Sudden Drop in Battery SoC

Had my battery SoC drop from about 24% down to 0% over the course of the 5 minute sampling window yesterday evening. It then charged itself back up to 4% as expected.

On checking the old web portal the SoC of the battery was showing as around 14% which wasn't recovered by the recharge back up to 4%.

Has anyone else seek this sort of behaviour?

Still in BMS firmware 1039 and A449 inverter firmware so it's not related to that.

Not snagged it to tech support as it might be a one off and want to push the firmware update soon too, just trying to time it so the discharge completes for the SoC calibration about the same time as the Go lower rate tariff kicks in.

yandards I had a similar situation on 3rd and 9th Dec during the night where the battery discharged from 4% to 0% and then charged back to 4%. Was on Battery Firmware 1040

Is this deliberate and associated with Battery Conditioning - and if so, is it driven by GivEnergy or ?

KevinT I'd get that for the 4% stuff but to drop from over 20% to 0% in the space of 5 minutes with about 600w of demand is odd.

Was the SOC at a constant 24% for a prolonged period prior to the apparent sudden discharge? I have had the experience of the state of charge appearing to remain constant while the battery is actually discharging through normal operation. The problem with this is that the 4% floor setting fails to take effect and battery discharges right down to 0%. Once it reaches 0% it is as if it suddenly realises that the SOC it thought it had is incorrect and it then recharges back up to the floor setting. If your experience was the same as mine there will be no evidence of the sudden 24% battery drop in any of the other recorded readings (load, grid export, inverter temperature, etc.) suggesting that discharge actually occurred over a longer period and the SOC value had somehow got stuck.

Woodstock no sadly. Usual evening demand stuff from cooking then a fairly steady 500-600w drain with TV and Xbox on over a hour or so before and then a sudden drop from 25% (so about 2kwh left) down to 0% at the next 5 minute sample window.

I still don't understand why it's suddenly dropped when going into the old portal and looking at the battery status page shows the SoC about where I expected it to be.

I've been having battery com drop outs daily but just put this down to an odd software blip.

yandards Does it show the Bms Voltage and the battery temperature drop at the same period? You should be able via the old portal and the graph data. If so, I had the same issue intermittently for a few months. Worked with Givenergy on various tests and found the Bms circuit board at fault. Replaced that, all good since then 👍

Kevsla I've just checked, temp stays nice and linear but voltage does take a dip down to 46.7V fairly rapidly over a short period.

yandards What does the battery SOC line look like in the period leading up to the sudden drop? When it happened to me the battery SOC line was horizontal with no deviation (on both legacy and beta) for several hours before diving to zero. All the other displayed values reflected what was actually happening during the same period.

yandards Ok, my Bms voltage dropped to 0 and the temperature dropped to -20, so it basically lost all comms completely. Hopefully yours is nothing major, I would give the guys at Givenergy a ring and they will monitor it 👍

Woodstock steady linear drop at a similar rate to previous loads of that level looking back at the data. Or to put it another way it hadn't done anything odd prior to the sudden drop.

I've had it down to lower levels of SoC under load as well with no issues.

It'll get a recalibration when I push the firmware update so I'll keep an eye on it from there.

This happened to me this afternoon, rather worrying as I've only just got the kit installed

MrChaz I'm seeing similar odd behaviour from my newly installed battery system. It can bounce between 0% and 100% SOC.

Full description ...
Over the weekend I have been trying to get the system (AC coupled inverter, 8.2 kwh battery) to work properly. The inverter seems to stay online OK but loses communications with the battery. I have updated the battery and inverter firmware to versions 3001 and D0.534-A0.534 respectively.

Yesterday during a period when comms were working OK I set a limited time and percentage grid charge. The battery started charging and soared through the time and percentage limits to charge for longer and to 100%. It then went into an ‘idle’ (not standby) state and shows 0% in the dashboard but seems to jump to 100% at times. I think it is supposed to be in Mode 1 and meeting demand from the house. Also the Beta dashboard shows firmware version for the Battery as 0 and Battery capacity as 10Ah as opposed to the 160Ah it should do.

Lights on the battery show all steady green with a full charge.
DC isolator switch is UP (On).

Lights on the inverter are interesting at one time they cycle from steady green central light with charge to battery indicated to flashing green central light with no battery lights. Also I ocasionally hear relays/contactors in the battery apparently cycling.

