Having my set-up settle down over the last 3 months I am interested in what people are setting their Max SoC and Min SoC over the Winter and Summer. Seeing that in the Summer keeping the battery at 100% is not good for long-life. I am on Octopus Flux and have 100% max and 10% min but I want to ensure I look after the health of the batteries. Interestingly I was getting charge rates up to 6.8 kWh and have now set the max charge amps to, the EP11 documentation recommended, 13.5 Amps. It seems that in Summer something around max 85 - 95% and min 15% would be a start. I also see that a full charge and discharge to 10% is a good idea 1-2 times a month. What have people set and why?
Many thanks
John
Firstly it is important that the battery periodically hits 100% and 10% - at least weekly for the 100%, and at least monthly for the 10%. This is needed for the voltage calibration (at 100% and 10%) and balancing (which takes place at 96% and near 10%).
Typically I run up to about 85-90% on most days, and down to 15-20% on most nights, but I will hit 100% / 10% on days when calibration is needed, and also on days where I know I will need the power, or the PV forecast is particularly low (so generally speaking I cycle more in winter than in summer).
At least on my battery pack, from a warranty viewpoint, the throughput warranty will cover 92% of a full cycle daily (a full cycle being from 10% to 100% and back down, so 92% of that would be maybe 18%-92% for example). 16.5kWh nominal capacity, 15kWh usable capacity, and the throughput warranty covers about 14kWh daily charge and discharge. So I don't think there is much to be gained by doing **less** than that since calendar degradation will also affect the battery regardless of how much you use it. So I aim to average that much usage over the long term. I use home assistant to track the warranty usage, do the calibration cycles, etc.
If in doubt about automating the calibration cycles, it is better to simply use the full range every day rather than doing it manually (which will result in forgetting, and the loss of voltage calibration and balance). While there is a small advantage for longevity of limiting the range on LFP chemistry, it is minimal compared to NMC and other chemistries which are much more affected by SoC extremes.
Other factors which affect LFP battery degradation are:
* Cell temperature (it doesn't like getting too hot, so I have fans which will turn on if the bms_cell_temp_high goes too high).
* Charge speed (so I control the charge speed to charge at the minimal speed required to hit my SoC target at the target time)
* Calendar degradation (regardless of how much you use the battery).
The BMS will also make adjustments as needed to charge speed and so forth based on the temperature and SoC, I just go one step beyond that. Controlling the charge speed and turning on the fans seems to be very effective at keeping the cell temperature in the reasonable range, even in the middle of summer.
There are some references here:
* Calendar aging https://www.mdpi.com/1996-1073/14/6/1732
* Insights for understanding multiscale degradation of LiFePO4 cathodes https://www.sciencedirect.com/science/a ... 1722000283
* Degradation of LFP batteries at 100% SoC https://iopscience.iop.org/article/10.1 ... 111/ad6cbd
But overall I think the most important thing you can do, is to stop the batteries getting too hot in summer by proper ventilation and limiting the charge rate.
Typically I run up to about 85-90% on most days, and down to 15-20% on most nights, but I will hit 100% / 10% on days when calibration is needed, and also on days where I know I will need the power, or the PV forecast is particularly low (so generally speaking I cycle more in winter than in summer).
At least on my battery pack, from a warranty viewpoint, the throughput warranty will cover 92% of a full cycle daily (a full cycle being from 10% to 100% and back down, so 92% of that would be maybe 18%-92% for example). 16.5kWh nominal capacity, 15kWh usable capacity, and the throughput warranty covers about 14kWh daily charge and discharge. So I don't think there is much to be gained by doing **less** than that since calendar degradation will also affect the battery regardless of how much you use it. So I aim to average that much usage over the long term. I use home assistant to track the warranty usage, do the calibration cycles, etc.
If in doubt about automating the calibration cycles, it is better to simply use the full range every day rather than doing it manually (which will result in forgetting, and the loss of voltage calibration and balance). While there is a small advantage for longevity of limiting the range on LFP chemistry, it is minimal compared to NMC and other chemistries which are much more affected by SoC extremes.
Other factors which affect LFP battery degradation are:
* Cell temperature (it doesn't like getting too hot, so I have fans which will turn on if the bms_cell_temp_high goes too high).
* Charge speed (so I control the charge speed to charge at the minimal speed required to hit my SoC target at the target time)
* Calendar degradation (regardless of how much you use the battery).
The BMS will also make adjustments as needed to charge speed and so forth based on the temperature and SoC, I just go one step beyond that. Controlling the charge speed and turning on the fans seems to be very effective at keeping the cell temperature in the reasonable range, even in the middle of summer.
There are some references here:
* Calendar aging https://www.mdpi.com/1996-1073/14/6/1732
* Insights for understanding multiscale degradation of LiFePO4 cathodes https://www.sciencedirect.com/science/a ... 1722000283
* Degradation of LFP batteries at 100% SoC https://iopscience.iop.org/article/10.1 ... 111/ad6cbd
But overall I think the most important thing you can do, is to stop the batteries getting too hot in summer by proper ventilation and limiting the charge rate.
I was hoping for more information but thanks WyndStryke. Wondering if Will has any views on this?
Pretty sure his view is to use the full capacity of the battery, 10% to 100%, and don't worry too much about the slight degradation differences, since leaving unused capacity on the table is a waste. Whereas I tend to hyperfocus on those slight differences.
There's a lot of info in those links, what more were you looking for?hoping for more information
Sorry I didn't mean to be dismissive and I have implemented your minimal charge rate to achieve target in the target time so thanks for that content. I wondered what more people had adopted but I will work in the calibration cycles in automation. BTW what temperature do you have your fans kick in?
43c for the big inverter fan, and 30c for the battery fan. The inverter will de-rate at 50c I believe. The battery heater mat turns on if the temp goes below 12c.
However, I also have put 2x 140mm USB computer fans onto the inverter, and since I did that, the inverter temperature has never exceeded the threshold needed to turn on the big fan, even in mid-summer. In winter the USB fans are upside-down (to blow the warm inverter air down towards the battery), but for most of the year they point upwards.
The USB fans are not controlled, they just run 24/7 (they're very low power, but also very effective due to their placement just above the inverter heatsink fins).
However, I also have put 2x 140mm USB computer fans onto the inverter, and since I did that, the inverter temperature has never exceeded the threshold needed to turn on the big fan, even in mid-summer. In winter the USB fans are upside-down (to blow the warm inverter air down towards the battery), but for most of the year they point upwards.
The USB fans are not controlled, they just run 24/7 (they're very low power, but also very effective due to their placement just above the inverter heatsink fins).