Primary injection testing on 800 amp breakers? 3

[WellWhatHadHappenedWas]

Does anyone do primary injection testing on large breakers, say around 800A? We have a test set that can inject 2000 amps but I would think that at that size of breaker we would need 3 to 5 times rated current to get it to trip quickly. I believe in the past for breakers around this size we have just tested the CT's in the breaker for accuracy and then tripped the breaker with current on the secondary. I was just curious if anybody else in industry does primary injection testing and if you do what test set are you using?

 

[oldfieldguy]

Your test set is likely inadequate for tests on 800-amp breakers.

NETA specifies a time curve test at 300%. That 2400 amps. YOu can't get there with your test set. Typical instantaneous currents for your breaker might require pulses of 4000-8000 amps or more. Again, nope.

When I was in the biz of testing power components, the test set we lugged around for testing breakers in that range could run 20 kA continuous and much more on short periods.

If you're in the USA you can contact any of several companies that do this testing and they will have the equipment and personnel to help you.


old field guy

 

[mparenteau]

If they are molded case, I would say there isn't much value in primary injection testing. Different conversation if we're talking about low voltage power circuit breakers.

There is a paper on the internet, where a company used thermography as its testing basis and they came up with good results on impending failures.

This topic has as many opinions as there is "sand on a beach" (ok...that's a stretch).

Mike


PS In case anyone is interested in the paper I referred to:

http://web.eaton.com/content/dam/ea...tion-maintenance-of-molded-case-breakers.pdf.

 

[waross]

You can test the thermal trips with time current tests within the capability of the set for troubleshooting.
That is if you suspect that your breaker is tripping on less than rated Amps on the thermal trips.
It is not acceptable for direct injection acceptance testing.

[link ]Bill[/url]
--------------------
"Why not the best?"
Jimmy Carter

 

[dpc]

If they are molded case, I would say there isn't much value in primary injection testing

Based on what and what size range are you thinking about?

To OP: Yes many facilities do primary injection testing for larger low voltage breakers. As oidfieldguy noted, your test set is just not big enough. Some of these big sets require a 480 V 3-phase power supply.

 

[jraef]

The issue with doing primary injection testing of MCCBs is that they are not really designed to be tripped repeatedly and maintain their reliability. Power breakers are.


" We are all here on earth to help others; what on earth the others are here for I don't know." -- W. H. Auden

 

[dpc]

jraef - if you're talking about testing at the maximum SC rating of the breaker, then I agree. But most current injection testing is not done at the breaker's maximum SC rating. The big issue with MCCB testing is that it is a PTTA. Drawout breakers are much easier to test.

 

[mparenteau]

I found an updated paper which is pretty good which talks about primary injection testing of molded case breakers:

https://inldigitallibrary.inl.gov/sites/sti/sti/Sort_9994.pdf.

Mike

 

[dpc]

Thanks for the link. I know of one major semiconductor company that does primary injection testing on every new molded case breaker - in the thermal range, IIRC. They experience about a 2% failure rate on brand new breakers. I agree exercising the breakers is important and "routine" injection testing is probably never practical except for larger breakers. But the testing recommendations are all over the place and don't really take into account the impact of arc flash incident energy levels. The majority of arcing faults will occur downstream of a molded case breaker. The arc flash calculations are based on that breaker tripping per the manufacturer's published time-current curve regardless of age. Before we worried about arc flash, if a MCCB failed to open, generally something upstream would eventually trip and that was seen as not a big deal. When arc flash is figured in, the reliability of MCCBs is much more important (at 480 and 600 V).

Also, I've been in many manufacturing facilities where MCCBs were used for everything. They weren't willing to spend any many to put in low voltage switchgear or they just didn't know any better. It's common to see 3000 A and 4000 A MCCBs. It seems that we would want to know if these devices were actually functional.

 

[rcw retired EE]

When I was in the breaker testing business we had 10 kA, 20ka, and 50 kA "portable" test sets. With good bus bars from the set to the breaker we could test 2,000A breakers in place. Pushing 6ka for the 20 minutes on each phase was a pain. We didn't find that many problems except loose joints which a good thermal scan would pick up. Draw out circuit breakers were easier to test. We seemed to find more problems with them. At one aerospace facility we failed 90% of the 400-2000 amp breakers during a holiday shutdown. Trip unit's operated correctly but the trip pin never hit the trip bar hard enough to trip the breaker.

My preference is to push primary current to verify that the CT's or other current sensor is working and to measure voltage drop at rated current to check for loose connections. Use a test kit or secondary injection for checking the trip unit. If it has ground fault protection, checking polarity of the neutral CT is a lot easier with a primary injection set. Ground fault testing doesn't require as high a current.

 

[unclebob]

Three reasons why we still do primary injection:

1- Retrofit of trip unit in a power circuit breaker: check the tripping times while checking the secondary circuit of the new ct's to the T/U
2- Check the tripping times of old dashpot protection, again in power circuit breakers, but we slowly retrofitting them with electronic T/U
3- Run into issues where the breaker trips at a lower current , most of the times at the ground fault pick up, when one phase CT is mis-wired or defective. But I've encountered this 3 times in 15 years.

Our breaker tester bench is ''movable'', but come in 3 pieces: 600/120 V single phase transformer, CT module and a auto-transformer. Max amp you can get is about 12 kA, but the thing is whining like hell...

 

[dpc]

2- Check the tripping times of old dashpot protection, again in power circuit breakers, but we slowly retrofitting them with electronic T/U

I think I'd just tell anyone with an old dashpot unit that it failed the test and needs to be replaced. Those things were horrible when they were new.

 

[che12345]

Dear Mr.WellWhatHadHappenedWas

Q1. Does anyone do primary injection testing on large breakers, say around 800A?
A1. Yes, very often; in the workshop as well as at site.
Q2. We have a test set that can inject 2000 amps but I would think that at that size of breaker we would need 3 to 5 times rated current to get it to trip quickly.
A2. Yes, a 3-5x 800A say 4kA (short-time rated say 10min ) would be needed for "short-time" tripping test.
FYI a) Some (not all) 2kA test sets can be paralleled. The out-put would be 2x 2kA= 4kA. Short-time rating say 10 min. Caution: observe the duty cycle and the cool-down duration stipulated by the manufacturer. Allow the cooling fan to run continuously during the " idle period ", when not injecting.
b) If you do not carry out these test frequently, out-source to an external contractor would be cost effective.
Q3. I believe in the past for breakers around this size we have just tested the CT's in the breaker for accuracy and then tripped the breaker with current on the secondary. I was just curious if anybody else in industry does primary injection testing and if you do what test set are you using?
A3.1 "...just tested the CT's in the breaker for accuracy and then tripped the breaker with current on the secondary..." This practice is a compromise, which should not be taken as a norm.
A3.2 "...what test set are you using?" Carry out a Google search, there are numerous manufacturers, depending on your location. See also above A2. b).
Che Kuan Yau (Singapore)

 

[WellWhatHadHappenedWas]

Thank you all for your input I really appreciate all the help. I'll look into this further to determine if this is testing we'll need to outsource or if we'll need to order some more test equipment. Also just a heads up to mparenteau (Mike), I'm stealing the saying "This topic has as many opinions as there is "sand on a beach"! I don't know how I've never heard this before! And thanks for the links to the white papers; I love when I can give an answer with a reference.