Circuit breaker testing

[Singer]

I've been asking this question for some time but never seem to get a definitive answer.
Why are we compelled to test RCD's to verify their correct operation but not MCB's?
MCB's or simply circuit breakers are being sold on ebay - all types, all ages, all makes - and no-one has any idea what they've been through in their lives - how many flooded houses they've been in, how many short circuit conditions they've had to cope with, they could have survived a fire in a CU, any number of traumas and yet there is no requirement to test these devices to verify correct operation.
In all my 40 year career as a spark I have never known an mcb testing device.
We all just install them and trust what the manufacturer is telling us ie. These devices are manufactured to a British Standard so as long as you satisfy the Maximum Zs requirement for the particular device there is no need to test - we guarantee it will work.
Well if that's the case why can't they say the same for RCD's?

 

[malcolmsanford]

Well I'm sure you can get bench type Electronic Modular Units (EMU) similar to those used in high voltage applications for testing MCCBs for smaller devices, but not sure how much this would cost, as it is not something you can put into a normal multi function test meter

 

[Engineer54]

In all my 40 year career as a spark I have never known an mcb testing device.

There has been test equipment available to test MCB's/MCCB's for donkey's years, but as Malc states this equipment can be expensive and is generally outside the means of small contractors, sole traders or journeyman electricians etc . Your basically talking about ''Primary Injection'' test equipment. So the larger the OCD rating, the larger the output of the test equipment needs to be and can start getting very expensive indeed. ...lol!!

Looking at the test equipment that forms part of the contract (which will be handed over to the future maintenance dept) i'm now on, we have a primary injection kit that is capable of delivering 3000+A and looking at the current cost, well exceeds US $10,000. Not saying you need a unit of this size, but even for some MCB's your going to need a decent primary test current....

 

[Singer]

Thanks guys,
The question I'm asking is not about mcb test equipment but, why do we need to test RCD's (and include results on test certificate) but not mcb's.
So really we're assuming that all mcb's will work but have to test to prove that an RCD will work - that's my point.

 

[malcolmsanford]

Safety is one aspect of it, as E54 and I said your going to have to inject quite large amperages into the MCB to get it to trip on overload and fault and unless these are controlled rigorously and in safe environment you are going to open yourself up to quite a lot of risk.

So this moves onto cost and that is why.

RCDs are quite simple your only injecting milliamps into the device to get it to activate in fault conditions.

 

[robd]

The tile cement across terminals method seems to work pretty well for a tiler I know! :shades_smile:

 

[Knobhead]

Singer, have a look at the inverse time curves for say a 40A MCB. Then work out the amount of current you’re going to have to inject to get it to trip within a set time. OK it can be low voltage, but you’ll be playing with 100’s of amps.
By rights a MCB that’s broken a heavy fault current should be changed, carrying out the test would essentially render the MCB unfit for use. You have to take the manufactures “type testing” at face value.

As for the Ebay stuff, as you say you don’t know it’s history or if it’s a fake. Steer clear.

 

[Singer]

Singer, have a look at the inverse time curves for say a 40A MCB. Then work out the amount of current you’re going to have to inject to get it to trip within a set time. OK it can be low voltage, but you’ll be playing with 100’s of amps.
By rights a MCB that’s broken a heavy fault current should be changed, carrying out the test would essentially render the MCB unfit for use. You have to take the manufactures “type testing” at face value.

As for the Ebay stuff, as you say you don’t know it’s history or if it’s a fake. Steer clear.

Hi Tony,
Yea that's exactly my point.
Within our industry safety is paramount, which is why we carry out cold tests,hot tests, tripping times, Zs, Ze, etc etc again & again to make sure everything is ok, but what we don't test is the thing that's going to switch off the supply (or not) if a fault occurs and I think this is a major oversight. Yes I understand that the practicalities are a big problem but is it not true that we all stick a circuit breaker into a board and ASSUME it's gonna work without proving that it will.
I changed a CU the other week that had gone up in smoke and the loss adjuster asked me what I thought had caused the fire. We'll never know for sure but it could have been a faulty mcb.
We all put stickers on the covers telling the consumer to test the RCD regularly to keep it moving freely and verify it works but you then have to assume the mcb is OK without verifying that too.

 

[Richard Burns]

The risks to life and the damage possibilities of testing an MCB are too great in this environment to be permitted.
The method of operation of an MCB is very simple and if it is faulty then there would normally be no continuity through the MCB, the functional testing we do ensures that this level of function is available.

 

[Engineer54]

MCB's have been around for a very long time now and it can be said that they are a proved OCD design. The same can't yet be said of RCD devices, regardless of what others have stated in other threads. Yes there can still be failures or duff ones, but they are relatively few and far between and most will be due to units that have been unduly stressed. We for arguments sake, only test installed MCB's if there is a suspicion on it's performance....

Having tested more than a good few MCB's over the years, i can honestly say that when MCB's do fail it's generally on the thermal side, rarely on the instantaneous side. In fact i can't remember any that have failed on instantaneous...

 

[Marvo]

If it came to having to test the mag and thermal operation of domestic MCB's it would be far cheaper and quicker to just replace them. To even remotely plot the thermal curve of an MCB would take hours if not days on a bench.

I'd be interested to see the manufactures testing schedule and procedures for the thermal operation of MCB's and what percentage of their products coming off the line go through full QC thermal function tests as a fully assembled unit (not just at component level).

 

[Engineer54]

If it came to having to test the mag and thermal operation of domestic MCB's it would be far cheaper and quicker to just replace them. To even remotely plot the thermal curve of an MCB would take hours if not days on a bench.

I'd be interested to see the manufactures testing schedule and procedures for the thermal operation of MCB's and what percentage of their products coming off the line go through full QC thermal function tests as a fully assembled unit (not just at component level).

No not day's Marvo, lol!! Generally you only test at 2 points on the thermal curve, and rarely at the top end of the curve. The test equipment has an integral timer that gives the final trip time. So were talking about a few minutes or so for each test position, but obviously that depends on the current and time you have chosen for your curve test position...

No idea what percentage of units get checked by the manufactures, but at a guess i would say they take X amount of units from each batch run and subject them to a full cycle of type tests. The last thing any manufacture would want, is a recall of Lot or Batch numbers... very expensive business is that. ..lol!!!

 

[Marvo]

Yeah, maybe 'days' was overstating it somewhat but even at 1.5 x the rated current which would be a sensible lower point you'd probably be in for an hour or so test time depending on the curve.


According to this document kindly pinned by Lenny on this thread the UK regs state that where two or more faults have occurred between the Ics and Icu values of the MCB the continued operation of the circuit breaker must be verified? Do they outline the verification procedure anywhere?

 

[i=p/u]

well havent had read of all the thread, but all would be needed is a resistor to to allow an Ia current to flow..

its as easy as this ...... high current zs test trips 6A breakers some times...

 

[malcolmsanford]

well havent had read of all the thread, but all would be needed is a resistor to to allow an Ia current to flow..

its as easy as this ...... high current zs test trips 6A breakers some times...

:uhoh2::whatchutalkingabout