I may be looking at doing this possibly. My sparky friend must assist as I want him to rewire in any event so that my geyser is also on the critical load db with its own changeover switch.
Correct. Also known as an ELB (earth leakage breaker), or a GFCI if you are American (ground fault circuit interrupter). In North American homes they are generally installed at the socket. Which is kinda nice.
Another thing to consider, if you ever redo an install (or part of it), is to use RCBOs (residual current breaker with overcurrent). They fit into a normal DIN slot, so take no extra space in the DC. Then you have individual overcurrent protection for each circuit, and you don’t lose the whole house over one bad appliance.
The downside: Very costly. Imagine paying R500+ for every breaker and doing a house with 10 or more circuits. It’s like 5k before you’ve done a stitch of work 
Noted but eish! In my next life with my new big house. But this does sound like the way forward.
Sorry to resurrect this!
I bought a white elephant yesterday - a treadmill for me and the wife to get in shape. After assembling it, I proceeded to plug its kettle cord into into a heavy duty extension cord and then into a normal house socket and it immediately tripped the earth-leakage.
I thought bummer, reset it, and plugged it into a different extension cord and into a different socket (on a different circuit) – same thing. So I phoned the supplier today and asked them to come and pick up the machine. They did give me some pushback – asking how sure I am it is not my house.
Thing is - I do notice that treadmills do tend to cause earth-leakage problems. E.g.
Treadmill problems again… | Talk Electrician Forum (electriciansforums.net)
treadmill is tripping power in house | DIYnot Forums
Treadmill tripping GFI (RCD) | Electronics Forum (Circuits, Projects and Microcontrollers) (electro-tech-online.com)
Treadmill ground fault : electricians (reddit.com)
That being said, most of these occur at motor switch-on, mine occurs when I literally just plug it in (the switch on the unit itself is still off). Can I safely fight with the supplier and tell them that it isn’t a problem at my house?
All you need to establish is that the earth leakage trips at its designated mA setting…
1 Like
I bought a laser printer a while ago, and it started tripping my RCD. There is nothing wrong with the printer though. The printer added a small amount of additional leakage (which is normal, anything with a switch mode power supply or some kind of EMI filtering will have a small loss to earth) and it was the last straw that broke the camels back.
In other words, there is a risk that it’s not the treadmill on its own.
These days I have an earth leakage clamp meter, which makes this easy, but before I had that, I literally put a multimeter into the earth line of a small plug+socket cord I made, and carefully measured on the AC current line my various appliances to see how much each leaked.
Of course it is a little hard to measure if it continually trips the RCD, but if you have a multimeter that measures nice and fast (some Fluke models with the bargraph display for example), you may be able to get a peak reading.
Something else you can also do to get a reading (but obviously, keep kids away, don’t touch the darn thing, etc etc) is remove the earth wire temporarily, then measure from the earth wire to neutral (which should be bonded to earth if your house is correctly wired) to see how much current is flowing. If that measures more than a few milliamp (I’d say 3mA max), return the appliance.
Edit: Here is another example, from the UK, where someone uses a normal clamp meter to do the same thing.
OK so the distributor’s technician was here today.
He actually found an issue - a control cable pinched in between two brackets, bolted tichtly together (basically my fault for not checking where the cable went while assembling the machine). So he proceeded to disconnected it but the treadmill still tripped the RCD.
He tested a variety of other things and eventually he disconnected the earth - at which point everything worked fine (even the pinched cable).
He then warned me that it could be the house’s earth - and that this could simply be pushing it over the brink - but said he would take the machine back to their offices and test it there. So I couldn’t really take any measurements as @plonkster suggested.
I did take a couple of resistance measurements between the earth + live and earth + neutral and live + neutral and saw some resistance in the mega Ohm range, but Jaco suggested we should try a Megger next.
The technician suggested that we could disconnect the earth (saying there’s no exposed metal), but I’m not comfortable doing that.
If they come back tomorrow saying there’s no issue, I’ll measure the current between earth and neutral.
1 Like
If it is anything like ours (Trojan brand), it’s an absolute mission to move it. Ours doesn’t fit through a standard door. You have to remove the control panel, and then it will fit sideways through a door if you tilt it 45° up. Of course it not only had to move up the stairs, it also had to be moved from one room to another in the last year…
Luckily in this case it was close enough to a sliding door and there were two technicians - so they pretty much just carried it out. I hope it survives the trip on the back of the flatbed though…
We have stairs in the house. Gives the same exercise without power requirements and does not need to be moved (I hope not!) 
Pretty sure the neutral is electrically connected to the device’s earth if it trips as you plug it in.
Easy to establish with a meter.
The E/L should trip.
Disconnecting the Earth is not a fix, it just means you have no E/L protection on that device. It is a dangerous practice.
I agree!
So after they took the machine, they did some “tests” and “calibration” and when I got it back, it no longer tripped the EL.
I immediately opened it up to see if they just disconnected the earth, but lo and behold, all was connected.
Works well now!
