Help with earth leakage tripping

[8IN4RY]

Hi Guys and Gals!

 

I am hoping that you can assist me, as I am at my whits end.

 

A friend recently asked me to assist with his solar installation.

Just some background info on the installation. The installation is normally fed from a 3 phase Eskom supply. And then during a power outage break before make change over switch is used to change over to the solar inverter. The inverter is single phase, and as can be seen in the attached schematic the 3 phase at the bottom of the change over switch has been bridged out, so that no wiring modification had to be done in the existing DB.

 

The problem that occurs is that the earth leakage Q2 in the attached single line keeps on tripping when the supply changeover switch is selected to the solar inverter, but when the supply is selected as the normal Eskom supply it does not trip.

 

Things that i have already checked. The voltage between the Neutral point and earth point on the inverter is 0.2V. The inverter is solidly grounded.

Thank you for looking

Kind regards

Morne

 

As build DB.pdf

[___]

I looked at it and I can't see a problem. I'm not an electrician, but tripping the ELB means you have more current going in on the one leg than is coming out on the other side and I don't see it.

You didn't say what inverter is used, but I have an idea. Earth and Neutral will be bonded on the Eskom supply side, before entering your make-before-break changeover switch (which throws all three phases in parallel I see). You aren't allowed to make your own bonds on the consumer side, usually. When you disconnect the grid, you also lose the earth/neutral bond and earth is now floating. You should add your own earth/neutral bond after the inverter but before the changeover switch, so that earth and neutral is also bonded while on the inverter, but both bonds are never active at the same time (that is against SANS).

It still doesn't make complete sense to me how that would cause an imbalance, but I don't see the inverter-side bond -- unless the inverter itself implements it -- and that might have something to do with it.

Not all inverters can be bonded in this manner, some can, and some do it out of the box.

[___]

51 minutes ago, 8IN4RY said:

The voltage between the Neutral point and earth point on the inverter is 0.2V

Usually I would expect something more, something like 90V or so. That's what happens at my place if earth is floating. At 0.2V it looks like a measuring error, though perhaps go down to the millivolt (or lowest) range on the DMM to confirm. There should be no potential difference between neutral and earth, if the bond is proper.

[8IN4RY]

I also had the thought that it might be a floating neutral causing the problem. I asked my friend to put in a earth rod and bond the neutral and the earth and this is when he measure the voltage of 0.2V. The 0.2V might indicate that the bonding was not done a 100% correctly, and that there is still some resistance between the neutral and the earth point. I will get him to look at the bonding point an to make sure that is is done correctly.

I don't have the info of the inverter with me at the moment, but i will get it and post an update here.

THanks for the help.

[___]

Even without the bond, it's still more likely that the elb will fail to trip rather than trip too easily. All I can think of is turning off all downstream loads/breakers so there is nothing going through the elb, check that it stays on, then bring up loads and see if there is a pattern. I've had since weird stuff in the past, I had a case where putting the tumble dryer and the electric fence energiser on the same plug point trips the elb, but putting them on separate plugs and breakers and it works fine. Which makes no sense, everything goes through the same elb? If there was a leakage it would leak regardless of where it was plugged in, right? Some months later the elb itself failed, and after replacing it the strange behaviour mysteriously disappeared. I can only think some kind of distortion or something induced enough magnetism wherever to trip the breaker.

Sent from my GT-I9195 using Tapatalk

[Carl]

It could also be worth while changing out the earth leakage unit (assuming it is type A) and installing a Type B.

Your inverter could be generating a DC and / or High Frequency component on the output waveform.

Type B earth leakage units are less susceptable to these components.

From ABB manual "For example, a Type B RCD with IΔn = 30 mA, when direct residual current without ripple is applied, has a maximum tripping threshold of up to 60 mA to take into account the lesser danger of DC current. For the same reason, for a 400 Hz frequency residual current, the maximum tripping value allowed is 180 mA."

[___]

6 minutes ago, Carl said:

Your inverter could be generating a DC and / or High Frequency component on the output waveform.

I recall reading in NRS097 that there is a limit to the DC component that is allowed for grid-tied operation. The mere fact that it is mentioned obviously means something :-)