Transcript
Introduction to Power Factor Correction and Voltage Regulation
Today I’m here with David Keaton, and we’re going to talk about power factor correction, PFCs in utility distribution networks and how they relate to voltage regulation. As David will talk about, ostensibly, PFCs are used to correct power factor, to cancel out the reactive current from inductive loads.
Since most loads are traditionally inductive from motors and transformers, these capacitors cancel out that reactive power flow and make the power system more efficient so you’re not losing power with reactive current. But in reality, capacitors are more often used to prevent voltage regulation because indirectly, they change the voltage drop across the line.
So if you are changing your power factor, whether it’s for better or for worse, you also end up changing the voltage the customers see, and that is generally more important than the efficiency. Utilities value voltage being in the right level and voltage regulation is more important than efficiency.
These capacitors often are used for just pure voltage regulation just to make sure voltage is within limits, and then assuming that’s met, then utilities will start thinking about actual power factor and efficiency. So I’ll turn this over to David to describe the paper, and he’ll show you also how to see this data in PQ Canvas.
Capacitor Bank Switching in Recordings
With power factor correction, capacitor banks are often used, and they’re switched depending on time of day when inductive loads can be high or when certain factors such as voltage drop below a certain level. One of the ways that you can easily see that is in the recording here.
You’ll notice that power factor correction increases in steps and so does voltage. Reactive power itself will also decrease. So here we have an instance of that occurring. The reactive power goes from about 25 kilovars down to 15 and 10, and the voltage itself is also increasing in steps.
Oscillatory Ringing and Harmonic Resonances
One of the other things that occurs when power factor correction banks turn on is an oscillatory ringing that can occur from the sudden inrush. We can see in this waveform here. See right here, the ringing occurs.
Other things that also happen is harmonic resonances can shift resonant frequency which we can see in here, primarily in the fifth, seventh, and eleventh harmonic on channel one. And as the ringing occurs, that resonant frequency shifts over to the tenth, eleventh, and twelfth harmonics. So something else to be aware of when that happens.
In general, your resonant frequency of a system will change when you change the capacitance. The resonant frequency before the cap bank switch and the resonant frequency after will change. And these cap bank energizations are typically disturbing the system. And you’ll often see a step change in voltage either before or after.
Voltage Step Change Example
If you look at the RMS version of this waveform, for example, the voltage before, if we click on the live, and now we see the… So, phase A, voltage before is, say, 495.1, and the voltage after, 499.4. So we have a step increase in voltage due to this cap bank energization. So this has not only changed the amount of reactive current flow, but also changed the voltage that everyone’s seeing downstream.
Using Merlin in PQ Canvas
One of the other things you can do with our PQ Canvas software is look at Merlin, which will also point out several of these capacitor bank switching and oscillatory ringing transients. So here, we’ve already run an analysis on the recording itself.
It’ll point out several things such as voltage sags that happen from those inrush currents, inductive loads, and it also points out some of the waveforms that have some stereotypical occurrences that happen with capacitor bank switching. You can also ask it questions in our new Merlin chat. Here, I’ve asked where the bank switching occurs most commonly and where the greatest changes in steady state voltage occur.
How Merlin Identifies Cap Bank Switching
Merlin can flag these cap switches. It’s looking for those signs we just saw. You have an oscillatory transient ringing, and you have a step change in voltage from before the event to after the event, whether it’s a step-up or a step-down, with no step change in current.
If it’s a load-induced step voltage change, it recognizes the load is causing all the drop. But if the current doesn’t change appreciably, then that’s more likely a cap bank energization or de-energization. And so Merlin will separate those out, and here it’s showing you the overall, the entire recording, how often did that happen, and what’s going on here. So Merlin pulls that out of the waveform capture data, and will tell you about it here.
And here it actually notes the part of the strip chart graph that we were looking at originally, where they were switching around the 14th of October at around 10:13 AM. Also notes some transients, oscillatory current spikes. There’s other parts of the analysis as well. Here’s the capture that we were looking at originally. See with that rating, you can see that Merlin actually picked up on it and noted about it.
Detecting Cap Bank Events in PQ Recordings
These oscillatory transient waveform captures are really the key to spotting these in a PQ recording. When cap banks operate and the voltage level changes slightly, that’s really tough to trigger on in a waveform capture, or detect in the data because those are often a volt or less, steady state changes. And so, the cap bank ring is really the sentinel that lets you figure that out.
Another way of looking at it is looking at the reactive power, but there you have to be measuring, if you’re on the primary side, measuring primary side current. So, if you’re at a customer, a secondary, or a secondary with tap out transformer, or meter base, the current you’re measuring isn’t showing a reactive current flowing through the primary side feeder. So, you don’t have that reactive power step change that you would normally see if you were, say, measuring at the capacitor or at the substation.
