Testing Breakers

Merry Christmas
Status
Not open for further replies.
C

cirelo132

Member
Hi All, I need to test a circuit to determine if the breakers protecting it actually work. In order to do this, I was planning on simulating a fault at the outlet in question and simply seeing if the breakers trip. However, I am not sure how to do this safely. In case the breakers do NOT trip, I don't want to cause a fire or blow up a transformer somewhere.

Any ideas on how I can safely pull enough current to quickly trip a 50 Amp 240V breaker?
 
You cannot do it safely, nor can you do it in a manner that will assure you that you have not damaged the breaker by "testing". I am not aware of any method to "field test" a circuit breaker. To quicky trip a 50-amp breaker you would need to overload it by 300 to 400%.
 
If you really need to test this CB this is not the method, a molded case CB can draw anywhere from 4-10 times it rating for instantaneous operation, if it is defective even higher.

50 X 4=200 amps
50 X 10=500 amps
50 X defective trips at the rating of the upstream device.

The proper way to test a CB is with a high current test set, with a 50 amp CB it is not cost effective to test. Unless there is an over riding reason for testing, law suit, loss of equipment, curiosity.

A high current test set cost in the range of 5,000.00-50,000.00 depending on current output required. Then you should have a DLRO/Ductor (though you do not need for a 50 amp CB) and then there is training and experience......OR you could call a electrical testing firm.
 
I short the breaker under test with another breaker rated 2x or 3x the steady-state current of the one I want to test. I put an LED indicator on it so I know it when it trips. I use it only in my own house.

You could add a homemade resistor or inductor out of several feet of small gauge wire to limit the current to some value between the larger and the smaller breaker, and put it in series with the breaker.

#10AWG = 1 milliohm/foot
#20 = 10
#30 = 100 (but this fuses at about 10 A)

If you want to see the "exploding wire" phenomenon, use #40 AWG.
 
brian john said:
I gave my answer. To test as noted is fool hearty IMO.
Click to expand...

Both of Brians post were right on the money, the method given is dangerous, ridiculous, and in no way shape or form resembles a breaker test that has any purpose.

OP, WHY do you need to test this breaker? As brian said, if there is any doubt about the performance of this little thing, it is much more economical to replace the nreaker than test it properly and safer to replace it than test it as described above.
 
Facts about breakers supporting the "Don't Test" argument:

1) MCCBs (Molded Case Circuit Breakers)are designed to pass the basic UL489 test procedure as a minimum. In the "olden days", manufacturers would usually strive to exceed those test criteria, but in the down and dirty price wars we see now, I don't think we can count on that any longer.

2) The UL489 test specifications for the Thermal (overload) trip elements are 50 trips at 600% of rated current while maintaining the trip accuracy tolerances (roughly +- 20%). After 50 Thermal trips, the breaker is no longer required to pass, so all bets are off.

3) The Magnetic (instantaneous) trip elements is required to trip ONCE, then reset, then trip again at anywhere from 600 - 1000% of rated current. They are not required to be reset or trip again after the second trip. SO AFTER THE SECOND TRIP, ALL BETS ARE OFF AS TO RELIABILITY AND ACCURACY.

So what does that say? It says that testing the breaker even once by shorting the output will consume 1/2 of the expected minimum life in terms of accuracy and reliability. It also says that because of that, the breaker contact materials are not going to need to be substantial enough to handle any more than those two events, so it is also highly likely that you will cause severe damage to the breaker with testing. Bottom line: don't test MCCBs.

The entire issue of testing circuit breakers is, IMHO, an over interpretation of what people are used to with Low Voltage Power Circuit Breakers (LVPCBs), which are a completely different animal from MCCBs. LVPCBs are usually in the 500 - 3000A range used in plant wide distribution systems and are expected to hold in under fault conditions long enough to allow the fault to be cleared by something lower down the chain. This could be as long as 1/2 second, an eternity when you think about the amount of fault energy they could be subjected to. The trade off of having them built to be able to handle that is that they need to be exercised, tested and maintained regularly. That is why we (or testing specialist contractors) get breaker testing contracts.

But MCCBs are specifically designed for smaller branch and feeder applications where testing and maintenance is NOT required, possible or practical and in fact, you are not supposed to do any maintenance on them at all (other than maybe blowing out dust or tightening lugs, and even that is controversial).
 
Jraef said:
3) The Magnetic (instantaneous) trip elements is required to trip ONCE, then reset, then trip again at anywhere from 600 - 1000% of rated current. They are not required to be reset or trip again after the second trip. SO AFTER THE SECOND TRIP, ALL BETS ARE OFF AS TO RELIABILITY AND ACCURACY.
Click to expand...

