HomeBrewBattery

It’s been raining quite heavily for 2 days, and this evening the battery was down to only 3% state-of-charge. The lights in most of the house run off-grid from the battery, and so I needed to get them some non-battery power. Luckily I’ve recently connected the grid input to my inverter. It was time to test the “bypass” mode where the inverter feeds grid power through to the loads.

It works beautifully, and interestingly in bypass mode the inverter seems to provide a charging trickle of about 1A to the battery even when configured not to charge the battery from the grid. After 10pm when my electricity switches over to cheap off-peak rates I got to test one more inverter capability for the first time: grid charging of the battery! With numerous packs now in my battery build I was able to configure the utility charging current up to the max setting of 60A. This means charging at about 1.7 kW.

The wet and cloudy weather is forecast to hang around for a few more days at least. This grid charging feature means that I can play the merry game of electricity arbitrage: store up cheap off-peak electricity from the grid to use later during expensive peak-pricing times. It’s not quite as good as my own free off-grid solar, but it’s still a way to save money – as well as helpfully “load shifting” to make the grid run more smoothly.

Late last year I bought some used 18650 cells that had already been somewhat tested. This has dramatically increased my “good” cell count in recent weeks, and I’ve been able to put the second pack halves on the first good string – taking my operational battery to about 1kWh in capacity.
In the design process I’ve dropped from 20 cells per full pack down to 18 so that the packs are short enough to fit double deep in my battery cabinet (keep an eye out for a design discussion post coming soon). Obviously this helps reach the pack count sooner, but processing pre-tested good cells really slowed my accumulation of “poor” cells.
Well, on the weekend I simultaneously reached the next half-pack set of good cells (this is how I construct the packs so it makes sense to count in this quantity) and the first complete half-pack of poor cells. When I get these soldered up the modular design of my battery will be visible.

Oops!

I’m having a bit of difficulty getting my packs balanced to start an experiment to see how well they remain that way. I think an easy way to do this would be to connect the 7 packs in parallel and leave them overnight to exactly equalise voltages, but this will require some new “bus bars”. I grabbed a wire this evening and held it roughly in place to check what length I need, and then POP, FLASH!

I’d absentmindedly touched the bare wire onto my pack connectors for two packs, shorting them out and instantly blowing all their fuses.

My battery project time today has been used up with replacing 20 fuses.