Transformer Winding Resistance Measurement: Field Challenges
Winding Resistance Measurement: Field Challenges Q&A
This is the transcript of the Q&A session from our webinar Transformer Winding Resistance Measurement: Field Challenges held on 3/18/2022. If you would like to watch the presentation leading up to the session below please click here.
What could we do when we have a deviation between phases above 5% and how can we be sure of the type of defect?
"So, if in case your readings are more than 5%, I would start off by making sure that I am comparing apples to apples. Those are temperature correction readings. The second thing that I would do is I would re-perform the test because I always blame the user before I start pointing toward the transformer.
So, let's see the reading set temperature corrected. I then the test again, my readings are still above 5%. If that is the case, I want to validate that with another electrical test. And one way to do that is that if you're winding the distance shows higher, deviation, I would start with the excitation current test and make sure that my excitation currents are kind of trending in the same direction with the historical measurement.
The other thing I would do is I would perform the SFRA measurement because if there is something wrong with the winding, I would be able to figure that out in the frequency range from 20 kilohertz to 400 kilohertz. So I would perform these two tests to see what's going on. The other third test that I would look into is our leakage reactance test.
That gives me the idea about the core and winding kind of thing. So those were the three tests that I would look into if my winding resistance is above 5% between the phases." - Dinesh Chhajer
What is the best current to use to test winding resistance from 10 amps to 50 amps? And what about the tap changer rating and using 50 amps?
"I will say briefly that you don’t want to just blindly choose any current. What you want to do is look at the nameplate rating of the particular transformer. And With regards to the tap changer, the tap changer is going to need to be, in the worst case, the same rating as the transformer. If you think about it from a logical standpoint. If a winding is rated for 1000 amps and my tap changer is rated for 50 amps, I'm going to have a serious problem.
So that addresses the second part of the question. But for the first part, look at the rating of the winding, You want to stay within 1 to 15%. And I will also add that many common instruments, once you input the nameplate information, they will give you a recommended test current, which is usually going to be somewhere around 1 to 10% of the rated time." - Damien Robinson
Is there a level of combustible gases that DC winding resistance testing should not be tested at?
"So obviously, performing DGA testing is very important to determine the level of combustible gases that you have in your transformer. And if that exceeds a certain threshold, then obviously that needs to be addressed. But in terms of a winding resistance test, those gases would not necessarily affect that test or would not cause any issues at that point, because you're injecting current on the windings.
So you really shouldn't have any danger in that sense. But obviously, we want to be aware of those combustible gases that are present within our transformer so that way we can take corrective, actions if that's the case." - Charles Nybeck
How much of a difference in measurement do you have with a loose connection or carbon contamination?
"It's a very interesting question, and actually it's one that's hard to give a direct answer. I mean, there's a couple of things that that play into that. How much carbonization are we talking? Are we talking about the beginning stages or is it severe? Are we talking about a loose connection that say, for example, some sort of stranded conductor and it's only, one piece of that conductor connected? or are we talking about how severe something is torqued?
There's a lot of different elements that go into being able to quantify that. The other thing that that complicates it a little bit, too, is how much current you're using.
So, for example, if it's a loose connection and I'm only using, say, 1 amp versus 50 amps, the 50 amps is going to going to show up a lot more because it's going to heat that connection up. So unfortunately, that's a tough one to answer. But you will see a difference in the reading if you have those loose connection or carbonation. And it's going to depend on the severity of the situation." - Damien Robinson
Is it best to select the current level or auto ranging?
"Say you want to be in the range of 1 to 10% If the test set allows, I would go as high as up to 10%. So lets say I have a test set that can give me 50 amps of test current and my rated current is let's say 200 amp of that winding. 10% of that is 20. I want to start by 20. So, I will pick a certain level of current if the test set can provide that. So that's the first part, to select the current level.
