformer.

What do you mean by "is the secondary current under a fault going to be a problem"?.

Thanks.

 

[ppedUK]

As dpc said, it can be done if all CTs are identical (knee point, VA etc, not just ratio!), but remember that all CTs still have a 5A output, so if you had a fault level of say 3000A, then under fault each CT would output 50A (before saturation), so the protection scheme would then see 200A which I suspect would probably blow the relay, especially if you have a high fault level.

Paralleling CTs is sometimes done for metering, but metering CTs saturate at much lower currents than protection CTs.

Not recommended. Change the CTs or fit interposing CTs.

 

[cranky108]

Try to use an over current relay at the 300/5 ratio for large faults. Then use a saturating auxilury CT feeding another over current relay for smaller faults.

Or if you consiter the cost of the transformer to be small, and the risk low, you can just not try so hard to protect it.

 

[davidbeach]

I opened the first relay manual at hand and find that the AC current input withstands are (for a 5A input):

Continuous - 15A
1 Second Thermal - 500A for 1 second, 1250A for one cycle. It would take a fault of extremely long duration for 200A to be of any concern.

I'd far rather see the CTs paralleled if they can't be replaced than to see interposing CTs. I find it hard to believe that there is only one relay that needs to be connected to the CTs of that breaker though. One of the CTs just needs to be replaced with one of a better ratio.

 

[cranky108]

davidbeach, you are correct. I guess I'm showing my age.

 

[RRaghunath]

Low CT ratio and high fault level - Phase O/C protction / high set element should be selected to act on 'peak' rather than on RMS.
This is to prevent relay non-operation on CT saturation.

 

[SolkorSte]

Please excuse my limited knowledge, but we parallel 300/5 type "x" ct's on 11kv Solkor protection, to summate 500KV TX current from the feeder circuit. In paralleling this in our humble method is literally connecting the two S1 and S2 connections together at the solkor panel. However as the Primary currents travel different ways through each set (one set in RMU and one set in TX GHV termination box),they detract rather than add. However, if the CT wiring has been installed incorrectly, and the primary current is driven the same way through the CT's they ADD their currents, and still retain the 300/5 ratio. ie if on the incorrectly wired circuit, I inject 300A Primary, the secondary output is 10A. and for the same injection on the CORRECTLY wired circuit the output is 0A. Basically my question is how would connecting the CT's in parallel reduce the ratio...............Once again my experince is only on 11kv solkor systems, so i apologise if i am WAY of mark here!!!

 

[ppedUK]

davidbeach, I used the 3000A fault level and 200A secondary current as an example to illustrate a point. If the fault level is higher, which I suspect it will be, then the problem gets worse. The withstand of the secondary wiring would need checking as well, or you could end up with the secondary wiring acting like a fuse under fault conditions.

I still say don't do it if the fault level is significant.

 

[scottf]

I can't see paralleling 4 CTs to be a good idea. The mismatch in all of the exciting currents is going to cause a significant circulating current flow in the secondary wiring. This may or may not be a problem if you're just using them for over-current.

Also, when using bushing CTs, getting ratios this low is tough to do while maintaining some suitable protection accuracy rating.

It's just not possible to provide a 75:5A CT on a bushing, since that would only be 15 secondary turns.

I'm a CT guy and not really a relay engineer, but from my perspective, the correct way to protect this installation would be to use free-standing CTs, where a lower ratio is possible...assuming you can't use the 300:5A CTs without paralleling them or have ruled out using an aux CT (which I'm not sure exactly why you have).

 

[scottf]

Forgot to add that doing a burden calculation on 4 CTs in parallel would be more of a very rough guess than actual science. Aux CTs, would be much more simple in comparison.

 

[davidbeach]

Even better yet, use just one of the 300:5 CTs and use a relay with a 1A current input rather than a 5A current input. Now your 300:5 CTs instantly become 60:1 CTs.

 

[slavag]

Yep.
David. Im fully agree with you. It's standard solution in this case. STAR.

Best Regards.
Slava

 

[ScottyUK]

It's a good idea. Just be careful to very clearly mark both the non-standard relay and the relay housing if you do this: a mistake involving a 1A relay and a 5A relay took out a fair chunk of London's distribution system. As a Northerner I found this vaguely entertaining and probably wouldn't have bothered restoring the power at all, but it nicely demonstrates how easily it can happen.


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

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

 

[Ohmly]

In a certain region of our network, there are 1 Amp relay connected to 5 Amp CT's (being used for earth fault detection)

These relays are continually failing (open circuit) - this relay is normally very reliable - and only fails when used in this manner.

We think that being an electromechanical relay with a large burden at the low setting, the voltage from the CT is damaging the relays (any other theories would be appreciated !). We are looking to replace the relays in this area.

I guess with a modern relay, the buden presented to the CT is low, so it wouldn't be a problem - I just thought I would mention it.

 

[scottf]

ohmly-

What is the protection rating, ratio, nominal primary current of the CT and input voltage rating of the relay?

Also, failing open circuit is generally an indication of over-current and not over-voltage.