RCD Tripping Nuisance / Faults

[Gavona]

Q1: I am not far off getting professional in but which professionals and how to find a good one?
I have researched and investigated a bit myself and concluded that:

• My clamp meter (MP780050 £40ish from Farnell, so cheap not calibrated, no ability to detect dc etc etc) was really helpful in getting me this far. Along with accessible tails coming out of the meter clamp and an adapted extension lead to test individual devices.
• I have nuisance current that certainly warrants moving one or two circuits to RCBOs if possible without a new CU.
• My Dual Oven registers leakage (varying with which heaters are in use) up to 15mA for the top oven and 20mA for the bottom. When both on it appears to be around 38mA.
• My plug-in Combination Microwave oven registers 30mA for the microwave, 22mA when used as oven/grill and 50mA when used in combined way
• When testing the Oven and Microwave combi above were the results just for the device in question. However the leakage at the meter tails went up to 62mA, including probably 6mA through RCD B, so 56mA through RCD A. Neither RCD tripped! The test button on both RCDs works. Should I be asking electrician to look into DC or other issues affecting the sensitivity of the RCD? I remember reading something about this and different types of RCD being available.

My consumer unit was installed in 2009 when I had building works done and that and the Meter were moved to the garage. It is plastic, probably MK as all the breakers are MK. There is a main switch, two RCDs, 15 circuits (including 1 unused and one not protected (smoke alarms etc))

 

[Lucien Nunes]

Your leakage current measurements appear to be significantly in error, or your appliances are faulty.

E.g. re, the combi microwave (your reading 50mA). No appliance connected via a plug is permitted to leak more than 3.5mA. Normal leakage for that appliance, if its elements are OK, would be 1-2mA

The dual oven (your reading 38mA) is permitted to leak no more than 0.75mA per kW or 5mA total, whichever is lower.

Therefore any inferences about RCD behaviour and circuit alterations should not be based on these figures until you can confirm them.

 

[Gavona]

Further background: My Router tells me that I have 20 wireless and 10 wired devices connected at the moment. Most of them will have chargers/power suplies plugged in most of the time.

I forgot to mention that my 2 laser printers seem to leak up to 10mA very briefly when first switched on then return to normal levels.

Q2: Given the amount of electronic devices I have, should I consider whole house surge protection? eg the Lewden Unit (assuming you don't tell me that I need a new CU anyway).

 

[Gavona]

Thanks Lucien - yes indeed. I certainly need a sparky to confirm measurements etc with decent equipment and better know how than me! Will all electricians have sufficient knowledge and expertise to help or how do I find one who does? I presume an electrician will likely stop at pat testing the combi then leave me to find someone to sort the issue?
I certainly have a few jobs lined up for Electrician - just trying to avoid people all over the house at these times.

 

[pc1966]

The simplest check for RCD tripping is to get a professional in who will have a MFT or similar and can do a ramp-test on the RCD to find out the trip current. as well as checking trip times at the standard test currents.

It is not just total leakage that can trip the RCD, it can be down to switching that results in a lot of electrical noise. My own oven occasionally trips the RCBO when turned off, but not on its thermostatic control during operations.

 

[loz2754]

The dual oven (your reading 38mA) is permitted to leak no more than 0.75mA per kW or 5mA total, whichever is lower.

This is no longer the case. CoPISITEE Edition 5 now only has one limit for all AC equipment, that limit being 5mA.

 

[Lucien Nunes]

This is no longer the case. CoPISITEE Edition 5 now only has one limit for all AC equipment, that limit being 5mA.

Dammit, old habits die hard! Thank you for the reminder, unfortunately I can't find my Tipp-Ex to correct the previous post! FWIW I can't recall the last time I saw an item that would have exceeded either limit.

I have 20 wireless and 10 wired devices connected at the moment. Most of them will have chargers/power suplies plugged in most of the time.

