SF6 Hipot Myth?

[BigJohn1]

Apparently one of our guys had some commissioning techs from a major circuit-breaker manufacturer tell him that under no circumstances should a high-potential test be performed on their SF6 interrupter bottles.

I cannot find any data to substantiate this practice nor provide any technical information of why it would be prohibited.

Is anyone familiar with any restrictions when doing insulation integrity tests on SF6? Or is this just somebody passing along bogus info? Thanks.

 

[ScottyUK]

Hadn't heard this for SF6 bottles, but it's generally accepted as being a bad idea to hi-pot vacuum interrupters because they act as a reasonably powerful X-ray source when subjected to high test voltages. It's also a borderline meaningless test because it tells you nothing about the vacuum interrupter other than distinguishing one that is totally failed from one that might (or might not) be OK.

 

[BigJohn1]

Thanks for the reply Scott.

Yes, I'm aware of the x-ray hazard with vacuum bottles, but it's low enough that just a few feet of distance mitigates the risk. We do high-pots simply as a means of determining vacuum integrity, because we have found bad bottle assemblies that way, and that has also been our motivation behind SF6 interrupter testing.

The short answer is: It's better than nothing. But if there's a good reason to avoid it on SF6 of course we'd like to know.

 

[VLFit]

Hello BigJohn1,

All breakers are factory tested and can be field tested using an AC hipot to perform an AC Withstand Test, or an over voltage hipot test, proof test, pressure test, whatever you wish to call it. An over voltage is applied for 60 seconds to evaluate the breaker contact, or whatever other substation apparatus you may be testing. The load holds the voltage or fails: simple test. I know of no reason that any breaker, including SF6 insulated models, using vacuum bottles should not be tested. Your question is a little ambiguous as you mention SF6. A breaker of this type can be tested using AC voltage, although some manufacturers declare it unnecessary for certain reasons. I am assuming you are testing the vacuum bottles only and not the bus work within the breaker insulated with SF6. As for the x-rays, it is possible to produce dangerous x-rays if the voltage is too high, however the AC levels of testing do not pose this danger if safety procedures are followed. The concern was born from the use of DC voltage (not recommended for bottle testing) for testing vacuum bottles. Some of the old DC hipots used a half wave bridge design to convert AC to DC. This results in a peak voltage 2.22 times the average voltage read on the volt meter. This would definitely produce harmful x-rays. Most modern DC hipots use a well regulated, low ripple, full wave bridge design, where the peak and rms/average/effective voltage are nearly equal. However DC shouldn't be used on bottles anyway and most bottle producers discourage its use. An AC Withstand Dielectric Test is the method to use for testing most substation apparatus. Web search the subject and you will find dozens of sources for the information you seek.

I hope this helps.

 

[DTR2011]

On USA made Alstom/Areva HV SF6 breakers, there is information in the book that explicitly states that a DC Megger test should not be performed. If I recall correctly, it has to do with the DC causing particles in the interrupter to stick to one contact or the other.

Perhaps something like that is what was stated. I've (incorrectly) heard a Megger test be called a High Pot test. I can see absolutely no reason why an AC voltage test could not be performed.

 

[BigJohn1]

I appreciate the replies:

I'm aware of the restrictions on DC hipot use, but as discussed here that has always boiled-down to concerns about inaccurate half-wave rectified units causing unknown overvoltage. We use full-wave rectified units that have good filtering and do not create this problem.

I'll be honest in saying that I've never seen anyone use an AC hipot for interrupter testing. Regardless of manufacturer's instructions, DC seems to be the standard field test method.

But to be clear my original concern wasn't reported as a restriction on DC hipoting, it was explained as a prohibition on doing *any* hipoting at all on SF6 interrupters. Maybe the factory tech was inaccurately thinking of the DC hipot restrictions?

I'll see what I can find on the Alstrom/Areva bottles. Thanks.

 

[VLFit]

BigJohn1,

Just a quick follow up regarding vacuum bottle testing. Nearly everyone (that pays attention to the standards, current technical studies, and vendor recommendations), uses AC to hipot bottles. The use of DC is unfortunately still a hangover from the days where DC was about the only thing used to test nearly everything due to the smaller size and lower cost of the equipment (compared to AC) and before the physics of different waveforms were studied and applied against their effects on materials tested. We sell hundreds of DC hipots a year from 37.5 kVdc to 600 kVdc but not for substation apparatus testing applications. The reason not to use DC isn't so much the potential for over voltage and xrays, but the fact that DC voltage can have harmful, degrading effects on some of the materials used for the bellows and other dielectric materials within the bottle. Also, what are you going to use for your criteria, in terms of mA leakage current, to determine pass/fail? There is no such thing as an acceptable partial vacuum. Either the vacuum is there or not there. An AC stress test is really the best way to determine its integrity. There are DC hipots on the market for bottle testing and produced by good companies, but their use does not conform to most bottle vendor recommendations and they are expensive instruments made for a singular purpose. It is not the most prudent purchase, especially compared to a 0 - 50 kVac hipot that can be used for bottles but also many other needs within a substation.

I hope this helps you some. Good luck with your work.

 

[Zogzog]

News to me, we do a lot of work on ABB GHK breakers which are SF6 and the maintenance procedures specifically call for hipot testing and list both AC and DC values.

