Caravan hook-ups, RCD's not tripping @5XIn and earthing (PME)

Hi all.

A local caravan site has asked me to test their RCD's. I popped in to speak to them about what they wanted and had a quick nosey around with a view to returning and undertaking the work.

The 32 electric hook-ups are fed via a 3 phase system which splits into 3 circuits. Each circuit is protected by a 100 mA RCCB, I think these are time delayed. Then at the hookup point each unit has 2 16A MCBs and they share a 30mA RCD unit as per Reg 415.1.1. (I note 708.553.1.12 and .13 says that each socket outlet shall have its own MCB and RCD etc).

The earthing system is definitely PME - I've spoken to the DNO engineer that installed it

When I did a quick test on a RCD at the hook-up it didn't trip out at 1/2 In but did at 1xIn and both parts of the waveform (o and 180 degrees). However when I tested at 5xIn my Megger kept reading back >50V. Does anyone have any idea why this might be? I tried switching over to the time delay setting but this had no effect, and as I say, they are standard 30mA RCD units. I tested several just to make sure it wasn't a dodgy one off.

I have a sneaky suspicion that there is no additional earth to the hook-up points and the manager thinks this might be the case too. Where the main incoming cable comes into the distribution box there is a separate 16 mm earth cable next to it running out into a soil pipe and into the ground. The DNO has a vague memory that originally an earth plate/rod was put in by a spark, but can't recall how long ago. The campsite manager is really helpful, but has only been there 3 years and there are no site records to show what has been done in the past. The owner has little interest and no clue either. (Incidentally the manager dug a test hole for me and we found the SWA cable serving the pitches barely at 30 cm with no additional physical protection/ warning tape or anything).

Sorry to sound ignorant but can I test this mysterious 16 mm cable by carrying out an earth fault loop impedance test as per GN3 page 17 to see if there is a means of TT earthing for these hook-ups? I have a Megger 1552 but no other specialist earth electrode test equipment?

Does anyone know how long earth rods/plates last for in the ground (sandy soil)? I'm just wondering if the site had a TT earth and it has corroded away and now not viable?

They also have 6 statics on site (3 belong to the caravan site and 3 are privately owned), that are fed from a different source (single phase) but also PME). The caravans are not far from the distribution board and on testing the RCD's either the Hook-up points or the main board RCCB's are tripping out at all ranges. However there is no time delay on these units and when testing it is somewhat a game of 'russian roulette' as to whether the hook-up RCD or the main board RCCB will trip out! Obviously this needs sorting out, but again the individual vans need their own TT earth? Is there any obligation upon me to notify the van owners that their vans are not correctly earthed and thus potentially dangerous under fault conditions? I have no way off obtaining the owners contact details unless the manager/owner wishes to advise me?

Any comments/observations would be gratefully received.

Think your first check must be the earth at each point. As you know from the the Regs and the ESQCR 2002 you can not feed caravan pitches with a PME, the static IMO should also not be PME as I would not class them as permanent buildings, but that is another argument. What you might find is that the SWA to the pitches is protected by the PME, but the pitches itself are TT. So the SWA would be connected to the earth at the origin and not connected at the pitches.

What you are going to have to do it check to see if you have an EFLI at your Pitches. This can be done either by doing a Zs or geting your R1+R2 at the pitch. I would first go down the R1 +R2 route as this is a dead test and you will be safer with this.

It seems a strange set up that each phase distribution circuit is protected by a 100mA RCCB only, what is providing the circuit with overload protection?

As you have quoted in your OP about the socket protection, was this install done under BS 7671-2008 or the 16th, I can't remember off hand the regs regarding this in the 16th.

So I would check from your SWA if it is Earthed back to the origin, and confirm that it is not connected to the pitches Then I would check at the pitch enclosure where the earths go to and trace back from there. You may find that all 32 pitches just have the one rod and this as you say maybe corroded or not even connected. It is really a case now of tracing things out.

this may be a long shot but i recently had the same problem but in a domestic dwelling. A 100ma RCCB time delayed main switch feeding a 63A RCCB(30ma) etc.....TT earth system 5 ccts on 30ma side.
Ze was good, bonding in place etc.

Customer said board kept tripping out randomly, however when i tested at the sockets with megger, it would not trip at all but would show >50v?????

So, i tested the RCD at board from the incoming side and it tripped ok??
this really bafflled me.........................after about two hours of changing the RCD, speaking to makers of DB eqpt i was still at a dead end.
i did the same RCD tests on other ccts fed from the same RCD with same result? (>50v)??
this is where it gets interesting, after doing IR tests on ccts there was a nicked neutral cable at one of the socket plates by the socket plate screw in the kitchen......................it was tiny. after re sleeving and connecting the RCD tested out fine??????????
So, how it affected the other ccts and did not trip the rcd regurlary is a mystery to me? (it was very random)
SO what im getting at is check the ccts for a possible fault as it may affect the RCD test.