Hi,
I seem to be having the same issue, in October i had a 5kw inverter installed with 5 455w solar panels, my earth leakage is tripping and is ridiculously annoying, ive had my installer come out a number of times but he cant seem to find the issue, i think it could be a neutral issue due to fact of when it trips. When power is being generated from the panels and grid power is available there is no trips, when i switch over from solar to grid power there is no trips, the tripping issue only occurs at night when the inverter is feeding battery power to my house and is extremely intermittent, some nights it can trip twice other nights it can trip over 10 times. I do not have any surge proctor multiplugs, I have replaced ever circuit breaker as well as the earth leakage on the DB board, the installer has redone the wiring on the DB to make sure a Neutral isnt crossed over, ive checked all the plugs in the house to make sure all are wired correctly and have a tight connection to make sure it isnt arcing, i am running out of ideas can anyone suggest anything? Its becoming a huge issue and is damaging my appliances, ive already had to replace my kettle, microwave and TV
Waheed, my main suggestion remains to split your DB into sections, each with their own earth leakage (or RCD). Of course, I assume the installation is checked and we know for certain the insulation is good, there is no real leak. Splitting your installation into more than one section, with its own RCD, really helps a lot. You don’t lose the whole house when there is a fault, which is already a big improvement, and by halving (or more) your standing leakage it often completely solves the problem and provides massive room for future appliances. Since the DB was already redone, adding this one additional step (for the cost of an RCD, <R500) ought to have done the trick.
The other thing that I see becoming standard in the UK now, is you really must have a type-A RCD. The cheap type-AC is really no good for modern appliances with an SMPS. So when you split it, use decent type-A RCDs.
1 Like
RIP Type AC.
ZA regulations will no doubt catch up sometime. ( Or maybe they already have, and blissful ignorance prevails ( See who is listed as responsible for fitting the correct RCD in the Solis test report below).
Type A is able to deal with a pulsating DC offset + standard AC (Read: Bridge rectifiers as part of the load, also the minimum specification for LF inverters that I have seen. Its weakness is being rated for nominal frequency ( 50Hz), devices with a variable frequency element aren’t really suitable).
Superior to Type A are Type F and the Type B.
Type F is 1Ph Type A plus sub 1kHz AC leakages. (Read: 1ph VFDs, 1ph UPS, heat pumps, & my minimum recommendation for a 1ph HF PV inverter when Type A is allowed ).
Type B is Type F plus smooth DC offset capability (1Ph & 3Ph). (Read: Some HF PV inverters, 3ph VFDs 3ph EV chargers).
Other types are available with higher frequency capability, but I don’t think standards are agreed yet. ( B+ & Bfq)
More capable RCDs must be fitted upstream of lesser RCDs, not vice versa.
I recommend Type F over Type A. Even when PV inverter manufacturers specify Type A is allowed (see below). Type F is a recent introduction:
The new regulations actually talk about requiring the appropriate Type RCD.
Not necessarily Type A, ( or Type F) although Type A would tend to fit the bill for common-or-garden applications.
However, we solar enthusiasts are not the bog standard norm with our fancy high-frequency PV inverters, VFDs and inverter-type aircons.
Not necessarily for us, is the lowly Type A.
Taken from a European document:
“Where an electrical installation includes a PV power supply system without at least
simple separation between the AC side and the DC side, an RCD installed to provide
fault protection by automatic disconnection of supply shall be type B according to
IEC 60755, amendment 2. Where the PV inverter by construction is not able to feed
DC fault currents into the electrical installation, an RCD of type B according to IEC
60755 amendment 2 is not required.”
I include a cutting from the Solis inverters’ approval for use in South Africa:
All is not lost, I have seen, for example, SMA state:
Fronius & ABB(Fimer) are probably also specifically excluded. (Do your own due diligence).
An interesting link:
So do I, but the last time I checked my eyes watered at the price… and I bought a type-A instead.
ABB/Fimer inverters have a monitoring unit for earth leakage, so a type-A is suitable in such a system.
Fronius has an RCMU too, so also don’t need the fancy RCD, Correction, it seems the Fronius RCMU allows quite a bit more than 6mA, so a tad unclear.
Here is a paper from Schneider that states that an RCMU is not a substitute for Type B when galvanic separation is not present.
From the conclusion.
Yeah, that’s also what I picked up which led to me correcting my post. I know my ABB PV-inverter constantly checks the leakage and displays it on the screen as the Ileak value, but I would have to make sure if it protects to under 6mA.
Edit: So the ABB documentation recommends a 100mA earth leakage, which is an indication that it can leak a lot more than 6mA. It does however also react to a sudden increase of 30mA or more, but I doubt that is good enough. A constant DC component through an RCD is not a good thing, as I recall.
1 Like
It isn’t a good thing. It will saturate magnetic cores and render them blind. “Blind” means they will fail to operate when a real fault occurs.
Type B can deal with a constant DC offset without being blinded; lesser types cannot.
Source link:
1 Like
Okay so this is making me worry… how do I know whether my RCD is working as intended? Is it enough to just press the button and see what happens? Or will that not necessarily test a small enough (but still sufficiently deadly) current to pick up on issues like a DC component?