Battery Firmware 3017

Due to constant SoC drops on firmware 3015 on my 9.5 battery, GE support put me on 3017, which may or may not fix the problem for me.
Anyone here with that firmware? Is it any better?

alfwro Yes, I’ve been on 3017 for a few weeks. It’s proved very stable so far, no SOC corrections noticed, despite the battery being predominantly above 60% at this time of year.

Gen1 Hy 5.0 9.5 battery
I am 3015 and was getting BMS Over Voltage errors. I reported this to Giv Energy and was told "new firmware coming soon", but that was 3 weeks ago and still nothing. No update link or anything

alfwro

Yes I’ve been on it a few weeks as well (since 10th June). Too early to say yet. I’ve seen some spikes (but they go back to the original SOC value) but no big “drops”. I’ve got a slow charging and discharging issue since 2 replacement batteries on 20th June.
3017 was working normally for the first 10’days from what I recall. But I think it caused the secondary battery to start to get stuck at 100%. May have been unrelated though (or not?).

What sort of SOC drops are you seeing?

SteveCook I get those BMS OV errors too.

Pete UK
Battery performance (%) pretty smooth and responds as I would expect with Predbt and Flux.
No big drops

SteveCook Yeah, i tried to probe them after the firmware mess up a few weeks back... no time scale and no option to self update.

SteveCook "Soon"(tm)

TimI
It's been bumped up to " very soon ". There's only " extremely soon " to go before we reach the " imminent " sequence 🤔

If they bump the alert level up anymore they'll need to consider...

Tenkaykev then its imminent, impending, maybe... don't get your hopes up?

3016 will be released first ...

Personally, having been bitten by both 3013 and 3015 I'll be sitting it out boys.

Best of luck to those going over the top first 😉

hoggy Space corp directive 34124....

Hi Guys,
I've an 8.2 battery and tried a firmware update from 3007 to 3012 (I see there may be a 3015 version and a few in between) yesterday but it came back as failed. I tried to remotely restart the inverter but no change. On the inverter dashboard I have no options on the software page to do anything e.g. retry or download firmware. I looked through the givenergy website for firmware but it's not given out. I've emailed support so hopefully they get back without delay. I realise that I may need to take the cover off the battery and remove the USB key and update the file from a PC. Any thoughts or suggestions on this or anything else to try?
@TheDragon (GivEnergy) can you help here?

DazMav the service desk can send you the latest files and guide you through a manual load if that's what's required.
It's rare that a firmware gets stuck or won't load, but a manual route will rectify the problem. Providing the USB stick is actually present to be used.

TheDragon (GivEnergy)
Get a sharp Stanley knife to cut the glue holding the USB stick in place in the inverter if as (the Dragon says) it is still there

@TheDragon (GivEnergy) @SteveCook Thanks for your replies. The inverter is functioning OK. It's the battery firmware that I was trying to update automatically via the dashboard so since this failed it looks like I need to do this via the USB in the battery case. It's a pain having to do this. Seen a YouTube video on this from GivEnergy.

@TheDragon (GivEnergy) @SteveCook Quick update here, I restarted the battery and the inverter and the battery has started up where it left off with 40% charge. It's still showing no battery firmware version like picture above but it is working. I assume since all looks OK I can leave it running and then when I get the firmware I update as per YouTube video. https://www.youtube.com/watch?v=xws4RCOwNkY

@TheDragon (GivEnergy) @SteveCook So live updates here... 🙂 The battery firmware has popped up on the inverter now Version 3015. It didn't do a discharge and charge cycle but it looks like I'm all good.

Lesson for me. If the firmware says it failed then restart the inverter and battery and wait for any changes.

Memo to self: never do an upgrade on a Friday or the weekend. Monday to Thursday is better in case you need GivEnergy support to help intervene

geoffreycoan I have updated on a Saturday but only after the nice GE support chap confirmed he was working on the Saturday and would be around to assist - I needed him to advise as it did fail and I had to update manually. But generally yes, don't update on a Friday or the weekend!

Does this update to 3017 require a calibration ? I don't have it showing ( i'm on 3015), but do resist a calibration unless 1) absolutely necessary and2) the FW brings benefits.

what update to 3017. I dont think it exists unless GivEnergy install it remotely (in some way) or email it to you to install via USB (in some way)

SteveCook As per the title of the thread. I don't have it available via the inverter, so I am blind to it except via this thread.

GivEnergy remote updated me to 3017 a couple of days ago.

I'd started getting over voltage errors which were getting more and more regular so I put the ticket in and they offered me the upgrade within 24 hours.

