When you clamp a ring of magnetic material around a current carrying wire  you are in circuit diagram terms adding a small inductance in series - a few microhenries for a typical clamp meter.  When you open the jaws, this equivalent inductance drops sharply, so if there was current flowing at the time, as LI  has to be conserved, you create a small voltage spike. This induced transient may or may not be equal in L and N conductors, and an RCD near the edge of tripping, or susceptible to short duration signals (and some electronic ones are over sensitive to fast edges/ high frequency) could just be set off.

The RCD is definitely on the edge of tripping, the imbalance on the tails is around 20-27 mA. The main earth has 9.5 mA running through it.


I was pondering a collapsing magnetic field when the jaws are opened.


 Andy B.

I like testing caravans that are plugged into a site supply, assuming that is the arrangement you can basically unplug it and “portable appliance test” the whole caravan.


So if you are treating the caravan as an appliance with a flex and plug you need to consider how to test the RCD. 


When portable appliance testing you may need to test plug-in RCD devices, plugs and adapters. If you know what you are doing you can make up a fly lead testing adapter that uses the neutral as earth just for portable RCD testing and when for you are checking your own test meters, so you can do a high-loop test on a RCD protected circuit without the installation RCD tripping.


Andy Betteridge

Do please add neons N-E and LE to that trick lead, (or use a double socket and a plug in polarity indicator ) so that when you find a caravan site socket with L-N reversal you are not using L as the spoofed earth for the caravan.

That neons NE & LE is best with a high value resistor LN too.

Gives a missing N or E indication by striking both neons.


We used to make such as homemade polarity testers a few years back.


From memory (So I might be wrong, check it first) I think it might have been 220K ohms, therefore 3 x resistors (22K ohms) one each in series with the neons to give LE and NE the just a resistor for LN. This meant a missing N or E would effectively have 2 x resistors in series with the neon, it strikes a bit duller but still as noticable (or not!) as the neon.

Of course, if you have room on a socket, you could double up those neons and resistors just to give a little more resiliance to component failure.

Probably not much better than the outlawed "Mains Tester Screwdriver" but I suspect even the those might have saved a few from harm.

So top marks to MAPJ1 for adding this bit of advice

The little neons (the 1/4 inch glass tube ones) typically take a mA or 2 RMS, and go into explosive failure at about ten times this , the neon itself drops 60-80v depending on gas pressure inside, size the resistor on the remaining volts. For those I'd suggest 22k is bit low  for anything more than short bursts - 100 to  220k is more like it for long term reliability in UK mains, and 75 - 100k or so for 110V systems.  The panel mount ones usually have the resistors in-line already.


I have a number of unofficial home made boxes with neons to different plugs that everyone laughs at until it spots a fault like polarity or earth off - nothing more than a plastic box with a stuffing gland for the flex and plug, and some holes for the lights.

With 'power of opportunity' - gensets, line taps, leads from other buildings (don't ask) especially in parts of the world with less care than the UK, a quick sanity test can save expensive kit or even lives.


You can also do 3 phase rotation testing with a 3 Rs and one C  and a neon, if you know the mains freq. Or you can spend a fortune on shiny test kit and watch it get nicked as it crosses a border.

As with all live testing precautions need to be taken, PAT RCD testing of portable RCD devices can be undertaken of a controlled basis, but you really need to be ensuring you have completed all the appropriate dead tests before moving onto any live tests, particularly insulation testing, with a caravan or any other installation you need to be working from the intake or power supply with the final circuits isolated. Normally you should always do a loop test before testing a RCD.


The issues arise when you have to decide if you are going to remove the cover of the consumer unit within a caravan to test the RCD or just plug into a handy socket. If unsure what to do it would be best to put the RCD tester away and just press the test button, but even just pressing the test button can fire the upfront RCD off.


I take the point that Graham has raised about introducing a TNCS earth to a caravan, even just for testing.


 Andy B.

So, the correct answer is use. RCD testing isolation transformer.


But it doesn’t appear to be rated for a times five test.

Wouldn't this provide electrical separation to more than one item of equipment if used on a caravan or mobile/transportable unit?


I'm not a big fan of that idea, especially as the RCD test will actually introduce a fault ... and I may be in contact with exposed-conductive-parts at the same time ... for example when resetting the RCD in a transportable unit with metal CU !


I think I'd prefer to simply connect the caravan or unit to a socket-outlet with additional protection by 30 mA RCD in accordance with Regulation 411.3.3 or 708.415, and conduct the "up/down" test on the RCD in the caravan or mobile/transportable unit !