Transcript
Introduction to Event Timing in Power Quality Investigations
Good afternoon, and welcome to today’s white paper webinar. Today, we’re going to be talking about event timing and power quality investigations. Typically, what we’re going to be going over is an often underutilized and unknown feature of PQ Canvass, and this is the GPS waveform overlay feature.
Network Time Protocol and Its Limitations
Traditionally, the internal timing in a power quality recorder, especially PMI’s recorders, dependent upon technology called Network Time Protocol or NTP. NTP is only really available with products that have live network connection. So NTP will reach out to a known good time source, but is still subject to network latency and jitter in general.
If you’re using the Network Time Protocol for timing, it’s very good for logging and system events, and for typically within one second accuracy, you’re going to be great. When you start looking at events where you’re trying to synchronize over sub-second accuracy is where you run into some issues.
In some power quality investigations, knowing which power quality meter witnessed or recorded a power quality event first can be extremely important in determining fault location or the direction the fault’s traveling. With NTP, if that accuracy is jumping around by a few milliseconds, think in a sixty-hertz system, a cycle is sixteen point six seven milliseconds. So if you’re off by eight milliseconds, that’s half a cycle. If your time’s bouncing around in that window, you may see that an event may look like it landed at one PQ meter before or after another one, but the reality is that’s not actually the case.
The Seeker and Seeker+ with GPS
PMI offers two products now, the Seeker and the Seeker+, both of which come equipped with a GPS module. The feature in PQ Canvass, these GPS locked and time-synchronized waveform captures don’t just work out of the box with the Seeker by itself. You have to have a GPS antenna connected, and it has to have view of the sky.
We really don’t consider a device locked until you have a speedy lock, so you got three satellites that the module’s considered locked onto. Then we can start the process internally in the firmware of synchronizing the clock and making these highly accurate timestamps that come along with triggered waveform captures.
So those are really the only requirements: PQ Canvass and a Seeker or Seeker+ with a GPS antenna connected and a view of the sky wherein it can actually receive signals from three or more GPS satellites.
Using the GPS Waveform Overlay in PQ Canvass
What I’m going to focus on here the next couple minutes is showing you how we get into that mode of PQ Canvass and how you can look at these. We’ll start with PQ Canvass. You’ll want to look at a report. Click on recording. If it’s GPS locked, you’ll know you have a GPS lock by looking at the device. The device list will show you a little green GPS icon next to it.
Once you open the recording in question, when you’ve got your triggered waveform captures, you can click in the capture list. This column here on the left, this GPS icon, they’re all green here, which means that every single one of these had a 3D GPS lock when they were captured. So we know that the timestamps are accurate timestamps.
We can click on one of these, and when we load this, it brings up our default waveform capture analysis tool, PQ Canvass. Those harmonics, phasor, table view of all the different metrics that we report, such as THD, et cetera. You can look at it as RMS trace or the symmetric components if you measure more than two phases.
Searching for Concurrent Captures
The waveform capture loads, and you get the default view. To see if there were any concurrent captures on another device, remember, you’re not looking at these within the same recording because it wouldn’t make sense to search the same recording. You’re looking at these over disparate PQ meters, different Seekers or Seeker+ at different locations.
What you do is go down to the bottom right here, this little lightning bolt with a plus icon, and it brings up the GPS search box. We have a suggested search range. You can uncheck that, and that range is in milliseconds. We did default eight milliseconds, half a cycle. Plus or minus half a cycle — if you’ve got a half-cycle phase delay for some event that’s traveling through your network, that’s a pretty significant time delay. But we figured that that’s probably a pretty good, pretty tight default.
You can make it whatever you want. We default eight milliseconds, and you’ll find that in most cases you’re not going to get many false positives. You can bring that down to one millisecond or two milliseconds if you want to tighten that up.
Reviewing Search Results
Then you hit the search button. I’m not going to do that live to you. It takes a few minutes, especially in this particular account. The more data you have, the more recordings you have, the longer that’s going to take. This particular account that we’re looking at has probably close to twelve, thirteen hundred recordings in it.
You click search, and then you’ll get this view. What happens on the right is you get a list of waveform captures that match your search criteria, and all of these just happen to be within really about one, maybe a millisecond of each other, a little less than those marks.
The accuracy that we provide on these GPS timestamps is one microsecond. Our sample rate for the Seeker is two hundred and fifty-six samples per cycle. So that’s a little over fifteen kilohertz, meaning that each sample is about sixty-five microseconds. So the timestamp is actually accurate to a little less than the resolution of an individual sample in a waveform capture.
Overlaying and Comparing Waveform Captures
As these come back in, you can click on them, and they’ll expand. It’ll tell you which serial numbers, which devices saw an event at or very near the same time. You can come over here, and you can click the checkbox, and that will overlay this waveform capture with the other.
I find when I’m looking at this, it’s better to switch over to this view which groups voltage versus current instead of aligning them voltage in channel one versus current channel one. We can split these up, say, have voltage as one set of traces and current as another set of traces.
You can look and see as I turn these on that this particular event here, the voltage is very clearly similar between the source recording and the other recording. You can see the wave shape’s almost identical. Zero crossing dominance, so they’re phase aligned, everything’s almost the same. But what you see down here is that the current is significantly different.
You’ve got the same general shape happening here in the current, so it’s all the same fault. The fault current is slightly lower magnitude in this particular capture. Then we can turn on another, and we see here it’s slightly more attenuated, so this next recorder saw the same event but at a slightly attenuated current than what the other two have seen.
Now we can turn on the fourth one, and the current was recorded. This one here, again, we’re still lowering the max current. And then finally we get another one that looks about the same.
Interpreting Results at a Substation
We know that these two — and I can’t speak in great detail about the topology of what we have here at this installation — but what I can say is that these are all from one substation. They’re all monitoring each of the individual feeders at that substation.
You can see right here immediately, you can see voltage of the line, all the voltage sagged or had a dip rather in the same point on the waveform for two consecutive cycles where this current was also flowing at a slightly different rate and slightly more attenuated as they reached each individual feeder.
Timing, it looks like they all arrived at about the same time. If you look at the zero crossing, zoom in here, you can see that they’re all within the millisecond or so.
Summary and Closing
So it’s a very useful way to look at event propagation when you have triggered waveform captures from Seeker or the Seeker+ that are distributed across your network and you’re really looking to see who’s closer to the original event — how are these propagating through transformers, army duty transformers or pieces of equipment along the network.
That covers the feature. That’s pretty much it in a nutshell. It’s pretty simple. You can start with your waveform list, and then from there you set your criteria, hit search, and you’ll get your list of available waveforms to overlay. You can turn them all on, turn them off, and compare any two together however you’d like or any three, four, however many you find.
If you have any questions, you can always reach out to us at support@powermonitors.com. I appreciate everybody tuning in for today’s White Paper Webinar, and we will see you again next week.