Battery did a re-calibration immediately which was going to flatten the batter right in the middle of peak Agile prices so I reduced the discharge rate to 15% which had it empty at the cheapest possible time.

It's early days, but I've not had an over voltage since the update.

I find it very bizarre that 3017 is being supplied by GE Support but no one can install it through the dashboard.
Why is it good enough for limited release by GE Support but not good enough for general release to all GE customers?

P3tro I'd suggest that 3017 is effectively in a beta test. Givenergy want to see live customer feedback before pushing it out in the field to everyone and getting lots of feedback if it's not perfect.

Some of the previous releases have been publicly available before they were really ready so this sounds like a good improvement to me to get real customers feedback first.

Of course frustrating if you're waiting for it to be generally available, but I'd prefer to have it fully proven before taking an upgrade

Well I've been on it for 2 weeks with full calibration and have already experienced 20% SoC crash - that could be one of the reasons why it is not being pushed out - from talking to GE support I am not the only one having that problem.

P3tro seems that 3018 has already been mentioned ?
https://community.givenergy.cloud/d/100-sudden-drop-in-battery-soc/30

No issues for me on 3017 since 9th July, hopefully it stays that way

I’ve been 3017 since June 3rd. No issues whatsoever. Most stable BMS I’ve had in the 30 months I’ve had the battery.

For me the most stable was 3013 - I've been on that version from August 2023 until last month and I've had no SoC drops, over or under voltages, other then occasional BMS crash it was perfect. I deeply regret upgrading to 3015 and now 3017.

I've been discussing over voltage issues with 3rd level support which I've been seeing periodically on 3015. I'm managing them with forced discharges and watching cell voltages etc. I'm on holiday at the moment so didn't want to take the offered 3017 as this was noted as still not solving all the issues they'd hoped. Also 3018 was mentioned as "coming soon". I think I'll avoid 3017 and see what 3018 brings.

alfwro Indeed. 3012 was better for me than 3015. The latter is "OK" but still needs too much intervention.

geoffreycoan I agree with you.

I’m on 3015 from 3012. I’ve seen SOC corrections and the drops seem to have improved but not eliminated. It’s been a strange summer with weather where I live too so it’s been more obvious.

Extremely frustrating they can’t get this issue sorted either retrospectively or for when they sell new batteries like the AIO. If I had known of this issue before purchase I wouldn’t have gone with GE.

It’s better than it was but given this has been going on for years I’m really disappointed. I may feel better or worse about it this Autumn/Winter when SOC drops are more common and obvious.

Soon is a relative term.

I too had a direct update, requested via my installer, and initiated directly by GivEnergy to raise battery from 3015 to 3017 Although a work-in-progress release, this was expected to help with the sudden loss of SOC. It updated one lunch time and there was no subsequent calibration cycle. So far I believe it has improved the SOC tracking as a 'quick and dirty' review shows the overnight battery decrease rate to be at slightly higher than previously, which would be consistent with better tracking - rather than the previous occasional early morning drops in SOC from 38% to 6% in a few minutes.

I will say again, although I would love to have everything updated 'toot-sweet' I would rather they did get it right and well tested. I would however ask for an improved customer feedback page from GE as to what is being progressed and a reasonable estimate (not just soon) of when we might see it and what it will change/improve.

I was told by GE technical that 3017 was being offered to systems that had variances in individual battery cell voltages and not for the SOC (and temperature) drop issue that I am still experiencing. They changed both my BMS PCB's just in case but no change.

I have asked to be downgraded to version 3013 as 3017 was just unusable - constant SoCs drops.
With 3013 despite the occasional BMS crashes it at least does a very good job of reporting correct SoC and it does work with the new G1 inverter firmware. Also no over and under voltages for me.

Just popped up for me, I am reluctant to update as I do not know if it will re-cal.

I've now got the button to update to 3017, this morning. It only updated to 3015 less than a week ago.
The batteries (I have two 8.2kWh) are acting strangely since last night - one is down at 24% while the second is at 100%, with the 'total' battery showing as 24% as if the second one isn't there.
It's a cloudy day so charging is slow, just a few 00 watts.

The knowledgebase doesn't mention version 3017 so I'm reluctant to install it!
https://kb.givenergy.cloud/article.php?id=22&oid=25

The battery behaviour is sorting itself out now, I'm at 31% and 95% with a 'total' of 64%.
I suspect this was a 'calibration' routine in action. I have emailed GE about it.

