PV Array Wiring
Having finally got around to redoing the array wiring, here are a few photos of combiner, wiring and grounding. The objectives for this little project were to:
- (finally) bury the home run cables, previously laid on the ground
- fit a proper fused combiner
- add ground bonding between the rack and the system ground
- inspect for and reduce where possible dissimilar metal corrosion.
After gutting the old 4mm2 pigtail wiring and temporary combiner, new pigtails were crimped and fitted as before, by taping the male and female halves together during the entire wire up. This allows you to get the cable run lengths right and wire safely on a non-live system. With the new combiner in place, this system is simple and effective, it avoids arc damage to the mc4 connectors, and shock hazards to yourself. So, contrary to old folklore, you don't need to cover the array. But you do need to work methodically, because unfortunately the 4mm2 wiring on both the modules and the pigtails are all colored black, which leaves quite a lot of room for error. (Rumour has that the next round of codes will require red and black array cabling.)
This time around I wanted to make the positive and negative pigtails the same length and keep them as close together as much as possible. While internally the modules themselves break this rule, at least my bit is now done right, the point being to minimise induced differential surge voltages. As for the three strings there is no reason at all to keep each strings pigtails the same length, for the simple reason that the internal resistance of the modules is reasonably high. You do however have to deal with the rather long and ungainly 4mm2 fly leads on the modules themselves. The best that I could seem to do was this small coil and figure eight pattern and some zip ties.
Keeping positive and negatives together helps reduce transient spikes
On the module earth side of things, I upgraded to the stainless fittings from NAWS. They come with the correct toothed stainless washer, and with lots of petroleum jelly thats about as good as we will get.
This upgrade we took the opportunity to bond the array into the main system. Previously there were two separate earth rods, one for the system, and one for the remote array, as per previous code iterations. The latest code requires the two rods to be bonded, and so i ran 6mm2 to connect the respective ground buses.
The home run cable and conduit was all housed inside some old four inch drain coil and buried to 450mm in a hand dug trench. A string line was used to get the trench straight (which is always a good thing if you ever have to pull more cable). I left a pull cord in the outer conduit ( future monitoring maybe).
Using lots of colored electrical tape helps keep polarity clear. I doubt it would be a happy sight to swap the polarity of a string, that wouldnt be nice.
After the pigtails were all terminated in the combiner, with the breakers all off (both combiner and main disconnect), the 3 pairs of mc4s were connected, and the three strings were tested for voltage and polarity using a DMM set to volts.
All being good there, i went ahead and refitted the homerun cable, as best as i could with my odd collection of conduits into the various combiner knockouts.
Just to clarify the grounding which isnt clear from the image: the new bonding cable runs direct from the disconnect ground bus to the combiner ground bus. This is the most corosion resistant way of doing it.
Homerun cable on dual polypipe / drain coil conduits (what i had to hand)
More or less completed upgrade.
Just a final point about ground rod connections. We have three rods. The third one is for the radio mast, (which serves as lightening protection for the array), and is completely independent. The main house rod is a typical NZ galvanised compression fitting (copper plated rods aren't available here). The array and mast are both crimped lug and stainless nut and bolt arrangements, which i greatly prefer. See pictures.