Nuisance tripping on feeder breaker & 67P1T setting (Wind plant)

[adi 3291]

Nuisance tripping issue with one of the feeder protection relays (SEL-351) and the event file shows the 67P1T protection element caused the trip. Due to certain initial energization parameters on a few circuits around the fleet and transformer inrush amps cause nuisance trips upon energization and the feeder trips quickly. 67P1T element is set to 12.00 and there is no 67P1P pick-up setting.

This is a 34.5KV feeder breaker, that is connected to appx 25 wind turbines each rated for 1.5MW, and total feeder FLA is 580A @34.5KV. Please see the attached layout. Below are the current settings.

50P1P = 0.39A
50P2P = 8.79A
67P1T = 12

51P1P = 4.4A
51PTD = 0.5

This is a Nuisance tripping for us and been there for some time. Would like to understand and look at ways to prevent the issue rather than what we are currently doing i.e. remove the 67P1T element temporarily from the trip.

Would like to understand your comments on it, appreciate the support. Thanks!!

 

[davidbeach]

Actual event files? Pictures don't help much.

I’ll see your silver lining and raise you two black clouds. - Protection Operations

 

[adi 3291]

Please see the attached for one of event file on this feeder breaker

 

[stevenal]

I would suggest changing the traces to rms values. You can then see how high above 50P1P you get. With the calculation feature, you can even put in a .39*1200/5 line for comparison. Then adjust 50P1P for a comfortable margin.

 

[pwrengrds]

The attached file isn't valid, comtrade files require *.CFG and *.DAT files to open.

It's hard to judge settings without knowing the CT ratio, but a good guess is 240/1 (1200/5). If that is true then the time over current setting setting is 1056 amps (1.8xFLA) and the Inst 2 over current is 2110 amps (3.6xFLA).

The 67P1T is set at 93.6 amps (0.16xFLA) with a time delay of 12 cycles which is less than the operating current. This would indicate the 67P1T setting is used for directional tripping toward the turbines and should be blocked for a couple of seconds on inrush or blocked by harmonic inrush (we do both at our wind sites for some settings like 51Q). The actual issue with the site is likey in the torque equation for the 67 element (67P1TC).

Knowing the actual settings (which are in the *.DAT file) would be helpful so we aren't guessing as to what's going on.

"67P1T element is set to 12.00 and there is no 67P1P pick-up setting" is not true in "SEL" programming.
The way SEL logic works is to use the 50P1P as the pickup, if you want to use it without delay (50P1) in the trip logic you can, then if adding a time delay 67P1D, the trip equation is 67P1T. 67P1T = 50P1P (pickup) + 67P1D (time delay) + 67P1TC (Blocking bit). 67P1T element is set to 12.00 means there is a 12 cycle delay (0.2 second)

Best of luck,
David

 

[cranky108]

Look at the traces, is it doing what you asked it to do?
Question, why did you asked it to do that?

 

[protoslash]

if you're certain the relay is nuisance tripping on transformer inrush, then SEL already have a solution.

It's called 2nd harmonic blocking logic, basically during inrush, current contains high magnitude of 2nd harmonic signal. The relay detects this condition and uses it to supervise various overcurrent tripping. Check out this paper, it has some examples/logic you can try.

https://selinc.com/api/download/21474836652/

 

[thermionic1]

I used the SEL inrush blocking feature with good success. The Comtrade inrush tripping events were used to playback via a relay test set to the relay to verify proper operation. The transformers are cycled several times a month and have been operating without issue for about 18 months. The SEL solution suggests adding in an unrestrained element with a higher pickup to the trip equation as an extra layer of security.

 

[hiker_28]

If you suspect inrush download and view the raw/unfiltered event files from the SEL-351, it will be easier to see. The 4 samples per cycles is filtered and you won't see the harmonics in that file.