Foot

@LawnDolar yes please, that will be a great help.

Thanks, this is a really helpful and informative article.
I think I will give the installer the opportunity to rectify the problem by sending out a qualified to IE or MIE to inspect and test and issue a new CoC, before I take matters further.
The lesson learnt is to ask upfront to see their credentials before commencing work and not be so trusting.

Not necessarily if major changes was done to your AC electrical system( Db board) then yes but if only upgrade was done to your inverter system then they only need to do a supplementary CoC with test report then you attach it to your original COC that was issued in 2010. The only time you will do a full COC if major electrical changes was made or should you sell the property. It all depends on the scope of work done. But the IE or MIE should know the diffrence.

Does he need to be there, though? If he got all the test results etc, surely that's fine. What exactly more is he supposed to do when he's there? If it's a reputable installer and they didn't forge the results, then surely that's fine. I would sign off on results and pictures unless I think the installer has forged everything.

Best people to ask are these guys https://ecasa.co.za/

This seems to be standard practice, not only for electricians, but for all trades. The same argument applies to building inspectors, plumbers, civil engineers, assessors, etc. A proper inspection of installations on most sites by a duly qualified and ethical person normally results in a rework and sour relations

I am not an electrician, but logic and common sense tells me this way of coc sign off cannot be legal at all!! So this "Master Electrician" believes anything they tell him?
Scam way of doing things - imho

Based on these readings when the grid is on and inverter of you have a lot of high impedance earth faults. These need to be found as it seems none are actually inverter related. Splitting the inverter load over 2 E/L is not the solution 
With the inverter delayed by 5sec and the E/L tripping from the mains return is the clue to check the whole installation and I doubt changing to other types of E/L is the path forward. 

hi all
so i am pulling my hair out in frustration and getting no where in the process ..
to cut a long story short . I recently replaced a cheapie 2.4kw ( solarwise axpert clone)  inverter with another cheapie 3.5kw fivestar inverter  for 2 main reason.
- its quieter
- can connect more wattage of PV to it
and before any one says anything , i know its a cheapie for a cheapie and that is probably the problem 🙂

It was a direct replacement with no wiring modifications on the AC input or output side.  After the replacement the earth leakage on the output side of the inverter for the essential load was tripping after load shedding ends and it tripped only when the inverter was running from the grid to supply the load and not when there is PV available.
Recently it started tripping randomly whenever it is running from grid to supply the loads.
So it obviously has something to do with the Grid powering the loads through the inverter.
 

However I can not replicate the tripping if i switch off the AC input to the Inverter and then switch it on again.

To eliminate sensitivity as the problem i have tested the earth leakage switch with a tester and it trips between 25 and 30ma .
I have also measured the voltage between Neutral and Earth on the output side of the inverter and i am reading in the millivolt range both when the Grid is supplying the load or during load shedding and the inverter is in "island" mode powering the load from battery.

As for the wiring side of things. The inverter draws its AC input direct from the mains and not after an Earth leakage switch. On the output side of the inverter it runs through a change over switch , then the earth leakage and then I have the homes lights and plugs CB's connected.
 
I have not installed any neutral earth bonding as the previous inverter ran perfectly fine and i am not entirely sure on the topic and how to implement it.
 

Interestingly when Load shedding ends and the inverter is powering the load from Grid , the earth leakage does not trip immediately but rather about 30 seconds in and then for a few seconds it is not possible to recover the earth leakage switch as it keeps tripping out as you switch it up.

I have by passed the change over switch as part of the fault finding method and checked the wiring at least 100 times and am 100% sure that there is no cross connection of any neutral from the non essential to the essential side. The more i check it the more blind i seem to become. Maybe i need a 2nd pair of non biased eyes.
there are no surge plugs used in the house and i have also disconnected all plugs from the wall sockets and the earth leakage still trips.
 
 
My next step is to disconnect all but one of the essential load circuits from the DB and their corresponding neutrals and wait for load shedding to end to see if the earth leakage still trips.  If it does not then start by re-connecting 1 circuit at a time , wait for load shedding to end. then connect the next one.. this could help with the following :
- isolate the circuit with that actually has earth leakage taking place
- or it could prove nothing as the total sum of earth leakage over all the circuits + the inverter is the actual problem and not necessarily one circuit considering that the earth leakage never tripped with the previous inverter.
So I will consider splitting the essential loads over 2 earth leakages but before i spend the monies surely there is something more i can test? Could the inverter actually be faulty?

any advice from the masters to the grasshopper?

Earth leakages doesn't "see each other", but they do see EMC spikes in switching events - and not all ELs are created equal. I'll bet if you replace the GEWIS with a CBI one (like you've got on your input), your problem will be solved. Friend had the same issue - worked for him. Very easy fix.

Is this within the SANS code  for wiring. We see how many times this is recommended but with geysers now also on E/L I just wonder about inverters. 

That image is from Annex P:

The actual regulations are written in the wordy sections...

Inverter goes before RCD, if you put it after an RCD, it will cause issues like this. It can also still kill you even if the RCD is tripped, as it is a source, so there is no point in putting an RCD before it. Geysers have to be after an RCD, because plumbers generally make poor electricians.

Remove the feed to the inverter from the E/L. This causes random trips as the 2 E/L see each other.