I could probably reset the battery once again and coax it into meeting the house demand but I think something is fundamentally wrong with the battery communications. It shouldn’t need manual intervention several times a day.
I have tested the battery comms cable continuity and it is fine.

Good morning guys.

I had the same thing last night (19th to 20th December 2021) Inverter is on A0.533 and batteries on 1039. The LED indicators on each battery are different. One 8.2KW pack is indicating at (what looks like) 20%, the other is showing one flashing LED, presumably 4%. I'm wondering if the guys at GivEnergy have done some form of update (UPDATE: No, they hadn't. Adam emailed me telling me they don't do that without asking and only when customers report a fault). It's really annoying as I'm data gathering at present to make the system work for us and last month spreadsheet now won't have any meaningful data to use.

Apparently, I need to have my system updated and the batteries optimised. Roll on a working system 🙂 Happy Christmas everyone.

MrChaz

Might want to get hold of support MeChaz...

Had it happen again yesterday, about 1700 sudden drop from around 25% SoC down to 0 and then charged back up to 4% SoC as per the system settings.

I've finally pushed a firmware update last night, albeit the battery is on the up to date firmware with the inverter on D449 and A451!

Battery calibration has completed successfully, interestingly the previous 100% SoC it used to show around 0600 has gone this morning (charged on Go up to 0435 to won't discharge again until then).

I'll be keeping an eye on it but hopefully it just needed a battery calibration.

Spoken to GivEnergy support yesterday who were mildly helpful - told me it hadn't been commissioned properly but because it'd started a calibration then it should do that.
Was still all over the place (charged to 80% then dropped to 0%) so I called my installer who said they'd contact GivEnergy.

Two further conversations with the installer today and apparently GivEnergy still haven't called them back. Battery still not working.

All seems ok post battery calibration, down to 18% SoC with no issues so far.

I manually triggered the Start/Discharge command last night and after it finished it charged all the way up to 100% and things are looking good now 🤞

spoke too soon.
The primary battery now won't turn off and the status light is flickering orange and the first battery level light is flickering green.

Did any of you folks have a successful resolution - mine has just dropped from 27% to 0 then charged itself back to 5.....
I have raised a ticket with GE but it would be good to know of a general solution :-)

TimBishop it's been fine since then and I didn't set the overnight charge to accommodate the weather today so it's fully run down as expected.

Think it's just an occasional nuance of the setup.

My SOC has been pretty consistent - given that I've yet to fully charge off solar so haven't seen 100% in the wild so to speak.
But it does run down to 4% all the time.
Last night it dropped to 5%, then around midnight dropped to 4% - pretty normal so far. But at about 1:30am dropped to 0% where just shy of 1kW was pumped into the battery to get it back up to 4%.
This happened at exactly 45V on the battery voltage.
Its now sitting at 51.11V and 4%.
Will be interesting to see what happens today as its meant to be reasonably (Yesterday was horrible and only got 2kWh off the solar 🙁 )

I charged to 40% last night anticipating a sunny day. Forgot to account for the coffee machine in my calcs.

Battery dropped to 0% then charged from grid to 4%

Wish I’d gone a bit higher e.g. perhaps 60% evening charge.

anglefire that sudden voltage dip is definitely a data point trend for the battery suddenly dropping down to 0% SoC in the space of a single 5 minute snapshot window.

I've looked at the battery data tab when mine last did this and could see voltage dipping under load (more so than usual) prior to its sudden drop off so another early warning it might do it.

PV generation was lousy yesterday, a whopping 1kw for me, oddly forecast.solar plugin on HA which has been within a kw of production kept insisting I was getting 10kwh over the day.

I had a similar experience last night. About 12:17 AM suddenly dropped from 16% reserve SOC to 0% and proceeded to charge from Grid back up to 4% and held there until this morning when solar charging has begun. At the time of dropping from 16% to 9% there was no significant load in the house, just a very small background load while we slept. Where did that electricity in the battery go to? Interesting that others have had the same exoerience about the same time of night last night,

The above sudden drop in SOC occured with a battery voltage of 44.36. There was also a Battery Undervoltage notification at that level which cleared 5 mins later.

Something similar happened to me at 4.55 am this morning... charge dropped from 20% to 0% in 5 minutes. During which time the inverter current shot up to around 10A to seemingly get the battery back up to the 4% and then dropped back down to 0A at 5.20 am once the battery was indeed at 4%.

