EVE 280Ah battery capacity test.

[Dacian Todea (electrodacus)]

I managed to fully charged the battery yesterday so decided to do a capacity test by fully discharging.

It was actually at full charge a few times so I used some loads to both test the inverter with 2800W+ for about an hour or two and use the available solar to heat the house.

Then I stopped the loads a bit to late to get again to full 100% SOC but 97% was close enough to start the test and the 3% delta will be considered at the end.

You can see the energy counters just before resetting and the fact that I got just over 14kWh from the 3.1kW PV array  

as seen from solar graph there where a few clouds but not bad and peak was around 2900W

The start point at 97% with a load around 900 to 1000W (a resistive heater).

I took an one hour pause from discharging here so that discharge will end at a more reasonable hour in the morning.

The inverter idle is quite bad at 55 to 60W but other than that is a great inverter and very capable.

Test was restarted here.

So test ended probably at around 4.9% or so with the declared capacity as 540Ah witch I was already considered as conservative.

The energy counter was exactly at 500Ah about 13kWh but we need to add the 3% deficit at the start so 515Ah or 13.4kWh usable capacity in the 3.55V to 2.8V range.

There may be a max 1 or 2% due to small imbalance as pack did not had time to properly top balance as it was just the first full charge since installed.

And there may be 1 or 2% if discharged to spec 2.5V and charged to 3.65V but is a bit of a stretch to get to 280Ah.

In any case I'm more than happy with the results as the capacity is in range and less relevant than degradation over time so real test will only be available in 1 to 3 years when I can observe the amount of cycling and calendar degradation of this cells.

Now the attempt to charge the pack as the sun just rised.

Discharge was enable somewhere between 7 and 11% when all cells exceeded 3.2V but I manually turned off the inverter to give the best chance of having the battery charged before the end of the day.

Unfortunately just about 10kWh where available today due to clouds so no full charge.

Will try to do a more controlled test in January or February when day is longer and I can do also a full charge.

[Dacian Todea (electrodacus)]

I did now a proper capacity test as the battery is now balanced after all this time and was properly fully charged.

The result is not that impressive with the 3.55V basically fully charged down to 2.8V basically not much left to 2.5V the capacity is as you see in photo below just 512.89Ah that means each cell is 256.4Ah a bit far from the 280Ah spec but to be fair there is maybe a bit more down to 2.5V but I do not like to stress the battery even for an annual battery test and it will still be far from 280Ah maybe it could get to 260Ah

While charging I put back about 513.65Ah so just 1Ah more. Maybe it could be 2Ah if I will do a few top charge cycles at lower charge current to get to exactly where I started from but anyway the difference between what you put in and what you get out is insignificant less than 3% loss in therms of energy. 

Here are a few photos in the order that test was performed as I started at about 14:02 yesterday from full battery with a discharge test about 25A discharge current with some occasional 48A at the beginning.

It worked overnight until today around 11:04 then from that point charging was started and it was fully charged in about 6 hours at 17:01

Still have the battery capacity set at 540Ah will need to set to the realistic 510Ah to have the SOC correct indication 0% not 5% when empty but I was still hoping it was just imbalanced battery but that was not much and was corrected since initial install and test. Will test next year to see degradation after first year.

Now charging starts. It was a bit hazy but no thick clouds.

Hope it will still be here for next year in March to compare the results.

Also hope for less than 1% degradation but I will be disappointed if it will be 2% or above.

The good thing is that cells have equal capacity and 13kWh usable is still great even if the 4.5kWh battery is still enough for our needs this is mostly for testing as at least 80% of you are using this type of cells and for good reason as they are the best value. 

[Dacian Todea (electrodacus)]

Short update about the amount of energy delivered by battery since the test in November as I was able to find a photo with the about page from that initial test even if the date was not set in the SBMS it was done in 12 Nov so about 4 months ago.

So 572.803kWh - 11.996kWh = 560.8kWh in the last 4 months.

As the usable battery capacity is about 13.3kWh that will be the equivalent of 42 cycles at 100% DOD so not that much and likely from now we get in to summer there will be barely any use until next winter

It will be more about the calendar aging that anything so I hope is below 1% or not far from that.

On Saturday, November 12, 2022 at 5:02:28 PM UTC-6 Dacian Todea wrote:

[Dacian Todea (electrodacus)]

I decided yesterday afternoon to do a battery capacity test as I was testing the new app anyway and battery was already fully charged.

As some may not want to read the entire thing I will start with the results of the test
Total usable capacity is 12.56kWh (482.6Ah) a capacity loss of about 6% from the last test a bit over 2 years ago when battery capacity was 13.34kWh (512.9Ah)

During this 2 years around 2000kWh where output from the battery representing the equivalent of just 150 cycles 100% DOD so the cycle plaid basically no role and all this capacity loss is due to aging degradation.
Thus around 3% per year capacity degradation for this particular battery.

I will have hoped for 2% degradation per year but 3% is not outside of what could be expected. But this means in 10 years it will be at 70% of original capacity instead of 80% I was hoping for.
For me is not an issue as I do not need this capacity and will work fine even after 10 years and this is still the most economical battery at this time.
It could be that my cells where some stranger quality rejects as they where fairly far from the spec capacity when new but is the only sample I have access to and purchased this specifically to test as as much as 80% of you are getting this particular cells due to excellent value it provides.

Here is a screenshot when I started the test yesterday 18:25. I used a 900W or so air heater as a load same as last time.

I reset the E_user disconnected the PV and started the test

This is the screenshot from this morning at the end of the discharge test 8:45


There is about a 2% cell imbalance when at the bottom but that can be part due to delta in capacity between cells and slight imbalance at the top. But that was the same last time also as I did not manually balance the cells for perfect balance but they where just slightly maybe 0.5% more balanced than now after two years based on the old test results thus maybe capacity degradation is 5.5% instead of 6% still not quite the 4% I was hoping for.
The reason it showed 10% SOC when load was disconnected has to do with the fact that I optimistically set the battery capacity to 520Ah will need to set it to around 480Ah to get accurate SOC calculation.


Regarding the amount of energy used from battery
Seems to be around 2000kWh over the past two years so around 3kWh per day from battery as a lot of energy is used directly from PV 4727 - 882 = 3845kW / 740days = 5.2kWh/day from PV mostly cooking, and some heating this of course excludes the diversion heating that is way more.

So say battery is 13kWh then 2000kWh/13kWh = 154 equivalent cycles and that is nothing maybe 0.5% total degradation over 2 years due to cycling and most of the degradation is due to calendar aging (just normal non reversible chemical reactions happening in the cell).

Some more screenshots taken during the test

And now the charging

Cell voltage graph

SOC and internal temperature graph. Very accurate SOC calculation other than the fact that I have the battery set as 520Ah when it should be 480Ah then SOC at the end of discharge will have been 0% instead of 10%

In conclusion this is still the best value battery that you can buy for energy storage.