Many of the wireless devices are likely to be floating relative to earth, with class II chargers, in which case they won't generate more than a few microamps of leakage. Those that can see earth from their ELV side will have a touch leakage of a fraction of a milliamp, so I would be surprised if in practice those 30 items are contributing more than a few mA of differential leakage to your total.

That is not to say that you don't have an issue with cumulative leakage; my point was more that the figures you mentioned for those specific appliances are significantly out of spec and I would not take them at face value.

 

[littlespark]

On your circuit chart, you mention that a circuit is disconnected as you drilled a hole in it.... from the photograph, nothing is disconnected, only turned off.
This only interrupts the live feed, not the neutral which the rcd also monitors.
you may have a slight short circuit between Nuetral and Earth on that circuit causing the rcd trip.

 

[James]

Firstly,
congratulations on a well thought out and documented post about a potential problem.

We are starting from a much better position than " I have changed this light fitting and now this wont work"

As i was reading through, i thought firstly, are you using the clamp meter correctly?

option 1
clamp round both L and N cables together.
in my opinion, the best approach, it will give a true reflection of the leakage current on that circuit.
The current used by the device on test is precisely and naturally removed from the meter reading.

option 2
clamp round the cpc (earth) cable.
Many things may sway the reading away from the true leakage.
parallel paths to earth being the main one that will reduce the measured value,
but there are other more in depth reasons why a large current could be traveling to earth but a much lower current "missing" from the L to N flow.

without some in depth analysis, I think that @littlespark has probably hit the nail on the head.
if i was trying to figure it out, the first thing i would do is disconnect the faulty circuit (both live and Neutral)

If I didn't have anything to safely terminate the bare ends of the cable after I disconnected them, I would attach both of them to the earth bar in the board.

 

[Gavona]

@littlespark Good thought but I don't think that is it, sadly. I did switch on one circuit at a time to do the testing in the table (well almost I had the Heat and Smoke alarms connected all the time) - that circuit recorded as 0.3mA. The hole I drilled is on a warm wall so I doubt condensation etc would affect it and I exposed and separated the conductors. Either wrapped them in insulating tape or put a choc block on, or both.

@James: love your strap line ?
Option1 - Yes I did this. Yes the jaws were fully closed. For the plug in devices I mention (and many others) I used an extension lead where I stripped off the insulation so I could avoid the earth. (I also had decent results with that wrapping the live and neutral around the clam twice to double the current reading - that was a good tip I found). For the oven I tested by removing the switch near the oven. I found almost identical results when I did the testing again around the meter tails. Having done all the tests twice on different days I am pretty confident that they are indicative of the issues - but it is only cheap meter.

Option 2: I did check this a bit but recognised that it wasn't as good an approach. The one thing I didn't do was remove network cables from those printers but when I retested one of them today I saw similar issues having disconnected it.

Totally convinced that it isn't the half disconnected circuit but its good to understand how I should do it.

"there are other more in depth reasons why a large current could be traveling to earth but a much lower current "missing" from the L to N flow." That is intriguing - I have to know more even if I only half understand it! (Not relevant to this thread though).
My thoughts were that by testing at the meter tails I was hopefully avoiding any issues with shared neutrals.

 

[Gavona]

So how to find a good sparky? (HP16) Or rather how do I tell a good one from a .... not so good one for this sort of work.

 

[pc1966]

So how to find a good sparky? (HP16) Or rather how do I tell a good one from a .... not so good one for this sort of work.

If you have folk locally how have used the services of someone and happy - go with that.

Otherwise the sparks on this forum are above average as they actually have shown an interest in the subject and the best practices. There are some search options here, but not terribly good.

 

[Gavona]

Thanks, Couldn't see any in the Trusted Electricians part of the site
I suppose I could use the top one listed as "The Best 10 Electricians near [me]" but how rigorous will their testing have been lol?

 

[DPG]

I don't think a £40 clamp meter is going to give accurate readings at those low levels.

Hopefully someone on here can have a look.

 

[123]

Which of the RCD's is tripping?