Metering CT saturation 2

[undervoltage]

At how many times of rated current normally a metering CT saturates? I have a metering CT of 0.3B0.5, 600/5A class and want to use it with a protective relay which will pickup at 600A. Can I use this CT with the relay or I need a relaying class CT?

Thanks

 

[waross]

I believe that metering CTs will maintain accuracy up to 200% of rated current.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[jghrist]

IEC metering class CTs will have an Instrument Security Factor (Fs) that defines the overcurrent at which the metering core will saturate. This limits the secondary current to protect meters during faults. This obviously restricts the accuracy during faults.

Secondary current is not deliberately limited in ANSI metering class CTs, but there is no assurance of any accuracy during faults if there is no relay classification.

 

[scottf]

There is absolutely no standard multiple of primary current that metering CTs will saturate. It varies widely and is also a function of the applied burden.

The only certainty you have is the rating factor, e.g. if the rating factor is 1.5, then you can be assured it wont saturate at 150% Inom, etc...

From a practical perspective, it all comes down to the burden that the relay will present. If you can stay below a 0.5 ohm burden (round-trip leads and relay input impedance), then I suspect it will work for you, if you simply need it to ride to 600A or slightly above (up to the rating factor of the CT).

 

[pwrtran]

At how many times of rated current normally a metering CT saturates? I have a metering CT of 0.3B0.5, 600/5A class and want to use it with a protective relay which will pickup at 600A. Can I use this CT with the relay or I need a relaying class CT?

That is a no no answer. Never every use metering class CTs for protection, they are not meant to.

0.3B0.5 just means 0.3% accuracy class and 0.5 ohm burden, it will not tell you when it will saturate. You never size the metering CT less than 10% minimum load conditions and more than rating factor.

To find out when the metering CT will saturate, get the typical excitation curve from the manufacture and find the knee point, it will close enough to the saturation voltage. But I can tell you it will very likely saturate under fault conditions!

If you use metering CT's for protection, your relay will not operate properly. However, you could use protection CTs for metering, but not revenue class, and it have 10% error.

 

[LSpark]

Most of the metering CTs on the sites I look after are only tested to 120% rated current. If you only need it to remain accurate to 120% rated current then you should be OK but a protection class CT would be preferable.

I'm suprised there aren't any suitble protection CTs, I usually find several protection class CTs but no metering class ones when a site is looking to install metering on a previously unmetered circuit.

 

[scottf]

pwrtran-

I agree with your sentiment, but in this case, with a 600:5A metering CT being used with a relay that will trip at 600A, it should work.

Of course it also means someone sized the metering CT very poorly

 

[bacon4life]

Although the CT work for a 600A fault, the relay may fail to operate as expected at higher fault levels.

For a much higher fault, the CT only produces secondary current for a small portion of each half cycle. If the fault had a DC offset, the CT could saturate in the first half cycle and not begin producing any output again until the DC component had fully decayed away.

Depending on the relay, it could also respond unexpectedly to the high levels of harmonics generated by a saturated CT. If it operates only on the 60 Hz component, it would vastly under measure the current.

SEL has a number of tech papers that describe CT saturation criteria such as

http://www2.selinc.com/techpprs/6038.pdf

If your fault levels are low and your relay can detect CT saturation very quickly, you might be able to use the existing CTs.

 

[scottf]

If the relay trips as current passes through 600A (nominal current rating of CT), what difference does it make how the CT performs at higher currents?

 

[davidbeach]

If it is an instantaneous trip set at 600A, it would probably work. If it is a time overcurrent element with a 600A pickup, it will take far longer than expected to trip, and might even drop out under severe saturation.

 

[bacon4life]

Even an "instantaneous" relay needs some time to operate. For example, GE UR instantaneous relay elements take 1.1 to 22 milliseconds to operate depending on the ratio of actual current/pickup threshold. I suspect it wouldn't take much DC offset or residual magnetization to saturate the CT before this particular relay could pick up.

http://store.gedigitalenergy.com/FAQ/Documents/F60/GET-8400.pdf

There might be higher speed fault detector relays for use with bus differential or breaker failure schemes that have faster response times.

 

[pwrtran]

If the relay trips as current passes through 600A (nominal current rating of CT), what difference does it make how the CT performs at higher currents?

You are right if the secondary of the metering class CT is able to detect the primary current without distortion when it saturates. However, it may saturate in less than ONE cycle.

Protection class CT normally can withstand 20 times of the primary current (100A secondary) with the secondary distortion less than 10% or we can say it measures 90% of the primary current. I doubt a metering CT can handle 100 times of the primary rating.

When a metering class CT saturates, the flux in the core cannot alternate, it basically maximum out, thus no alternation on secondary. In other words, the secondary output is cut off.

There are some graphs that show secondary distortion in the article attached below.

So, get back to your point, the CT may not have the chance to get 600A primary pickup point before it saturates since the fault happens rapidly and the flux have no chance to change its direction because the core flux is already built up! Another fact is once the CT saturated due to the DC component of the fault, it will reside in the core unless you de-magnetized it.

 

[pwrtran]

Sorry forgot the attachment

 

[scottf]

All of the above statements are correct, but they don't consider the connected burden. As I stated, if the burden is 0.5 ohms or less, the metering class CT will ride up long enough for an instantaneous trip to work correctly.

Now, the metering class CT would ride up to 20 time overcurrent with 8 ohms applied (which is what a C800 rating means), but it doesn't mean it work under the prescribed conditions. If the relay is a modern electronic relay placed fairly close the to the CT, the connected burden is likely well less than 0.5 ohms.

 

[pwrtran]

Without having the metering class ct's excitation curves and the secondary winding resistance values, it is safer no to say "it should work" - my thought.

Here is an IEEE calculation sheet using that one can find out when the selected CT will saturate or not under a given fault and some other conditions.