LV BREAKER TEST PROBLEM - PRIMARY INJECTION

[EMT01]

We have recently tested numberous Square D Mastpact NW & NT circuit breakers and are having a 50+% failure rate, when performing primary current injection tests. All other tests are OK (insulation resistance, contact resistance, Trip Unit pick-up). We are finding that the breakers, in question, may test okay on one or more tests and then fail on any given function. We then re-test and it may pass or it may not.

This is what we experienced this week on a single breaker... We had one test bad on INST.(day 1); then on day 2 (in the presence of the customer) it failed on Long-Time and passed on all other tests; day 3 (today) in the presence of a forensic engineer it passed on all tests. We then tested four additional breakers and all had some sort of failure on LT, ST, INST, or GFP. These were intermittant failures that we could not get to repeat.

The bottom line is that the breakers are inconsistent in their failures and these failures are seldom repeatable. Square D is not acknowledging a problem, but one another project (a few months ago) a Square D service technician used (8) different trip units in order to get (3) breakers functioning without any problem.

If someone would like additional information about our testing process, test results, other experience, etc. please let me know. Does anyone else have similar experience with these breakers? Any help would be appreciated.

 

[dpmac]

EMT01:

Are you saying that you find Pickup to be OK then do a LTD timing test and it may be up to 4X higher (longer). This happens on other functions too. All timing is longer then mfg curves by a long shot.

After doing the tests a few more times the "timing" gets shorter, or falls closer to the mfg curves. I assume this applies to all functions. LT, ST, GF (and Inst I guess).
Does pick up ever vary like time? What happens when you simply run 30% of RP (In) for 60 seconds between tests. Does it change results in timing?

Have you ever continued to perform the tests until the breaker tripped in too little time? Will it continue to get shorter and shorter until its out of spec?


I will check the Square D part number of our test set in the AM.

 

[alehman]

I wonder if the "thermal imaging" function is causing the decreasing trip times. If you disconnect and then reconnect the secondary test set, does it return to the long time delay?

Alan
----
"It’s always fun to do the impossible." - Walt Disney

 

[trosepe]

EMT01,

In the case I had mentioned, they did not have a secondary injection test set. They tried testing through two poles (as you can do with Versa Trip) but the results were inconsistent' changing from test to test as with yours. The test set up that worked was to common all three load side stabs together with bus. For testing 'A' - bus line side 'A'to one side of test set; then bus line side 'B' and 'C' together and to other side of test set. They set ground fault to max during these tests so the unbalance would not be a factor. After the Contact Resistance, LT & ST tests were performed they performed ground fault on each phase individually. They said out of 30 cb's, two failed. I was not at the site but I understand the work was performed by a well known NETA testing company. You may want to try it on a few breakers and see how it goes. If it works; write a paper and submit it for the next NETA conference.

 

[EMT01]

DPMAC: LTD @ 300% has gone up tp 15X longer than published trip time and we have terminated the test. The same goes for other functions. Once the breaker "cooks" for this amount of time the breaker will usually fall within the curve on the next attempt, but if not by the third or fourth attempt 97% of the time. We have let the breaker(s) sit over-night and re-test and they will usually fall within the curve on the first attempt, but if not they seem to fail on another function and the process gets started over again.

The pick-up does not seem to vary (except in the case of a bad trip unit).

We have not continued to test the breakers on a particular function once it has fallen within the curve.... I see where you're going...thermal memory / imaging?


ALEHMAN: We have not tried to determine if something in the thermal imaging is bringing the breaker into the curve, but that's an idea.


TROSEPE: The only problem I see with your method is the 20+ minutes of cool-down required between tests, since thermal cannot be defeated without the secondary injection set and will cause the breakers to trip sooner, otherwise.

EVERYONE: I THINK WE MAY BE GETTING SOME GOOD IDEAS HERE, BUT WHAT I'M HEARING IS THAT NO ONE HAS HAD THE SAME PROBLEMS AS US. I SURE WOULD APPRECIATE IT IF EVERYONE COULD KEEP THE IDEAS FLOWING. WE ARE CONSIDERING TAKING SOME "VIRGIN" BREAKERS TO ANOTHER TESTING COMPANY AND SEEING WHAT / HOW THEY DEAL WITH THE SITUATION.

 

[dpmac]

alehman:

Thats what I am getting at in my above reply. It really sounds like the Thermal Memory or Imaging circuit is amuck.

EMT01:

The test sets we used were Square D S33595 Full Function Test sets.

 

[alehman]

I think it might be worthwhile to take a sample of the failed breakers to another shop, preferably with a different brand of primary injection test set and see what happens.

Alan
----
"It’s always fun to do the impossible." - Walt Disney

 

[davidbeach]

Sounds to me like there is something in the trip unit that needs to be charged, and it only charges very slowly.

 

[dpmac]

Any further developments on this situation? I would really like to know what the deal was.

 

[EMT01]

5/25/2009:

As a follow-up to this problem, we took some breakers that we had already tested to a well respected NETA firm in the Chicago area for re-testing and also testing of some virgin breakers. We had them re-test (2) Masterpacts with the 6.0A trip unit (PRI. INJ.). One tested OK and the other failed on GFP. The unit that failed on GFP was re-tested after lunch and passed. We then moved onto the units with 5.0 trip units (no way to defeat thermal) so each test took 20 minutes. There were no failures on the two virgin breakers.

Last week we had Square D in our shop and they witnessed our testing procedure on breakers that had been previously tested (now 5 times) and there were no failures, as expected since they had been tested before. Remember once they pass a primary injection test on all poles they are almost sure to pass any subsequent test (so far that is). We then moved onto some virgin breakers with the 5.0 trip units and had 2 out of four fail. As a diagnostic test, one of these passed secondary injection the other did not.

Square D, satisfied that our procedure was correct, wanted to disassemble the breakers and begin troubleshooting. We stated that the purpose of this testing was to validate our procedure (again) and for Square D to witness the failures. We have yet to receive a response from Square D...