Overcurrent relay at generators: True RMS / DC ? 1

[stason]

Hi guys,

what current is usually measured by the overcurrent protective relays installed at synchronous generators below 10 MVA?

Do they measure the true rms value (that is, all harmonics) or are some harmonics filtered out?

How about the DC offset during a fault or transformer inrush? Is it included in the rms value?

Is there a general practice or does it depend on the manufacturer and his definitions?

Thanks.
Regards

 

[AMBMI]

I think it depends on the manufacturer.
Basically any relay has a low pass analog filter.
Then simple over current relays are usually calculating the RMS value simply integrating in a discrete way (samples) the sinusoidal current wave. In that case the DC component and harmonics not cleared by the analog filter are part of the RMS calculation.
More complex relays (like most of the recent generator protections) are performing a DFT calculation. In that case the DFT works as a digital filter removing up to the harmonic number (nsamples/2). Greater harmonics should be removed by the analog filter. The DC component could be removed in a digital way subtracting to all samples the value equal to the difference of two samples at 180°.
Suggestion: read the relay manual looking for any clue about what the relay is doing

 

[dpc]

I agree - this will be relay-specific. In general, most newer digital relays should include both harmonic filtering and dc offset filtering. Some relays allow you to choose rms current sensing or fundamental current sensing.

Old electro-mechanical relays (instantaneous trip units) do respond nicely to dc offset.

Regarding fault current data - it is generally given in symmetrical current - this would not include any dc offset. Asymmetrical rms current would include the dc offset.




David Castor

http://www.cvoes.com

 

[stason]

Thank you!

Another question: are instantaneous pickup currents defined in Apeak or Arms?

If an instantaneous current relay should operate with a 0,04 sec time delay, I suppose the corresponding curent pickup value is defined in Arms, since the relay has 2 whole periods (50 Hz) to measure the rms value.

Thanks!
Regards

 

[Ohmly]

This is also relay specific, but usually the inst o/c is fourier filtered to the fundamental (this also needs at least a cycle to evaluate).

Some relays give you the peak as an option if you want, it gives a faster time, but DC is then included

 

[stason]

Hi guys,

I am still new to protection system, so don't judge too hard.

A 3-phase overcurrent relay at a generator normally monitors each phase and will pick up and subsequently trip the generator even if the current at ONE phase only exceeds the threshold, right? For example due to too high current at one phase when a transformer is being energized.

Thanks in advance
Regards
stason

 

[AMBMI]

Yes, one phase current exceeding the threshold is usually enough to trip a 51 (or considering that we are protecting a generator a 51V ) element.

 

[stason]

Thank you!

a 51 is an inverse time OC element. What about the definite time OC element and instantaneous one?

 

[dpc]

What about the definite time OC element and instantaneous one

What's the question? They will trip on any one phase, two phases or three phases. Just like the 51 element.

 

[stason]

dpc,

According to wikipedia:
# 50 - Instantaneous Overcurrent Relay
# 51 - AC Inverse Time Overcurrent Relay

so I asked if the definite time OC relay would also trip due to overcurrent at one phase only.

BTW, does the Def.time OC relay have an ANSI number as well or does it share 51 with the Inverse Time OC relay?

 

[rbulsara]

51 is AC time overcurrent relay, not specific to inverse time. It includes inverse, very inverse, extremely inverse and definite time. May mfrs use some kind of a suffix like Y,D etc after 51 to describe various curves, but there are no hard and fast rules for them. Multi-function relays have blurred the lines further.

If you want to learn about relays, look up manuals of a few relays of reputed manufacturers and not Wikipedia.

Rafiq Bulsara

http://www.srengineersct.com

 

[stevenal]

How about IEEE C37.2 as a reference?

3.1.51 Device number 51—ac inverse time overcurrent relay
A device that functions when the ac input current exceeds a predetermined value, and in which the input
current and operating time are inversely related through a substantial portion of the performance range.

Note that EI, VI, STI, etc. are all types of inverse. Definite time is not.

 

[jghrist]

A definite-time overcurrent relay is now 50TD.

 

[mgtrp]

...and of course different manufacturers implement this in different ways - some seem to consider definite time as a "curve" for a 51/TOC element, others treat it as a 50/inst element with a time delay. Go figure!