Metering equipment for capturing sub cycle transients 3

[rockman7892]

I had posted a little while ago that we are experiencing several issues in our facility that we believe are a result of voltage transiens originating either from outside our facility or coming from within our facility.

In an effort to analyze some of these transients I want to try to capture them on some sort of power metering device. I believe some or most of these transients are sub cycle transients lasting less than 1/2 cycle in duration in which case most relays or metering equipment are not quick enough to capture these transients.

Does anyone know of a good power meter for capturing these sub-cycle transients? I would want something that could provide an osilography of the transient as well a indicate magnitude and durations. I would want something that I could permenantly install and put on our SCADA network for viewing present and past events.

Does anyone have any experience with a particular unit?

Does anyone know if the Multilin 750 relay, or Multilin PQMII Power monitor is capable of this?

 

[jghrist]

Did anything noticable happen at the time of this transient? I suspect that this is just a measurement anomoly, and that there was no transient in the actual current and voltage. I would not expect such a clean jump in currents and voltages. If the current really changed so fast, there would be ringing instead of a clean jump to the new value.

 

[davidbeach]

As far as I can tell, I_a lags V_a by about 90°. If V_a is really V_ab it should be so labeled. V_ab leads V_a by 30°, so if it is V_ab, then V_a leads I_a by about 60°, still far from a resistive load.

But in any case, I know of nothing that can happen downstream and cause that discontinuity in the voltage wave.

 

[rockman7892]

I was able to capture (5) more transient events over the weekend. The strange part is as I mentioned all of these PQM's are on the main and feeder breaker looking at the same PT's on the main switchgear lineup, however all (5) of the attached events seem to be different. I would think that they would all see the same event since they are all looking at the same PT? Relays are set so they must be manuall reset after a trigger.

I have attached the (5) triggered waveforms, along with the trigger time and date proceeding each waveform.

It appears that three of these events seem to have the phase shifted transients as the first one I posted, while the other two events (4/24/10 @ 06:11:48 and 4/24/10 @ 10:53:14) only seem to have a transient on the Va waveform. After seeing this, do you still thinks its possible that the phase shifted transients are a result of a source transfer since it was mentioned above that this was very uncommon? Could the two non-phaseshifted transients be a result of capacitor switching or some other event?

Do these transients look like ones that could trip VFD's on bus overvoltage had we not gone ahead and put line reactors on all of our drives?

The transient on 4/25/10 @ 17:31:10 looks interesting for it appears to have alot of current distortion associated with the current waveforms.

I was able to figure out how to display the Vc waveform and was also able to determine that these Va, Vb, and Vc waveforms are indeed L-L voltages and this can be verfied by the peak of each waveform being close to 5.8kV. Multilin has told me that becuase I am using an Open Delta 2-PT arrangement I am not able so see a Vb waveform, however I questioned this and are waiting for a response from them. Elsewhere in the relay it gives me Vab and Vbc magnitudes but the relay does not give me a Vbc phasor in the phasor metering. I'm not sure whats going on here.

I tried to use the cursers as best I could and came up with close to a 70deg phase shift between the voltages and currents that davidbeach was questioning. And since we now know that voltages are L-L so we need to subtract the 30deg which leaves us with about 40deg and this matches close to the p.f. of about .75 which I was seeing on the relay. Therefore waveforms seem to be correct.

Thanks for the help.

 

[dpc]

Is there something happening in the real world at the same time as these events that is a concern?

David Castor

http://www.cvoes.com

 

[rockman7892]

dpc

No there is not currently something happening here at our facility at the time of these events I've captured. We have had over the past few months several issues with drives tripping out (some of which I've posted on this site) and power supplies failing. Almost like clockwork we would have drives trip while sitting idle overnight and would have to reset them all every morning. I was never able to see what exactly was causing these drives to trip so just assumed it was some sort of transient and installed line reactors on all the drives and havent had a problem since. This whole time I've assumed it was something coming from outside the facility, and put is on my radar to get ahold of a meter that could help me verify this.

I finally got a chance to get back to this issue and after looking at the PQM found that I could possibly capture some of these transients. After seeing some of these (again none caused any issues currently) I became curious as to what I was seeing and thought I would seek the advice of the experts here on this site. These transients look kind of strange to me and I'm wondering if these should be something I should be concerned with and contact the utility, of if these are just natural occurances? I find it hard to see how the phase shifted transient would be a natural occurance?

Thanks for all the help with this.

 

[rbulsara]

Without any other first hand information, my view remains the same as before. I believe that the voltage anomaly is creating the current anomaly. So find the cause of the voltage anomaly.

Rafiq Bulsara

http://www.srengineersct.com

 

[jghrist]

The 25/04/2010 17:31, 24/04/2010 5:32, and 24/04/2010 5:31 events have the phase shifts shown in the initial event. These look to me like there are missing samples in the data. It's like all of a sudden, the trace skipped about 300 degrees from one cycle. There is no ringing (high frequency oscillations) in the current when this occurs, which would occur if the current really did change rapidly. I think it's a measurement problem. I suggest sending the results to the PQM manufacturer to get their comments.

The other two events appear to be voltage spikes on Va.

 

[rockman7892]

I've attached (2) more transients that I've recorded over the past two days. I've recorded others but they were more of the phase shift type that appears to be a measurement error of some sort. I am going to ask Multilin for their thoughts on this phases shifted measurement error.

As far as the transient that appers on the Va most of the time, does this appear to be a real transient? Can you tell if this is something that is coming from outside our facility?

Do the peaks of 9kV during these transients indicate that this may be something that could have been causing a disturbance to our drives before we put the line reactors on. (I'm almost half tempted to take the reactor off this drive temporarily to see.)

jghrist

Can you briefly explain why there would be ringing associted with current changing rapidly? Is this ringing only associated when current changes rapidly, or would it occur when the voltage changed rapidly as well?

 

[rockman7892]

The other thing that I find strange with these transients is the fact that all of these PQM's are looking at the same PT inside the switchgear, however not every device records the same transient. There may be one transient that shows up on one device and none that show up on the rest. And it doesn't appear that they are even at all triggered from the same transient due to the fact they all have different trigger times.

 

[jghrist]

All all of the trigger points the same? What are the trigger values? If the voltage spike is close to the trigger value, I could understand how one PQM would trigger and others wouldn't, but if the voltage is considerably higher than the trigger value, all of the PQMs connected to the same VT should trigger.

 

[rockman7892]

jghirst

Yes all trigger points are the same set to trigger at 110% overvoltage.

Some of the transients we have seen are peak at about 9kV which is well over the peak of 110% of L-L voltage (5.88kV)

 

[jghrist]

Can you briefly explain why there would be ringing associted with current changing rapidly? Is this ringing only associated when current changes rapidly, or would it occur when the voltage changed rapidly as well?

It is caused by inductance in the circuit which prevent instantaneous changes in the current. It could occur with a rapid change in voltage if there were a lot of capacitance.

Some of the transients we have seen are peak at about 9kV which is well over the peak of 110% of L-L voltage (5.88kV)

Sounds like something else to ask Multilin about.