Traces of 2MW transformer inrush 3

[rbulsara]

Recently I came across a generator differential trip due to relatively excessive inrush of the load transformer. The issue is currently fixed by opening up the settings, and I am looking into more details.

I just recorded transients for a 2000kVA, 13.8kV/480V , cast coil transformer fed by a 2MW generator. In our case the transients last for more than 90 cycles in many instances. It appears normal inrush but I have nothing to compare to. (Also this unit is not arranged to be fed from the utiltiy and I do not have taces of energization while on a utitilty source.)

I am wondering if anyone has traces (waveform captures) of inrush of a similarly sized transformer, in order to compare them to mine?

I will be contacting some mfrs. too.

Thanks in advance.

 

[alehman]

The inrush doesn't seem abnormal. This is a common problem most likely related to CT performance.

thread238-59831
thread238-103145
thread238-89165

http://pm.geindustrial.com/FAQ/Documents/489/GET-8402.pdf

 

[rbulsara]

alehman:

Thanks, useful link to know what others have run in to. I share your opinion that inrush is OK. The CT's are not identical but are matched as best as they can (accroding to Russelectric).

The relay is GE SR489 and I am planning to implement inhibiting the 87 function during transformer enegization.
The relay did capture waveforms of one event (it happned twice in single day, no other time dozens of other starting events).

We recorded more transients with Astromed and just wanted to compare to transients of similar units.

The differential current peak is 10A at 20 cycles and 6-7A at 30 cyles, in some cases when the inrush peak is in the rane to 200A or more. Would this be normal for 'matched' CTs? I set the setting for now at 15A and 30 cycles, was at 10.5A and 20 cycles.

 

[electricpete]

One small comment (probably does not address your question)

It is normal to have a differential current during inrush. That is why many transformer differntial relays incorporate harmonic restraint.


=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[rbulsara]

Thanks pete..

 

[dpc]

I used to have a relay waveform of a new 60 MW transformer being energized - the inrush went on and on.

I don't recall that the SR489 differential has any harmonic restraint, since it's intended strictly for generator protection, so I'm not surprised it operated on the transformer inrush.

I just looked into possibility of using SR489 to protect a generator and unit transformer combination but decided against it because of this issue (among others), so I'm planning on a separate transformer differential relay. You can get gen relays that all harmonic restraint to be applied so that you can use them to look through the transformer as well as the generator. (Or you can use a transformer differential and include the generator.) But normally a generator differential can be more sensitive than a transformer differential, so very large units will use all three differential zones.

 

[rbulsara]

dpc:

Good point. Yes I will check again, but I did not see harmonic restraint in SR489. Something to remember on my next project, although I can work around this issue.
I should clarify that the diff zone does not include the transformer, but it does see the inrush.

Apparently GE knows this issue hence they recommend inhibiting the function during switching. What is furstratig is that they do not have a imple way (say a logical function)of blocking the function. They recommend a most cumbersome way you can think of in one of their publicaiton, mentioned in links posted by alehman. I plan to implement that however.

 

[RalphChristie]

rbulsara:

I am not familiar with the relay you are using and I do not have any waveforms, but this inrush (from the values you are providing, and if I read it correctly) looks normal to me. Just a few comments...

When a transformer is first energized, a transient magnetizing or exciting inrush current will flow. This inrush current appears as an internal fault to the diff relays, and may reach instantaneous peaks of 8 to 30 times those for full load. The factors controlling the duration and magnitude of the inrush are:

Size/location of transformer bank.
Size of power system.
Resistance in power system from source to the transformer bank.
Type of iron used in the transformer core and its saturation density.
Prior history, or residual flux level, of the bank.
How the bank is energized.
Instant of switching.

The resistance in power system from source to the transformer bank determines the damping of the current wave. Banks near a generator will have a longer inrush because the resistance is low. (L/R) Also, larger transformer units tend to have a long inrush as they represent a large L relative to the system resistance. At remote subs, the inrush will be not so severe, since the resistance in the connecting line will damp the current. Inrush tends to be bigger if the transformer is energized from the LV-side. Theoretical the max. phase-current inrush occurs at the 0° closing angle, (O° voltage) but maximum inrush will not occur on every energization - the probability of energizing at the worst condition is relatively low.


There are basically three types of inrush:
Initial inrush (Energizing a bank - most severe)
Recovery inrush (An inrush after a fault external to the bank is cleared and the voltage returns to normal)
Sympathetic inrush (When a bank is paralleled with a second energized bank, the energized bank can experience a sympathetic inrush)


There are two aspects of significance for inrush currents affecting differential relaying:
Current flows only in one winding of the transformer (winding being energized) and therefore appears to the protection as an internal fault.
The inrush current differs from an internal fault current insofar as its wave form comprises a high percentage of harmonics. Of these, the second harmonic has a particular prominence under all switching in conditions.


Hope it is helpful

Regards

 

[RalphChristie]

Sorry, I've seen now the transformer is not included in the scheme. Is the generator diff tripping when you are energizing the transformer?

 

[stevenal]

Energization of a downstream transformer should look like through current to a generator differential relay regardless of the harmonic content. Harmonic restraint would not help. If energization looks like differential current, what will through fault current look like? Maybe time to check CT ratios, polarity, burden, etc. Maybe excitation testing and demagnetization. I had a similar problem with a transformer differential tripping when downstream loads were connected. The fix was to replace the relays.

 

[dpc]

rbulsara,

Sorry - I assumed the transformer was in the gen diff zone.

Now I recall the previous threads on this issue with the SR-489.

 

[alehman]

Another GE article on the subject:

http://www.geindustrial.com/products/support/489/get8403.pdf

 

[mc5w]

If this "inrush" is happening well after the transformer is energized, you could have a problem with some brief ground faults on the secondary that clear themselves before a circuit breaker opens. That is, you could have some type of corona discharge somewhere or something just getting wet.

Lots of luck finding this before the intermittent fault turns inot something major.

 

[rbulsara]

mc5w:

Its just plain old inrush, and there is no fault in the system. I wanted to compare my captured data to those captured by others for a similar installation.

And also wanted to know that inrsuh lasting for several cycles (90 plus) is not uncommon and the if seeing some differentical duing switching is expected, before we tear apart the system for no reason.

Reply by electricpete, dpc and others have helped affirm that.

I really appreciate everyone's input, opinions and sincere desire to be helpful.

Thank you.