Hope that helps and makes some sort of sense? took me half a day to sort it out?!!!!!!
Next time i will disconnect each cct at a time from CU to see if i can eliminate the problem cct earlier.

If the pitches are rented then its up to the site owner to sort it out , i would carry out all of what malcome has said , i am taking it that the hook up points are the standard 16 amp sockets that you plug into , if this is the case make up a test lead with a 16 amp socket and 13 amp extension socket then its easy to test at this point and for stability reasons the readings should be <200 ohms TT at each point , one of the easiest ways to rectify this is to install an earth rod at each point and disconnect the PME earth to the point and connect the Earth rod to the protective conductotor at the hook up each hook up point should be RCD main switch ,MCB overload protective device with a maximum rating of 16 amps hope this helps

Great post richy !

This will occur when the Ra is greater than 333 Ohms

5 X 30 = 150 Ma which equates to .150 x 333 = 50 volts

so anything above the meter wont test.

Does the owner of the site have an EIC or previous PIR that you could refer to,this should give the info on the type of earthing etc.

SPARTYKUS said:

Great post richy !

Click to expand...

Cheers dude. I didn't realise you were active over on this forum

55 days

Thanks everyone for the quick and detailed responses. The work was definitely not done to 17th Ed and most probably 15th or 16th, but the manager cant say when it was installed.

I had forgotten that I had taken a couple of pics on my phone which might make things a little clearer.

The hook-ups are 16A and appear to be connected to the SWA and no other earth.


This is the cut out. Hopefully you can just see the 16 mm earth cable disappearing down the pipe (bottom left).


The other end connects into a Henley block arrangement and then feeds out to the 3 phases as per this photo.

wirepuller said:

Does the owner of the site have an EIC or previous PIR that you could refer to,this should give the info on the type of earthing etc.

Click to expand...

Afraid not wirepuller. There has never been any EIC's and certainly no PIR's supplied to the customer - This is the Highlands Last year the spark that checked the site just said it was 'OK'. Despite the site manager asking for the report, he never received it, and neither was an invoice forthcoming either. The customer would have been lucky if the spark was actually qualified and not just someone that owned a screwdriver

richy3333 said:

Thanks everyone for the quick and detailed responses. The work was definitely not done to 17th Ed and most probably 15th or 16th, but the manager cant say when it was installed.

I had forgotten that I had taken a couple of pics on my phone which might make things a little clearer.

The hook-ups are 16A and appear to be connected to the SWA and no other earth.
View attachment 4784

This is the cut out. Hopefully you can just see the 16 mm earth cable disappearing down the pipe (bottom left).
View attachment 4786

The other end connects into a Henley block arrangement and then feeds out to the 3 phases as per this photo.
View attachment 4785

Click to expand...

oh please tel me you dont have to take the cover off to get at the MCB's if that the case all the enclosure will have to be replaces so you can get at the MCB's as having to remove the cover to reset the RCD or MCB is totally wrong and exposes the user to live parts it should look some thing like this

Don't like the use of an earth clamp on the SWA - although better than the site I have seen last week - none of the SWAs earthed at all - not a gland in sight!

nickblake said:

oh please tel me you dont have to take the cover off to get at the MCB's if that the case all the enclosure will have to be replaces so you can get at the MCB's as having to remove the cover to reset the RCD or MCB is totally wrong and exposes the user to live parts it should look some thing like this

Click to expand...

It's OK Nick.
The MCB's and RCD can be reset by lifting the usual weatherproof cover. You do not need to remove the whole front cover to get at them

richy3333 said:

It's OK Nick.
The MCB's and RCD can be reset by lifting the usual weatherproof cover. You do not need to remove the whole front cover to get at them

Click to expand...

Phew im assuming its like the one in the pic i posted , i got worried for a minuit lol i have come accross the MCB's RCD in an enclosed box
well thats one good thing

In the process of doing PIR's on a load of caravans on a big park. Half of the boxes do not have the flap to gain access to the RCD/MCB's and causes no end of teouble for the on-site spark.

WIth regards to the tripping on 5x, the earth will most likely be the problem. Is the site located close to the sea??? as salt in the atmosphere will speed up the corrosion of the metalwork. Pull back a couple of the SWA gland sleeves and check the SWA strands. COrrosion is apparant on the earth stud on the first picture posted. A simple Loop test will confirm the integrity and condition of the system, and yes the higher the reading, the less likely the RCD will trip at 5X....