Giv pushed 3017 to me yesterday with a cal from 4pm yesterday ( so that I could get 6hrs from Octopus at 7p a unit. No idea what it does, but on 3015 my batteries were hitting over volts then discharging/charging so wasting energy. 3017 was the Giv "solution". Time will tell.

Roberto I had 3017 about five weeks ago and so far seems to track the SoC better than 3015. It was installed in 10 minutes one lunchtime and did not need to have a calibration cycle.

gwynnej3 I have the same message... not quite sure what to make of "The generation 1 battery will be updated" since my battery is gen-2 AFAIK.

Though looking at the release notes, gen 1 and 2 seem to be lumped together, with gen-3 as a separate device.

I note that the release notes call this "BMS2.0" - that sounds like quite a substantial change! Most of the benefits seem to be for multi-battery setups

DD

The generation refers to the battery not the inverter.

If you have a 9.5kwh battery. As far as I know there has only been one version. There are more than one generation of the lower sized batteries.

hmm...

I've tried upgrading from 3015 to 3017 and now the inverter is rebooting every 5 mins, Does anyone know of the fix? Or is this a call to GE support?

Hook If you have a 9.5kwh battery. As far as I know there has only been one version.

https://givenergy.co.uk/wp-content/uploads/UK-Datasheet-Giv-Bat-9.5.pdf describes a gen-2 9.5kWh battery.
(That's not to say that there hasn't been only one version - perhaps it was just called gen-2 from the outset, if it came as part of a refreshed suite.)

Midlands_Solar Or is this a call to GE support?

I'd say so, yes. Though presumably they're now out until tomorrow, which is unfortunate.

DD thankfully it's failed and the reboots looked to have now stopped! I'll be leaving this update

DD I spoke to GivEnergy when i was going to add another 9.5 battery.
They said there is only one type of 9.5 but the reference to Gen 2 comes because of the inbuilt isolator switch and plug/connect cables to ease installation

No sign of 3017 here, but I’d imagine it’s being done in stages?

I noticed the 3017 update this afternoon and tried it. It was a long long time before I got a message saying it had failed. Meantime I had a sense that it was restarting a few times as my home assistant seemed to repeatedly be working with it and then losing touch with it. It's settled down now and I remain on 3015. I'm about to go away for a few days so I'll contact support next week.

The_Engineer I managed to get my cal started at 4pm, so that it was finished just before my 6hrs at 7p a unit. Worked quite well. I'll see how it performs with two 9.5s

I've just seen the 3017 software update but strangely it says I'm running 3007 when both my batteries are running 3015. Closed the app and rechecked and it now says 3015. I thought 3017 was for batteries with individual batterie voltage variations, which my two 9.5's don't suffer from. I wonder if it will sort out my dips in temp and SOC as well

The 3017 update appeared on my Inverter software portal page yesterday. Currently I'm on 3015 firmware with 2 x 9.5 batts which previously had been suffering with over voltage errors until these notifications were wiped from my notification history a week or so ago.
Also my batteries don't currently charge overnight from the grid beyond 90% (even though set for 100%) & only get up to 100% charge if there is solar the next morning to top them up. Soft buffering to prevent over voltage errors?
I've not instigated the 3017 update yet & I'm waiting for the right time of day to minimise an discharge / re-charge losses. That said there's no mention in the notification that a re-calibration or discharge / re-charge will take place.

I had an email from octopus regarding a free energy session (above average usage at that time) today between 13:00 and 14:00. I was hoping to charge up my batteries but being on OIF I doubt the octopus API will be intelligent enough to make this happen.

gwynnej3 Stop Press: There is now a description of 3017 on the knowledgebase...

Ortwin what percentage are you currently at. Mine is at 47%

Octopus do sometimes charge outside of their usual pattern, so I'm hoping that they are planning to do the rest of today's charge between 1-2pm. 🤞

TX200

I am currently at 72% (75% on primary and 64% secondary with 53.4v on both) if I wait till after 14:00 I could update the battery software (hopefully on both) and then do a recalibration which might bring the batteries into alignment (yet again) for a bit longer than the last calibration.

does the upgrade to 3017 force a calibration cycle?

locked Mine did not.

I spoke to GivEnergy who did a remote upgrade to 3017 and during the call they told me it would not do a calibration. It did not do a calibration

Forced change from 3012 to 3015. Just changed to 3017 - no calibration.
@Rubikcube app message - Sorry, battery information is not available from your inverter because the protocol doesn't allow it.
GE don't want me to see individual cell voltages, why not?

