Things That Go Wrong
In this post I’ll try to describe a few of the things that didn’t quite go as planned.
I’ve paid my school fees and I hope you won’t have to.
Once I got the whole DB completed, it was time to fire it up and hopefully live happily ever after with no more worries about loadshedding. My best mate and I cracked a cold one and sat in the garage admiring my handiwork.
It wasn’t long before the RCD tripped for no apparent reason. It was reset and it held.
Another ten minutes and a beer later, the RCD tripped again. I tried to reset it, but it kept tripping. Obviously there was a fault.
Now for the fault finding; switch everything off; reset RCD; reset circuit breakers one by one to try find the fault. Everything stays on! After about 20 minutes the same thing happens. At this stage just about everything being fed from the inverter was off, barring the light circuits, so it seemed as if there was a fault with one of the light circuits.
Another reset and everything stayed on. About two hours later and I’m sitting in darkness again! By now I’m at a bit of a loss and it being 9pm on a Sunday night, I decide to switch off what appears to be the offending light circuit circuit and work with the lighting by table lamps and bedside lights fed from the plug circuits.
This worked for the night and the next morning I switched everything back on. Nothing tripped for the whole day and the lights worked until around 10pm that night when the RCD tripped again. Once again, I did what I did the previous night. The next morning I called in a favour from friend who is an electrical contractor to see if he could find the faulty light circuit. After a couple of hours of crawling around in the ceiling and testing no fault was found.
In order to get rid of the problem, at least temporarily, I decided to use one of the spare sections on the DB and changed things around a bit:
- I changed the RCD to a normal 2-pole circuit breaker;
- moved the RCD and plug circuit to what was a spare section;
- and then added a new circuit breaker to feed the new plug section.
No more tripping!
The likely causes of this problem are as follows:
- Faulty fluorescent fitting on the lighting circuit. I’ve subsequently changed the only fluorescent fitting I had in my garage and will re-test the circuits.
- Damaged neutral wire somewhere on on of the lighting circuits. Hard if not almost impossible to trace.
- Floating neutral on the inverter output. Possible but unlikely as the RCD behaves properly with just the plug circuits connected.
Some people recommend bridging the earth and neutral at the inverter output to overcome the floating neutral. I haven’t done this yet.
In my haste to get the system running, I didn’t double check my battery charge settings and left them at the default settings for Flooded Cell batteries.
The default charge current is 60A and for a battery bank my size, the recommended charge current is 20A.
I didn’t set the Depth of Discharge (DoD) properly and discharged the batteries too deep the first couple of times. These “cheapy” Excis batteries are NOT designed for that. They are “High Cycle” and not “Deep Cycle”.
The result of this is that I now have a battery bank that only manages to give me around 45-minutes of backup even when there is hardly any load on them.
To compound matters even more, I decided to go smaller than the recommended battery capacity of 200Ah as a minimum, and installed a smaller 100Ah battery. I may be wrong, but the way I see it is that the inverter sees the discharge rate and underestimates the remaining capacity and shuts down even though the battery voltage under no load shows very little discharge.
My solution to this is to do what I should’ve done in the first place: install the correct size battery bank using batteries that are designed for this.
I’ve ordered 4 x Omnipower Gel/VRLA Hybrid 260Ah batteries from Sinetech. I’ve also ordered a Victron BMV-700 battery monitor so I can see exactly what is going on with the batteries and not make the same stupid mistake again.