We have come a long way since we first began experimenting with our off-grid power supply 18 years ago.  During the years that our house was just used for holidays, we could make mistakes with the projects without much impact, and at that time we learnt a lot of lessons.  But in 2008, when we settled here, we needed a better power system.  We chose to alter things incrementally rather than all at once, continuing to learn as we went.  We bought a new set of Trojan T-105 batteries at that time, when we still ran a 12v system, and added a pair of 200w solar panels to the mix.  Then we went 24v, (should we have gone 48v?  Probably...) and the battery bank was reconfigured to allow for this.  During that time, the poor batteries were not treated very well, although we rarely, if ever, took the batteries to below 50% state of charge as measured by voltage.  But I do not think our charging systems ever really fully charged the batteries.  Given their earlier treatment, I am surprised the 12 batteries, in three strings of 4 (they are 6v batteries) lasted as well as they have.  But for the last couple of years, we have only just got a full day's power out of them before they were approaching the 50% state of charge mark.  Clearly they had lost a lot of capacity over the years.  So we thought that this year, in late autumn, in time for winter, we would replace the battery bank.

Then came the wretched Westminster squabble that has taken the UK out of the EU, and the pound plummeted against the dollar.  Now most deep cycle batteries available here are either US or Canadian manufacture, and therefore the price was likely to rise dramatically.  As I thought the new bank would be between £2000 and £3000, this meant quite a difference to the final price.  So we decided to bring forward the purchase.


As Hugh Piggott, Scoraig, has been so helpful, as written about earlier, and as I knew he supplied proper deep cycle batteries, I planned to ask him to quote for a suitable bank.  But Hugh was interested in our use of Trojan T-105 traction batteries, which are designed for use in golf buggies, and even the G-Whiz little electric car used them.  He thought the price point was remarkable, probably because they are produced in large quantities.  So I contacted Tayna to ask for a quote.  An Internet search seemed to indicate that they were competitive on price, and batteries are their main business.  As usual, getting the batteries to us was going to cost an additional arm and leg, but they did what they could to minimise the additional cost.  I ended up ordering the T-105's bigger brother, the T-125, which has a slightly greater capacity, though it is the same physical size.  I ordered 16 of them, planning to run four strings of four batteries, for a total capacity at 24v of 960AHr.  We use about 125 - 150 AHrs a day at the moment, so a fully charged bank should give us 3 to 4 days of power, assuming we do not discharge the batteries deeper than 50%.  If my sums are right, that is a 23kWHr power supply.

But there were other things to think about too.  Battery inter-connects is one.  For such a bank, 12 connecting cable are required, three for each string to connect four 6v batteries to give 24v, and we have four strings.  This is known as a series-parallel configuration.  I also needed new take-off cables for the 24v, another 8, in other words.  Now these cables with fitted ring ends, are wickedly expensive.  The cheapest I could find still meant spending £120 on them. I looked at making them up myself, but adding the tool to do this meant spending £30 more, so I ordered the cables.

The batteries were delivered on a pallet by Caledonian Logistics, all 480kg of them. Yes, the battery bank weighs half a tonne.  They were very carefully packed, with polystyrene protection at the top and sides, as well as a small sacrificial box on the top to prevent the carriage companies stacking anything on top.  Nuts for the connector posts were included, as requested.  And Tayna, understanding the requirement, made sure that the batteries were from the same batch.  The lorry can only get so far up our track, so Mungo the Land River had to collect them.  Here is the pallet on the left hand side of the delivery lorry

TinSlave-113241-13072016.jpg

And here they are fitting neatly into the back of the Landie.  This was the time to check each battery, using a voltmeter to see whether they were looking to be in a similar condition.  Sure enough, the variation among the batteries was low, just 3/100ths of a volt, each one being around 6.2v

TinSlave-113325-13072016.jpg

Meanwhile, I cleared out the old bank in the battery shed, finding all kinds of things I had dropped under the platform on which the batteries sit, as well as a lot of dirt and dust.

TinSlave-114128-13072016.jpg

I also had to alter the strongly built wooden platform on which the batteries sit, which raises them a little for maintenance and prevents them getting too cold in winter from sitting directly on the floor.  The cardboard sides of the battery shed, with bubble wrap underneath, provides remarkably effective insulation for the little wooden battery shed, keeping it cooler in the summer and warmer in the winter.

Slowly, I started lugging each 30kg battery into position, knowing that they are likely to be in place for 7 to 10 years. This time, I placed a sheet of cardboard on top of the wooden platform to prevent anything dropped from being inaccessible without moving the whole half tonne.

TinSlave-114617-13072016.jpg

Once the batteries were in position, I checked them to ensure the polarity of each battery was as I expected, and then asked Helen to check them too - you do not want to make any mistake at all when connecting up a battery bank.  The battery tops are symmetrical, so other than a POS+ and NEG- mark on the plastic tops, but confusingly, some had instructions on one side, while others had them on the other.

TinSlave-113344-13072016.jpg

Eventually, it was complete, looking neat before the cat's cradle of cabling started.  I dipped each end of the inter-connects into a tub of petroleum jelly - good old vaseline, wiping off the excess, before connecting them up, and wiped each connector in vaseline too.  The intention with this is to prevent air gaps in the connection from allowing corrosion to build up.  It is possible that acid vapour coming off the batteries can lead to corrosion at the terminals and the build-up of salts, which alters the electrical connection quality, but here we are also near the sea and the possibility of salt corrosion borne on the winds is high.  This extra step, though some say it is not necessary, may prevent issues down the line.

TinSlave-113350-13072016.jpg

Now the "rules" say you should not use more than three strings of batteries.  This is because any variations among the batteries may result in some getting over-charged and some getting under-charged, which if unchecked gets worse over time.  But Hugh Piggott's pragmatic and practical experience suggests that you make do with what you have, and an occasional equalisation charge, when necessary, should reduce the downsides of the theory.  Also in theory, if you do use multiple strings, the charging and discharging circuits need to add power and remove it evenly across each string, and if you try hard enough, you can find complicated drawings of how you should connect up such strings, on the assumption that even different cable lengths create different resistances which could result in uneven charging and discharging.  I'm not so sure batteries are that finicky, and the at-rest voltage of each component battery of the old battery bank bears that out, as they weren't treated according to the theory.

But to try to approach things as ideally as possible, my intention was to use a busbar of some description, and a length of copper pipe seemed to fir the bill.  This will allow us to clamp charger cables when we need to use the petrol charger in winter.  The ends of some cleaned-up copper pipe have been clamped shut, and bolted through for attachment points for the batteries, the inverters, the charge controllers and the turbine and solar inputs.  This acts as a manifold for all the cabling, and here it is in all its gory detail:-

TinSlave-115549-13072016.jpg

While I don't know if it is necessary, I have drilled two air expansion holes in the round centre section in case the closed-off pipe warms up excessively.  The holes make convenient holders for a multimeter's cable.

And so the bank is complete.  The copper manifolds no longer hang as in the picture, but are held on little wooden stakes attached to the platform.

TinSlave-165501-12072016.jpg

It is inevitable that the cabling ends up looking like a cat's cradle, even though the busbar idea makes things a lot clearer.  Over the next little while I will neaten up the cabling, making new cables for the inverters, controllers and inputs so that they are the correct lengths and can be chanelled more neatly.

So that's the power project's latest phase completed.  Since the beginning of last year, we have a new wind turbine, which performed really well over last winter, a new set of solar panels, providing all the power we need for around 9 months of the year, and now a new set of batteries.  Hopefully we will not need to make any significant changes for a number of years.