Regarding vacuum breakers, all a hipot test does is tell you if you have a bottle that has already failed, we do MAC testing on every VCB we service, this allows us to determine internal pressure (Vacuum), determine a leak rate, and predict the remaining life of the bottle (Assuming conditions are the same as they have always been).

 

[VLFit]

Hello Zogzog,

Again, we should specify what we are testing when we recommend a method. In this case the requester mentioned both SF6 SG testing and vacuum bottles: two different types of apparatus each with its own suggested method of testing. With his in mind, my comments and others may be misinterpreted depending on which object is under test.

You are correct that many, and i guess ABB's, service manuals list both AC and DC voltage test levels, however, most also footnote that chart with wording something like "DC voltage testing of vacuum bottles is not recommended. DC test voltage levels are provided should only DC high voltage equipment be available." Many of the charts, and not just for bottles, will not include DC test levels for the higher voltage ratings, like >25 kV switch gear. The breaker/SG makers recognize that many still use DC so they include the numbers, but that doesn't mean that it is the technically correct test. Speaking of ABB, I was looking at some of their maintenance manuals and they recommend using the VIDAR DC output bottle tester, designed by Programma, for testing the bottles and also recommend an AC withstand test for the primary input test, the test for all the bus work within the switch. I don't get why they would not just recommend the AC hipot for both rather than subject the user to buy two different instruments. The VIDAR is only useful for bottle testing,and if you accept the DC output, making it an imprudent purchase when there are alternatives.

Your second point is incorrect. The AC withstand test on a bottle or anything else does not tell you if it already failed. It is an over voltage test that will fail the object if it cannot withstand the over voltage. If an bottle or anything else can't take at least two times normal operating voltage, the insulation material is not sound and should be replaced. The MAC test is rather new and is a diagnostic test that attempts to predict remaining bottle life. I don't know but I am sure it is an expensive tester that most will not see the justification in buying when a simple AC hipot test is all that is really needed. I know there are those out there that try to dispute the need for the AC test in their commercial attempt to market their product, but it is a little hard for most to accept someone telling them that an AC over voltage test is not a reasonable way to test a bottle, and everything else in a substation with the same AC hipot.

All these instruments and test methods have their place. None should be up to opinion as to which to use. Physics, research, and field experience should dictate which should be used where.

 

[BigJohn1]

"...Nearly everyone (that pays attention to the standards, current technical studies, and vendor recommendations), uses AC to hipot bottles...." You would have to provide some serious data to substantiate this statement. I was not being hyperbolic with my previous comment: I have quite literally never seen a single manufacturer or private commissioning company perform an AC hipot interrupter test in the field.

Further, there is nothing in any of the testing standards that I've seen that says when testing interrupters an AC withstand test would be any more effective than a DC withstand test. While many manufacturers *do* have restrictions about DC it is for the overvoltage reasons already discussed here and because of that those restrictions are usually followed with caveats like these:

"CAUTION: IF DC HIGH POTENTIAL TESTING IS REQUIRED,
THE DC HIGH POTENTIAL MACHINE MUST NOT PRODUCE
PEAK VOLTAGES EXCEEDING 50 KV."

"DC high potential testing of the bus or
equipment is NOTE RECOMMENDED, except for
the Vacuum Interrupter Integrity Test."

Those warnings have nothing to do with any inherently inferior quality of DC testing.

 

[Zogzog]

"Your second point is incorrect. The AC withstand test on a bottle or anything else does not tell you if it already failed. It is an over voltage test that will fail the object if it cannot withstand the over voltage. If an bottle or anything else can't take at least two times normal operating voltage, the insulation material is not sound and should be replaced. The MAC test is rather new and is a diagnostic test that attempts to predict remaining bottle life. I don't know but I am sure it is an expensive tester that most will not see the justification in buying when a simple AC hipot test is all that is really needed. I know there are those out there that try to dispute the need for the AC test in their commercial attempt to market their product, but it is a little hard for most to accept someone telling them that an AC over voltage test is not a reasonable way to test a bottle, and everything else in a substation with the same AC hipot."

OK, let me rephrase, AC hipot will only tell you a VI is good at the moment you test it, it could very well fail tomorrow. A VI will not fail AC Hipot until it is very, very close to end of life. If you look at Pachens curve for dry air you will see the dielectric strength (kv/cm) is very high (About 400kV/cm) from the day it is made (1.E-05 Pa) all the way to 1.E-01 where the dielectric strength rapidly decreases, an AC Hipot cannot tell you if you where you are on that curve, MAC testing can. And there is nothing new about using the penning discharge principle to determine pressure insode a VI, the manufacturers have been using this test for 30 years, the only new thing with the MAC tests is that it is a portable test set you can use in the field or in a shop as a predictive maintenance tool.

 

[VLFit]

I will be happy to expand on my comments and cite many sources for the notion that AC testing on most apparatus that is designed for AC operation and experiences AC voltages stress every second of the day, is best tested with AC voltage. I believe all can agree to that. Yes there are DC methods that are acceptable with some insulation types and some apparatus. Right now I do not have time but next week we can continue the discussion. Have a nice weekend.