Givenergy Battery Capacity Problems

ispookie666

The Only analogy I can compare the impending FW upgrade is Flash Sale which runs 24*7*365

awans

I called GE customer services 3 times already. Every time there's a renewed promise for a fix. Nearly 9 months now since first reported.
If it doesn't get resolved I will ask for a new battery, as to my understanding the SOC fault is not universal and only occurs in some batteries, this is according to their customer services. 
But it's really good that this forum is used, on Facebook they've guys who won't let you present your case.

pete-20-11

The fix for the soc bug is in the battery firmware 3012 which is a available to install from the web portal (on the MyInverter page). 

There’s further firmware updates to follow for the inverter, but I’ve not seen a “soc crash” since using the new firmware. 

steveconrad

Great, I'll get that installed tonight, I don't fancy doing that in the middle of the day and 'wasting' a 90% full battery.

Oddly with my battery hitting 100% a few times a week I've not seen one of those massive jumps. Fingers crossed the new firmware works.

94JDH

steveconrad said:

Great, I'll get that installed tonight, I don't fancy doing that in the middle of the day and 'wasting' a 90% full battery.

Oddly with my battery hitting 100% a few times a week I've not seen one of those massive jumps. Fingers crossed the new firmware works.

The SoC bug was mainly noticeable in those that didn't fully cycle the battery

NearlyLost12

I saw one of those jumps this morning: 65%->0% at 4.45am. This is the second one; the first was 'only' 40%->0. Then of course the inverter immediately used the grid to charge the battery to 4%.  This is firmware 3009; hopefully the new firmware - when I want to allow the inverter to calibrate the battery - will fix it.

Speaking as an electronics / software engineer it looks like whatever counts charge in and out isn't very good at it. There must be an overall cutoff based on battery voltage which panics and resets the charge to zero.  I hope they are not using battery voltage to determine charge: lithium batteries have an almost flat charge / voltage characteristic.

DougMLancs

3012 worked a treat for me

ispookie666

3012 does not fix anything. The problems are not fixed with beta firmware. 
@NearlyLost12 they are using Voltage and lookup tables to determine SoC and believe its superior 

NearlyLost12

You are right about firmware 3012.  It was fine during the summer / early autumn when the battery never got down to its minimum but now strange behaviour is manifest.

Even when the battery has got to 4% and is not being used it will continue to show a discharge down to 1% when the inverter will trigger a recharge from grid to 4%.  This in itself is annoying but just about OK.  However it gets worse...

On several occasions the battery is observed to accumulate charge - seemingly from nowhere - to between 10 and 12%; at which point the inverter starts supplying the load.  This lasts until the battery is again at an indicated 4%. All of this is taking place around 6am.  This morning at 7:15am a final twist when the battery charge dropped from 10% to 0%, probably the fail safe cutoff since the voltage was now 44.8V.

As to the use of lookup tables and battery voltage to determine charge I despair.  As soon as the internal resistance of the cells increases - which all Lithium based cells do - the lookup tables are essentially out of date.  The battery voltage under load now represents a different state of charge; and this assumes you can read the voltage with sufficient accuracy.

ispookie666

a new firmware 3015, and nothing has been fixed - maybe a cosmetic/superficial touch-up.  
Over time, the battery SOC and capacity diverge and it reaches a point when the battery is empty with a state of charge of > 10,  drops till the voltage hits 45V.  
Givenergy just obfuscate, all of their batteries have this problem