At the beginning of this year, we experienced a really heavy thunderstorm here in Assynt. It caused quite a lot of damage across the parish, to power, phone and internet connections. We got off relatively lightly, and as noted in my blog post about it at the time, although our off-grid power system was best switched off, the only damage we suffered was to electrical goods in the house, rather than to our power system itself.
Our power system is remarkably robust and stable. We rarely have any issues other than having to run a generator occasionally in winter when solar power is virtually nothing, and the wind may not be blowing for a few days, so the batteries need topping up from a different source, but that is an operational issue, rather than one of the system itself. But when something does go wrong, it re-inforces the fact that, when you are off-grid, you are responsible for your system. That means, if you don't have the expertise yourself, calling in someone, or, if you do have the expertise, rolling up your sleeves and getting things done. There was a possibility that the inverter, the unit than converts low voltage DC from the batteries to conventional 230v AC for the house, was damaged by the lightning. Until now, we have had a small 300w unit, nothing like as industrial-strength as the main inverter, as a backup. However, when the lightning struck, I realised, with my heart in my slippers, that if the inverter was damaged, the little emergency one would probably not be powerful enough to start the fridge.
Our power use is quite frugal. During the day, running our laptops etc, the average power draw at 24v is around 6-8A, about 150-200W of power. However, items like fridges, which have motors, need a big spike of power to start, even if, like our fridge, it needs just 50 or 60W of power to run. I have seen the ammeter spike to 50A, which is 1.2kW, for a split second just as the fridge starts. Even though the emergency inverter might be able to cope with such a surge, it would be touch and go.
Now the inverter we have been using for the last 11 years has been fantastic. It's a Victron 1200W unit, which has been humming away in the battery shed with nothing other than the fridge starting bothering it. When the fridge starts, it makes a noise - dahing - and then takes up the slack and continues. One can help the inverter with these surges by ensuring that the cables from the battery bank are as short as possible, and also as thick as possible, to reduce losses in the low voltage side of the circuit, and, to some extent, I have done this.
But we have long said that it would be nice to have just a little more headroom with the inverter, so that running items of, say 1000W, would be no big deal. The issues arise really on the low voltage side of the circuit. If the inverter is producing, say, 2kW, that means that over 80A is being drawn from the battery bank. At those amperages, any defects in the joins, or any imbalances in the battery bank are soon found out. It's close to the amperage drawn for just a few seconds when starting a car. Now we should have thought about this, and installed a 48v system, which would halve this amperage draw. But my thinking, at the time, was that there were other advantages, such as the amount of equipment available for 24v car and lorry systems, that may make maintaining a 24v system easier in the longer term. So it was a choice that was made with eyes wide open. Still, for occasional high power draws, I thought our system would cope.
So we decided to go for a 2000w inverter, staying with Victron as we have come to trust the way they work and the way they are built. Then came the usual round of internet research to find suppliers who had the devices in stock, and who had them for a reasonable price. It also turns out that Victron are now offering this range of inverters with built-in bluetooth, for a certain amount of remote monitoring, which would be useful. The newer model seemed quite a bit cheaper than the old model too, though that fact made me nervous. The inverter is such a critical component that I would rather pay extra for the best device I can find, but perhaps new manufacturing techniques make the new one just as robust and cheaper too. I duly chose a supplier who claimed to be able to deliver to us in Assynt (honestly, one would think we're on the ISS, the way some suppliers carry on about delivering to the Highlands). There were some issues with the supplier, but eventually the unit arrived. And of course it was the wrong one. It was the old style, the one that was more costly, but that also the range with which we are familiar. I came to an agreement with the supplier. They would lose money by having to collect the wrong device and send out a replacement, while I was happy to take the old style unit, so we called it quits.
I also ordered, from a different source, new lengths of battery cable, substantially thicker than the previous cables, so as to be able to cope with the extra power demand, if we ever asked for it. While I was about it, I decided to add an extra fuse too. The inverter has a built-in fuse, but if this could be saved by having an external one, should Something Bad happen, all to the good. I could also choose to ensure that the cabling was good, as it is now four years since we replaced the battery bank, and the connections have been left to themselves over that time. Here is the Big Red Button, although it is blue, that isolates the battery bank from the inverter, the cable leading to the new additional fuse and then to the inverter.
And here is the new inverter, looking pretty much the same as the old one, but rather taller and a lot heavier. The white ring dangling off the black negative lead is a sensor for an ammeter, the display for which is inside the house, in the kitchen, so power output and input can easily be monitored at a glance. The final part of the jigsaw was to re-wire the RCD earth leakage protection units, which I believe may have saved worse damage during the thunderstorm in January.
While we were about these changes, I decided to add some better instrumentation. Actually, the instrumentation we have had until now has been rudimentary, restricted to two moving-coil ammeters in the battery shed, so pretty inconvenient. If I wanted to check the amperage of the outgoing power, I had to clip on a hand-held ammeter. These were supplemented by the really important instrument, a digital voltmeter in the house. This is important to check that the batteries do not fall below 50% state of charge, which can be estimated using voltage. In our system the voltage should never fall below 24.0v. So in addition to the little sensor described above, I also installed two ammeter shunts on the solar and wind inputs to be able to monitor them better. I chose not to use a shunt for the outputs, because I did not want two more joins in the high-amperage cable. Apologies for the poor images, but you should be able to see what's going on. The top image shows the batteries at 30.0 volts, so nicely charged, and in a while, the control systems will reduce that the a "float" charge at around 27.6v. The green number is the output amps, the reading from that little white sensor in the picture above. So at the time the picture was taken, the system was using 30.0vx5,2A = 156 watts of power battery power. Actual AC power inside would be a little less than that. The inverter is around 94% efficient. The lower image shows the power coming in from the wind and the sun.
So, back to where we started. The new inverter is now in place, and the old inverter is in a box, along with everything it needs to function. I had thought of wiring it in as well as the new one, but then it no longer is an emergency spare and is just part of the system, and if an event occurs which I can't foresee, there may be some unnecessary risk.