High ratio CT with less load current

[kssschsekhar]

Hi,
We would like to use high ratio CT with less load current with the following details:
The Maximum load current is 75A
Protection CT ratio: 800/1 A (5P10)
Numerical protection Relay- O/C & E/F protection (50, 51, 50N & 51N)
Available fault current 40kA
Max.Load current is 8% of CT primary current.
As continuous operating current of the CT is less than Ankle point of the magnetising current and relay pick-up current is 15% of CT primary current, is there any anomalies in this configuration.
Advanced thanks...

 

[slavag]

Hi
It will work fine.
Aren't power meter and metering core.
Best Regards.
Slava

 

[scottf]

I agree with slavag...unless you're doing any steady state measurements, then it will work fine.

 

[bacon4life]

I am not very familiar with IEC ratings, but wouldn't this transformer saturate at 8000 A?

My gut feeling is that this would work fine, but I would like to be able to to show for sure that saturation does not cause the inverse time overcurrent element to slow down.

 

[dpc]

With modern, low-burden relays, this is actually a pretty good approach. I wouldn't recommend it for an electromechanical relay.

Too often we see multi-ratio CTs such as 1200/5 that are tapped down to 100/5 for relaying. Using this low of a tap greatly reduces the accuracy of the CT during fault conditions.

 

[scottf]

bacon4life-

He's not talking about saturation, but "under-excitement" meaning operating in the point of the excitation curve often called the "boot", which is where the curve becomes non-linear again as the current gets lower.

This doesn't come up a lot in protection and often isn't even shown on excitation curves generated.

"The boot" is effectively what causes metering CT accuracy to fall off exponentially as the current gets below some point, depending on the type of core material used, etc...

The effect is often worse for protection CTs, since the cross sectional area of the core tend to be larger than metering CTs, hence the flux density is lower and accuracy falls off more rapidly as the current drops below rated...normally very pronounced as it gets down below 10% of rated current.

 

[cranky108]

So in relay applications the boot normally dosen't matter because we are looking for larger currents, right?
In metering applications it is important as it effects the accurcy of the meter at the low values.

Interesting.

 

[kssschsekhar]

Thanks for all your discussion. As a conclusion, we can agree,
For relay applications the boot normally dosen't matter because during faults currents are higher than boot region of the CT.

 

[stevenal]

Interesting new terms, boot and ankle. Are these equivalent (it's an over the ankle boot)?

Scottf,

So the rule of thumb is 10% of rated for protection CTs? Do you have a different rule of thumb for metering CTs? How about dual rated CTs?

 

[scottf]

Stevenal-

For metering CTs, the error at various percentages of rated currents are defined by the accuracy class.

For instance, in the IEEE world, a CT with a 0.3 accuracy has the following performance:

0.3% error limit from 100% Inom to rating factor
0.6% error limit from 10%Inom to 100%Inom

 

[cesgibis]

some Toroidal protection CTs need to be sensitive enough to pick up low earth fault currents. should a low ration CTs (e.g. 50/1A) be selected ? They will provide a good accuracy down to 5A primary (10%)
The CT will saturate quickly for high fault currents in this case. but this should not affect the tripping of the relay.
However some toroidal CT used on LV systems and rated 1000/1A are used to pick up on very low earth fault currents.
Can anyone please provide assistance on this ?

 

[ScottyUK]

bacon4life,

To answer your earlier question about the 5P10 CT rating, at 8000A (10x rating) it would be within 5% of stated accuracy.


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

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

 

[EP007]

We have 5 Amp neutral grounding resistor (NGR). In this case, high fault current is not occuring. The CT in the transformer neutral ground connection is 200:5. The CT has been tested at low currents so we can see the errors and be satisfied on performance.

 

[Kiribanda]

Kssschsekhar,

Under steady state conditions it serves your purpose.

But, donot you think that at 40 kA fault, this CT will get saturate within no time and the protection objective is lost?

Knowing the X/R ratio, if you calculate the time to saturate it will be less than even one ms.

So in my opinion, if it is for a medium voltage application, you may have to think another way of achieving the protection objective.

 

[ScottyUK]

Kiribanda,

How do you expect to get a 40kA earth fault in a high resistance grounded application, other than through a gross failure of the earthing system?


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

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

 

[Kiribanda]

ScottyUK,

The post does not say

1) 40 kA is ground fault current

2) it is a high resistance grounded system

It implies that 40 kA is the 3-phase fault current.

Did I missed something?

 

[slavag]

Hello.
Kiribanda.
40kA is only 50Inom from 800/1A.
In case of ( sorry for term) "normal" secondary burden/load
of CT, nothing do to CT, no saturation.
Please pay attention, saturation isn't function of primary current, or maybe better say not only of primary current.
We ahve so much application with 75-100/A primary CT with fault current up to 8-10kA, no saturation in the bolted 3-ph fault too ( from time to time personnal forgot earting on the trafos phase).
Second, back to OP, 8% is nominal current, 15% setting, no problem with saturation too.
And last, on the HV side of trafos installed 100,200,400A CT with 40kA fault level no problem with 87T, 87B protections.
Best Regards,
Slava

 

[bacon4life]

This paper from SEL cautions that CT's rated much less that fault current could cause the instantaneous overcurrent elements not to operate.

http://www.selinc.com/techpprs/6142.pdf

SEL notes that the current inputs are only linear up to 20A with a 100A for 1 second thermal rating.

Beckwith simply notes that current inputs are rated 30A for 2 seconds.

As a newer engineer, I would have been hesitant to design something that operates well outside of published specifications, but it sounds like the voice of experience on the forum says not to worry about it.

 

[ScottyUK]

Kiri,

Mis-read EP007's post 23/10/08 "We have 5 Amp neutral grounding resistor (NGR). In this case, high fault current is not occuring...." as being from the OP. My mistake, sorry!


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

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

 

[slavag]

Bacon4life.
Please pay attention again on the OP and Socottf's answer to you.
You are right in all your points, but OP about other issue.
This thread is mix of several terms and issues.
and of course, newer and "older" EE design ONLY according to
mnf. instruction.
Good Luck.
Slava