I've got the 9.5kw battery, there's no timed discharge or timed export setup and the reserve is set to 4%.

cmanifold A more current thread is this one.

cmanifold

Interesting its also affecting the 9.5kwh new batteries too.

Are you guys still having the dips in SOC as my Gen2 5KW and 2 x 9.5Kw batteries are suffering the dips every couple of days. I also have dips on the temperature readings. GE Support say that BMS 3018 will offer some improvements on this.

Wierd - my batteries (2 x 5.2) also lost about 30% of charge around 8am this morning, no significant loads from the house.

They have been 100% stable for over a month since an issue back in May that was sorted by support.

Ortwin I often sees these kinds of transient dips to 0 in the app presentation. Which I would put down to comms problems uploading from inverter to cloud. However, when looking at the raw data on the portal, the gaps aren't there. Which is a bit odd.

But I think these are just transient reporting problems, whereas some people are experiencing genuine, dramatic changes in SoC.

wilfleek I think having two batteries can complicate things. Have you checked that they're running the same version of the firmware, and that they are at a reasonably similar level of charge ?

DD thanks,

Yes, they run pretty close to each other on SOC, after an issue that was causing them to drift apart (a register that was incorrect, IIRC) was fixed remotely by support during May. I loaded 3015 onto each of them myself using the memory ticks a whole back.

wilfleek when you look at the detailed info on the app, did they both drop SoC, or was it just one of them?
(I only have one battery, so can't offer any personal insight.)

Both, and at the same time. First time I have seen an unexpected SOC drop in a while.

Massive SOC correction this morning, from 69% to 11% over 35 minutes. This is a 8.2kW on a Gen 1 GIV-HY3.6:

chris_p Tha’s pretty awful. Have you been running the battery towards full all the time in the summer, in which case the inverter may just have lost track of the SoC.
Worth trying a full discharge and recharge so it gets back the idea of where the SoC is.
Also check the individual cell voltages (Rubik cube’s Android app can show this) in case there is a dodgy battery cell

geoffreycoan I could do all that or GE could admit that they will never get this right. I've had my kit for nearly three years and this SOC issue has been happening from the first months of installation. After years of promises that the next firmware either solves the problem or improves it, we are still at a situation that the equipment cannot track the SOC accurately.

chris_p I’m not saying the GivEnergy equipment is perfect, and the AIO seem to have terrible trouble tracking the SoC, but I’ve had my two Gen 1 hybrid’s for 18 months and I don’t think I have ever seen a random SoC drop on them.
I do use the batteries quite a bit, cycling from empty to full and back again most days and I know this helps keep track of the SoC.

With such a massive drop then to me it points to either a battery cell issue or the inverter loosing track because the SoC has been kept near full for an extended period of time.

Assume you are on firmware 450/451 the main public release for Gen 1’s

geoffreycoan Inverter is D0.499-A0.499 and battery is 3015. A few people have said that those who cycle full > empty > full have better SOC tracking, which is great for those where it happens organically. But for me to achieve that I'd have to nanny the system daily, not what I expected when purchasing a system like this.

Mine behaves the same way at the moment; only cycling 100 - 80% or so gradually loses track such that indicated 100% is in reality less and less. Eventually 70% indicated is pretty near empty and it dies. Crazy really because I can look at the voltage (under ongoing low or zero draw) and estimate the real % better than the BMS!

It's actually not a big problem because it behaves much better when it's needed and hence cycling more fully, just the unpredictability at this time of year is a bit poor. I guess one positive is that it does mean the battery isn't sitting at pretty much 100% for months, which may not be good for it.

I was having this issue with sudden drop of battery charge in a very short time, normally combined with a BMS under/over voltage message.

I contacted GivEnergy support and their engineer said it was a known issue with battery fw 3015 and upgraded mine to 3017. Its been over a week now and I've had no repeat of the problem.

Just thought I would let you know in case it helps others.

Thanks!

My Gen2 inverter and 2 x 9.5Kw batteries often loose the SOC and battery temperature for which GE say there will be a fix in 3018

joon79 I too suffered from regular sudden drops in SOC. I need the EPS for personal reasons and a sudden loss of energy was not good.

Through my installer I too have updated to 3017and am on 499.499 for the inverter; for another issue. Apart from one over volt I would say the system is tracking the SOC better as it generally falls a little further overnight with the same load as before.

I would rather they get it right than get it early- though both together would be even better.