Departures from bs7671, certification.

[odawire]

Hi, I have always made sure my installs have no departures from bs7671 but I'm doing a ccu change next week and have done some preliminary testing and the heating circuit (boiler and heating controls) has low IR readings, had a look in all obvious places and cant find any problems, had trouble breaking the circuit up as its a bit messy, client has just decorated so he doesn't want any damage to the house, seeing as its all fixed equipment (I'm unsure on the cable routes but id guess its buried less than 50mm at some point) is it acceptable to leave this circuit unprotected and note it on the departures section? all the earthing/bonding is mustard. Any experience on this would be appreciated, thanks.

 

[Guitarist]

With the greatest respect, don't "try" to rectify the fault, but make it clear that it MUST be sorted before you proceed further. If your test was 7 meg and that was an end to it, then all well and good. A note that IR tests were low would have covered you. However, as it is taking the RCD straight out, then there is something very wrong somewhere.
How would you feel if there's a fire due to this fault, and you'd walked away saying "I had a look, but nothing I could do"?

 

[odawire]

It would only be a c3 if doing a periodic and it was done to old regulations, but u are effectively updating the installation to 17th edition regs by applying Rcd protection, and as u say u think cable is in wall less than 50mm so to 17th obviously needs Rcd protection. But I would explain to customer that it needs sorting before changing Fuseboard, as you have tested circuit and it fails test so further investigation is required.
i also applaud you for testing installation prior as many don't, and has also potentially got you more money out of the job rather than sorting later at your expense/time finding and fixing fault.

Thanks for the reply, BS7671 does not require circuits to be upgraded to current regulations to be connected to a replacement ccu, only for it to be ascertained that they are not an immediate danger (which I would read as no C1's or C2's), GN3 even says there is no obligation to inspect or test any part of the installation!

Also found this : www.niceic.com/Uploads/File1950.pdf Guidance on cu replacement, which also says the same. interesting reading.

 

[odawire]

So...Would you sleep soundly at night knowing that you had connected this known-faulty circuit to the non-RCD side so that the fault wouldn't show up? Daz

Its not necessarily "faulty"

 

[Platinum lad]

Just a thought I would disconnect any zone valves or circulation pump I know from experience they can be a pain

 

[telectrix]

disconnect at the FCU> if it's only single pole, then take the N out of the terminal. then test your circuit. if the fault is on the heating then you need the customer to get it sorted, either by you or a heating " engineer" (that's usually a plumber that can read and write).

 

[DPG]

Its not necessarily "faulty"

Is the circuit faulty or not? It trips the RCD - yes? Daz

 

[odawire]

With the greatest respect, don't "try" to rectify the fault, but make it clear that it MUST be sorted before you proceed further. If your test was 7 meg and that was an end to it, then all well and good. A note that IR tests were low would have covered you. However, as it is taking the RCD straight out, then there is something very wrong somewhere.
How would you feel if there's a fire due to this fault, and you'd walked away saying "I had a look, but nothing I could do"?

That's not what Im saying, its a bit more complicated then that, it may be that the client does not want any work done in which case I need a reason for the circuit to be on a rcd, which to be honest there isn't

 

[telectrix]

odawire, if you get the same reading L-E as N-E, then its one or the other with a load still connected. do continuity L-N first, then 250V L-N this will give you a better idea.

 

[Rockingit]

If you've got an IR measurement of 7MOhms, then you're chasing up a tree. 230/700000 = something too small to bother with, certainly not enough to trip a 30mA RCD. So, test and test again. Have you considered something as mundane as a borrowed neutral somewhere?

 

[odawire]

Is the circuit faulty or not? It trips the RCD - yes? Daz

Yes, which I assume is down to IR breakdown

 

[DPG]

That's not what Im saying, its a bit more complicated then that, it may be that the client does not want any work done in which case I need a reason for the circuit to be on a rcd, which to be honest there isn't

I don't get this, maybe I'm misunderstanding you. It's not a case of whether the circuit should or should not be on an RCD. The issue is the fact that the circuit should work fine if it is on an RCD, but it doesn't. The reason you are putting forward for not protecting it with an RCD is not a regs issue, it's because it won't work when it is fed from one. Daz

 

[odawire]

If you've got an IR measurement of 7MOhms, then you're chasing up a tree. 230/700000 = something too small to bother with, certainly not enough to trip a 30mA RCD. So, test and test again. Have you considered something as mundane as a borrowed neutral somewhere?

Yea, I have thought that, I will be checking that

 

[odawire]

I don't get this, maybe I'm misunderstanding you. It's not a case of whether the circuit should or should not be on an RCD. The issue is the fact that the circuit should work fine if it is on an RCD, but it doesn't. The reason you are putting forward for not protecting it with an RCD is not a regs issue, it's because it won't work when it is fed from one. Daz

My original question was about departures being noted on an EIC, but its turned into me turning my back on a dangerous situation which I haven't.

- - - Updated - - -

Just a thought I would disconnect any zone valves or circulation pump I know from experience they can be a pain

Thanks, ill keep that in mind

 

[odawire]

odawire, if you get the same reading L-E as N-E, then its one or the other with a load still connected. do continuity L-N first, then 250V L-N this will give you a better idea.

Good point, ill keep that in mind

N-L was the same too

 

[Octopus]

Hi, should have explained, I have a rig I made up I use sometimes, its basically an rcd adaptor that I wire into and plug into a socket, when I did this it tripped, the IR reading was bouncing around I think it settled around 7Mohms across all points, I remember it was the exact same number. The reason Im asking is, I want to know if its acceptable to leave it off the RCD side if I have to, as I don't think I can leave them with no heating at this time of year, can I note it as a departure as the cable is probably less than 50mm deep

Maybe you should go back and retest the IR as I wouldn't expect the value you claim to trip an RCD if it really was across all 3 "points"