Can't test RCBO as main RCD trips first - what to do?

[cornburn]

As above, came to commission a circuit yesterday and got to the last step (test the RCBO) and on 3 attempts the main RCD went instead. TT installation, no markings on front of RCD but assume must be 30mA non time delay.

The obvious answer is quote them to swap for a type S RCD but it's a big 3 phase one and will be expensive!

My main problem is what to do about certificating? I've commisioned the circuit as everything else tested fine and even if the RCBO I've put in was faulty it still met disconnection times because of the main RCD tripping.

It's a 6A lighting circuit. Strangely enough there are loads of other RCBOs on the board too so I can only assume that these either haven't been tested or someone bypassed the main RCD to do them!

 

[Plonker 3]

Well assuming is a dangerous thing to do, have you ramp tested both RCDs to ensure they both work correctly?

 

[telectrix]

why an RCBO if it's already RCD protected?

 

[AndyL]

Ramp test main RCD, may give indication of its rating if you cant find it marked

 

[richy3333]

Surely a pre-change survey highlighted that an RCD was already fitted and you would have noted the type etc? I guess customer will be ****ed off at having to fork-out for more work?

Just fudge the cert, cheat, lie, else like plenty of other sparks don't send them the certificate. That should avoid the problem

Sounds like the up-front RCD isn't 100mA time delayed?

 

[Guest55]

No need to cheat or lie , just write a note in the " Comments on existing Installation" box saying something like ;

Unable to complete Rcbo tests due to unsuitable main switch device not providing discrimination to downstream devices.

Well its the truth aint it ? lol

 

[richy3333]

^^^ Yeah, Apologies. Bit fed up with the industry today. A combination of the **** Part P theme and a local 'electrcian' advertising as an NICEIC spark when he isn't Why cant people just do work properly, or not at all?

The above wording is good.

 

[cornburn]

Thanks all. Only got a quick look at the board prior to the work and as recent we assumed by the number of 30mA RCBOs in there that the main RCD was a 100mA type S. Clearly it isn't :S They've just had the PIR too so this obviously wasn't picked up on which is strange (as they noted other things on the same board).

Also there's about 20 circuits in total so an RCBO was specced to avoid knocking the whole building out.

Will note on the cert that not possible to test like you say and recommend having it changed. NIC assessor is out in a few weeks so might stop off at this one to see what his take on it is.

Thanks again

Edit: Other option is to swap to a breaker I suppose and put down the test results for the RCD already in place, loses the discrimination between circuits but the paperwork would be better.

 

[topquark]

Tell the customer there is a problem with the installation and note it as biff has suggested. Then give them an idea of the costs to change the main RCD for one that will provide the necessary discrimination and inform them that it should have been picked up on the EICR(PIR).

 

[Engineer54]

why an RCBO if it's already RCD protected?

Actually very sensible to have an up-front 100mA RCD on a 200 ohm or the like TT system, where your relying totally on a RCD, for your earth fault protection. But it should obviously be an S type unit!!

 

[vernam616]

if all other circuits are RCBO, change that main rcd for main switch, that will then let the RCBOs operate how the customer wants them to ( i.e if one trips it doesnt knock other circuits out)

 

[rich.250]

if all other circuits are RCBO, change that main rcd for main switch, that will then let the RCBOs operate how the customer wants them to ( i.e if one trips it doesnt knock other circuits out)

On a TT?!?!?

Either S type upfront, obviously this will cost them....
Or if there is a 30mA upfront take the rcbo out as you are just putting two 30mA in series which is senseless LOL...

 

[rich.250]

^^^ Yeah, Apologies. Bit fed up with the industry today. A combination of the **** Part P theme and a local 'electrcian' advertising as an NICEIC spark when he isn't Why cant people just do work properly, or not at all?

The above wording is good.

Your just fed up with that today...??

 

[LineProducts]

Just a thought...I assume that the testers check an rcd by leaking current to cpc. So, you could then test the rcbo by connecting the cpc end of tester to the neutral on the supply side but before the rcd. Only the rcbo would trip as the rcd would not detect any leakage. Or am I missing something?

 

[rich.250]

They create an imbalance, they don't leak current.
Its pointless even testing it really if one trips first.
Layout of rcds needs sorting 1st IMO.

I know megger had some issues with certain rcbos, where reversing lead connections could solve problems, i know its not a megger but may be worth a try lol

 

[vernam616]

On a TT?!?!?

Shouldnt matter if all circuits were protected by 30mA rcbos

 

[Jimmy Boy]

1) Establish by testing by ramp or device enquiry the upfront RCD type 100/30mA TD ?
2)If the upfront is 30 mA why would have other peeps fitted 30mA RCBO's ?
3) If all circuits on board are individually 30mA protected there is no requirement to have a 100mA up front, only if some are not.
4)You state the up front meets the disconnection times ? what on x1 ? and/or x5 ?
5) Establish the characteristics of the upfront, make sure it meets the regs, bridge out and test your RCBO, note the findings on the cert.

Your the last one to work on the board..It's your --- on the line if someone gets an --- winder mate do it once do it right !

 

[rich.250]

That is true, but personally would always put a S TYPE upfront if i could.

 

[rich.250]

Just cos he's the last one to work on the board doesn't mean he's responsible for it all!
Only what he's put on his cert in the extent.

 

[Jimmy Boy]

Just cos he's the last one to work on the board doesn't mean he's responsible for it all!
Only what he's put on his cert in the extent.

Agreed..BUT the OP suggested that he wouldn't test HIS RCBO and rely on an RCD ( One by his own admission he can not identify ) to give additional protection, so he either has to falsify the cert or investigate ? Failure to investigate would make him liable sorry