For the third time I have had a state of charge SOC drop by 30-40% in around 30 minutes when load was only a few hundred watts.
Support advised to run a low discharge rate calibration cycle which was done a couple of weeks back - but this did not fix problem. Today there was a 48% drop in 35 minutes.
What is happening to apparently lose half the capacity of a 9.5kWH battery.
Is this a faulty battery that needs replacing or is there another explanation?
If its like my hybrid and smaller battery, I understand there is some difficulty keeping track of the SOC [state of charge]. At some point the measurements (probably voltage) kick off some recalculation, and it realizes that the SOC isn’t what it was, and it looks like energy has been lost. I hope [live in] that there will one day be an update to improve the Gen1 software.
Others may have a better explanation.
mine dropped loads of times rang givenegry just give you BS,
i just set battery not to drop below 50% works for now.
We had this problem on several occasions. After several re- calibrations and a software update the problem remained. Givenergy engineer came Wed , removed battery cover, tightened various nuts ( which are known to sometimes come loose in transport from factory) and hey presto we now have a fully functional 9.5 kwh battery which charges and discharges to full capacity.
I can confirm I had this problem too. It’s not that the battery is suddely disharging - the energy has to go somewhere. It is more that the reading is false ie 70% SOC is actually 20% so when the system realises this it resets to 20%. I suppose you could compare this to a car with a faulty fuel guage. I too has an engineer visit and tightening up the battery connectors seems to have cured the problem.
We used to regularly see the ‘loss’ of lots of power.
GE remotely installed 3017 early July before it went on general release. It didn’t need a calibration cycle and since then the power tracking has significantly improved as we have had no obvious unexpected step change in power available.
My understanding is that the SoC [state of charge] is deduced from the cell voltages, and as the variation between full and empty is not huge it is somewhat difficult. However BMS 3017 is better.