Now the nice thing that works is that let's say you said that I want to inject 20 amps, but the transformer has very high resistance for whatever reason. So what the test instrument does is that even though you asked for 20 amps, it would try to strive for 20 amps. But it may happen that because of the high resistance, it might settle at 15 only amps.
So it is doing auto-ranging after you've selected a current level. So always start with 10% if you can. And then once you select that current, the test instrument would try to get to that level, but it would auto range based upon whether it can do that or not." - Dinesh Chhajer
We use a low resistance for meter for wanting resistance test measurements. It takes a lot of time to stabilize which test instrument is better?
"In my tenure with Megger, I have received this question less and less as the market becomes more educated that the traditional winding resistance test is the better way to go.
A low resistance ohm meter can have the current to push with. But one of the problems is going to be the voltage it outputs. I looked up the specifications on one of the popular units we sell and the compliance voltage or the voltage it is able to push is 20 million volts. And we've talked about, 50 volts as an example of some of our winding resistance test. So that should help to answer that question of why it's taken so long - You're not able to overcome the inductance of the winding in order to get that stable reading so that that answers that question.
Now there's a couple other important parts too, as why a low resistance ohmmeter should not be used and that is the discharge circuit. Once you have charged up that winding, you need something to properly discharge it and low resistance ohmmeters do not have that. And then lastly is the de-mag capability. Once you have finished all of your winding resistance testing, you need to de-mag the transformer and low resistance ohmmeters do not add that functionality.
So the long story short is that's not the right instrument to be using. You need to use a dedicated winding resistance test set." - Damien Robinson
In the presentation, “exceeding the saturation knee” is mentioned. Could you explain this more?
"The way we define the knee point is - the point beyond which the relation between voltage and current is not linear. That's kind of the definition for the knee point. And if the flux in the core is beyond that point, then we see that the core is saturated. So when you perform the winding resistance, the goal is that I want to overcome the inductance of the circuit. I shared one circuit (in the webinar), which is the equivalent circuit of a transformer, and it had a leakage reactance, component, X1 and x2. So what happens is that once you apply enough voltage for enough amount of time, you develop enough flux that now the cores the domains are aligned. What it does is that it reduces the effect of inductance, which is the resistance to the flow of the current, it neutralizes that.
Once you neutralize the effect of Inductance, then it is much easier for you to push the current and stabilize the current. And once you have stabilized the current, that's when the LDA by DT goes away. And if I don't have the LDA over DT, which is the rate of change of current, then it is obviously now easy for me to do V over I. So I want to set up enough flux which would take it to the saturation and then I can get rid of LDA by DT and get my resistance measurement. " - Dinesh Chhajer
What is the purpose of correcting to a reference temperature?
"It's a good question. And one thing to remember about resistance is that it's temperature dependent. So whether we're talking contact resistance or winding resistance or we're talking insulation resistance, it doesn't matter, resistance is going to be temperature dependent in an object.
Now when we talk about conductors, which is what we're talking about with the transformer winding, the higher the temperature, the higher the resistance - the lower the temperature, the lower the resistance. So, one thing we can't control in the factory, you can pretty much control the temperature of the winding. I could theoretically not have to worry about adjusting it to a certain temperature.
But when I'm out in the field, I have no control over the asset temperature when I'm there. The idea of temperature correction or doing a reference temperature is that no matter what temperature, I take the measurements at I have a reference to go back to for comparison purposes." - Damien Robinson
Which transformer current should be used as a reference, either base or force cooled?
"This is a good piggyback off what Damian just mentioned. As he mentioned, resistance is temperature dependent. When you're performing a winding resistance test, that transformer will have been taken offline and isolated. And what typically happens is that sets and comes to an ambient temperature at that point.
So, there will be no force cooling going on. We would want to utilize that base as a reference for our current measurement or for our current application. It kind of goes along with exactly what Damien said in the sense that it's temperature dependent, so we really want that set at an ambient temperature and then perform the measurement, where we would reference the base." - Charles Nybeck