This is not true.

A UL 489 breaker is required to trip once at FULL FAULT CURRENT.
 
Thanks for the help guys. I agree that testing a breaker is dangerous, and that in most situations I would just advise replacement of the breaker, especially since I am sure that any test that would cause the breaker to instantaneously trip would probably damage the breaker anyway.

However, I still need to test the system, more to prove that it works now than to prove that it will work. I wont get into the details, but the point is that there is an overriding reason that the system needs to be tested, even if testing it is expensive and means that the system needs to be replaced afterwards.

In any case, it sounds like the high current test set may be the way to go, either buying one or finding a testing company. Got any advice on that?
 
MCCBs

MCCBs

Are they likely to fail by opening late, or opening early?

Are there published guidelines or rules, or informal rules-of-thumb, for periodically replacing these since all testing is destructive or semi-destructive?
 
jim dungar said:
This is not true.

A UL 489 breaker is required to trip once at FULL FAULT CURRENT.
Click to expand...

I think that would be the Interrupting Capacity rating test, not the Instantaneous Trip test. In the IC test you are right, there is no expectation of being able to reset it. All it has to do is stop the fault even if it results in its destruction (without too much collateral damage).

I don't have my UL489 specs with me at this PC, but that's how I remember it. I'll check later when I have access to my full system.
 
Lxnxjxhx said:
Are they likely to fail by opening late, or opening early?

Are there published guidelines or rules, or informal rules-of-thumb, for periodically replacing these since all testing is destructive or semi-destructive?
Click to expand...

As a gross generality, a failure is if it trips outside of a window that is between 80 and 130% of the setting.

I don't know of any published rules as to replacement. The rule-of-thumb I've always heard is to plan on replacing MCCBs after they have tripped on a short circuit, but not immediately. In other words, buy the replacement and put it in at the next PM shutdown. But for thermal overload trips, I have never heard it discussed. Generally, thermal trips tend to start tripping early and become a nuisance, which begs replacement. But technically if the breaker has never tripped or seldom tripped on an overload, they are supposed to "never" need replacement. Never is a long long time though...
 
Last edited:
cirelo132 said:
Thanks for the help guys. I agree that testing a breaker is dangerous, and that in most situations I would just advise replacement of the breaker, especially since I am sure that any test that would cause the breaker to instantaneously trip would probably damage the breaker anyway.

However, I still need to test the system, more to prove that it works now than to prove that it will work. I wont get into the details, but the point is that there is an overriding reason that the system needs to be tested, even if testing it is expensive and means that the system needs to be replaced afterwards.

In any case, it sounds like the high current test set may be the way to go, either buying one or finding a testing company. Got any advice on that?
Click to expand...

You do realize a high current test set will cost you $10,000-$50,000 right?

I guess you have some good reason you need this tested. Does it need to be tested in place or can you ship it out?
 
zog said:
Thats a baby, you cant do much with that.
Click to expand...
Wow... you're almost as big of a fat-headed bastard as me! :grin:

Actually, that unit will be "perfect" for anything I'll ever need it for. Any bigger, and I'd have to sub it out to Brian.
 
Almost...:)

OK. let me rephrase that. "I cant do much with that"

We have maybe 20 primary injection test sets in our shop in various ratings and sizes (But nothing that small), along with maybe 50 seconday injection test sets for different breaker types.

What ever is best for the job, but 400A max output cant test much by means of primary injection, seconday injection yes, but that does you no good for series trip units.
 
zog said:
What ever is best for the job, but 400A max output cant test much by means of primary injection, seconday injection yes, but that does you no good for series trip units.
Click to expand...
Actually, I plan to tap into a market with it that is presently unsupported.

It tickles me when guys talk about not having work. I see work everywhere I'm at. I've got more ideas than I have hours in a day to make them happen.
 
zog: got any loose Masterpact test sets the one we use broke $13,000.00 for a new one 8 weeks for repair.

We use the small phenix, small a EIL and several small AVO's for GFP testing need more current and we bring out the HEAVY boys.

The phenix with CB bolted to the output bus and heavy cable on the other connection MIGHT test a 200 amp MCCB. 150 fer sure, what can be a problem is the long time test, this set has a limited time frame for current.

Single CB ship it out, even if there is a local test company they would most likely take the CB to their shop than lug the test set there. Mail it to me I'll test it no charge (ONE CB RIGHT not a panel worth).
 
Last edited:
Status
Not open for further replies.
Top