So, if you’re at the substation or upstream, so your current actually has that feeder current, you can see step changes in that reactive power. If it’s a 600 kvar cap, you’ll see a step change of reactive power that matches that. But if you’re measuring downstream, you won’t see that. But, if the customer has their own power factor correction caps on their side of the meter, then you will see that.
Cap Banks, Voltage Regulation, and Efficiency
The key here in the paper is that cap banks are, ostensibly used, as the name implies, power factor correction. Most loads are traditionally inductive. That’s not quite as true as it used to be, but everyone has transformers, so that does add some inductance to the system, and the substation transformer is a large inductor.
These capacitors are meant to undo that, cancel out that inductance. The capacitor has a reactance that’s equal, or hopefully equal in magnitude, but opposite in sign from the inductive reactance, and the net effect is a pure resistive system. And in that situation you have the lowest current flow possible, and that reduces the voltage drop to the lowest possible across the distribution conductors.
But, as we have been saying, more often the utilities focus on basic steady state voltage regulation. Are they within ANSI C84.1, range A ideally, and that’s more important than efficiency. First, it is reliability and voltage being within the right level, and so capacitors are more often kind of pressed into service to make sure the voltage is within the limits. If it is, then you can secondarily try to optimize power factor.
Now that said, utilities are also focused on large customers to make sure their power factor is within limits themselves, so that reduces the burden on the system. And in general, it’s more efficient to place the capacitors near the source of inductance. If you have a customer with a giant induction motor, it’s overall most efficient to put the capacitor right where the motor is, and that way the capacitor is switched with the motor, and you don’t have over-correction. So, that’s another reason to focus on customer power factor correction capacitors.
But these capacitors inevitably bring resonances into the system, which can create these transients that can bring their own problems. They can magnify harmonic distortion. So, there’s a whole host of issues that come along with capacitors, and you’ll find more in other PMI white papers.
Question: How Is Resonance Related to This?
In general, if you have an inductance and a capacitance in a circuit, there will be a resonant frequency. So, if you put a power factor correction capacitor on your system, you’re going to have a capacitor, and you always have inductance, if nothing else, the substation transformer is an inductance. So, you have an inductor and a capacitor, so therefore you have a resonant frequency.
I’m going to bring up an entire one-hour presentation I have on resonances, but just show you a couple slides to help understand this.
Estimating Resonant Frequency
Here is a very useful formula for estimating the resonant frequency of a system of a distribution. If you know the capacitance, there is a very useful formula to estimate the resonant frequency. The classic resonant frequency is one over 2πfc, for the reactance of the capacitor. The resonance is where the reactance of the inductor is equal and opposite to the reactance of the capacitor.
Here’s the formula for resonant frequency. One formula that is very handy is the short-cut formula. If you know the flow rate in the capacitor, and you know the short circuit current at the capacitor, this formula gives you an approximation for the resonant frequency in terms of harmonic number.
So if, for example, we have a 200 MVA short circuit current, we have a 4.2 Mvar capacitor bank, that gives us a resonant frequency of 414 Hertz, or seventh harmonic.
Impedance Versus Frequency With and Without Capacitors
What that means is that, at that frequency, the impedance to your system is very high. If you have no capacitors, your impedance versus frequency, the impedance looking back into your network from a customer load, is a straight line whose slope depends on the inductance. At 60 Hertz the impedance is low. At high frequencies, the impedance is high, because that’s what an inductor does, and your substation transformer is that inductor.
With a capacitor in your system, it’s much different. This blue curve is what happens when you have a capacitor in there, you have a resonant frequency. Low frequencies, the substation inductor is low impedance, like a short circuit. At high frequencies, the capacitor is basically a short circuit. And at the resonant frequency, you have a very high impedance, or a high resistance, which means that harmonic currents at that frequency will cause a lot of voltage drop at that frequency.
That causes high harmonic distortion if you have customer current at the same frequency as that resonant frequency. And if you don’t, if the customer currents are different frequencies, not a problem, but if that resonant frequency happens to be on a common harmonic, like the third harmonic, or the fifth, or the seventh, that’s a combination for bad problems.
Your voltage distortion will be much higher than it would have been, because the resistance to the network is much, much higher than you would normally have just from the wire itself. And changing the capacitor size moves that resonant frequency around. A larger cap makes the resonant frequency lower, a smaller cap makes it bigger. And of course, as the capacitance is switching in and out, that resonance is changing, or coming and going, as the capacitor switches.
Closing Remarks
That’s kind of a sneak peek at this resonance class. You can find the schedules on our website. If you want to hear much more about resonances, feel free to sign up. And if anyone else has any more questions, feel free to type them here in the questions box.
Well, that’s all the questions we have. Again, thanks for attending, everyone, and everyone have a great afternoon.