Update. Battery voltages have just appeared. I check these regularly, they are much more reliable than SOC.
Thanks to Rubik, apologies to GE.

Well, I gave a it a try about an hour ago, stayed on "Verifying 90%" for some time, dissappeared when I refreshed the dashboard.
Now still syas Software version "3015", no notification to say the update has completed an hour later. I'll give it some more time.
Never have been able to auto update battery firmware for some reason, have always had to pull the memory sticks and do it manually with help from Ge support. May well be the same this time. Not calling them today as the batteries are working normally, are 80% full, and I don't want a calibration run at this time of day - will wait untill tomorrow. Have always found support to be very helpful myself.

Update - saw the inverter was restarting every 10 minutes, so called support who promptly stopped that, and confirmed that the OTA update hadn't worked. So they are sending the file by email and I will apply it manually tomorrow morning. After that it will need them to trigger a calibration.

Yeuck. Good luck with the manual update tomorrow.
Has anyone managed to work out what it is that causes failed OTA updates? @wilfleek Are you prepared to admit to a particularly flaky Wi-Fi connection, or some other reason for failed OTA updates?
A misbehaving update process, I'm sure, will put many off keeping their systems fully up to date. I had to put on my biggest set of big boy pants to brave 3007 -> 3012. That update went reasonably well, but it didn't kick off an automatic calibration. Even that spooked me a little. I really don't fancy taking fronts of batteries and waving Stanley Knives around.

BobB My update failed at the first attempt today, resulted in the inverter restarting every 5 minutes for 2 hours until it decided it had actually failed (new notification). I tried again and it succeeded within 15 mins with 2 restarts. I sit on a pretty solid internet connection and have not had wifi dropouts to my inverter. I did stop GivTCP running during the second attempt (forgot the first time) which may have helped but can't say it was definitely the cause. In general my updates have succeeded previously.

3017 update has finished - stayed on 90% verifying for what felt an age. An oddity, in the app one of the two 8.2kwh batteries is showing a capacity of 0.0kwh. Prior to the upgrade the soc between the 2 batteries had drifted apart by about 15%. Will keep an eye and raise a ticket if it does not rectify itself.

Phil_H Mine did exactly the same! I gave up after three failed update attempts. Also I wasn't keen on the inverter being restarted so frequently as well

Oddly and this may have been done remotely by GE support having raised a ticket with them last night, however my battery is now showing as running firmware version 3017 from 3015. I left the firmware as pending update with the battery running 3015 as of early this AM

I've checked the logs and all entries point back to the three failed attempts last night.

Do GE just remote in and remotely fix??

BobB My wifi connection is rock solid, never had an issue with it, so no idea why OTA updates won't work for me, but they never have - at least not for battery firmware, inverter update went OK!

Did mine today and it seems sound, I am used to seeing loads of BMS notifications.

wilfleek interesting. I have a mesh WiFi system and have never had an inverter or battery firmware update fail.

I have a gen 1 running predbat as well so comms are very important.

Well I tried the update but it fails and I'm on a LAN connection not WiFi. So email sent to GE technical for them to send me the file and update locally. It reported that I was running 3013 but definitely I was on 3015 and always seems to get stuck at 73%

locked stayed on 90% verifying for what felt an age.

Just updated my 9.5KwH from 3015 to 3017. Took about 15 mins, 12 of which were at "90% verifying..."

No battery recalibration.

Upgraded 3015 to 3017 here, no calibration required.

Thanks all for your experiences of 3017. I see that this is being offered via the portal as an update for my two batteries (although as noted it classifies my 9.5 as a Gen 1 even though it was advertised as a Gen 2 battery).

From re-reading the thread I think the feedback boils down into:

• works great, SoC better tracked than previous BMS’s, less big jumps but still occasionally some

• much worse than 3013 and 3015, very unstable, had to downgrade back to 3013

• OTA update fails and gets stuck in a continual reboot cycle requiring support to intervene

I currently have 3015 on my 9.5 and 5.2 and it’s very stable, I never observe any SoC drops. Looking at the cell voltages, on the 9.5 (93% full) they range from 3.427 to 3.409, and on the 5.2 (97% full) from 3.320 to 3.329 so no massive differences. Cell voltages over the last 24 hours and last month show the cells all moving up and down together.

I’ve looked at the kb for the BMS changes and nothing jumps out as being essential to have so on the basis of don’t change what isn’t broken I think I’ll stick with 3015 as its working fine for me.

Put the update on the memory sticks manually first thing, Ge then triggered a recalibration, which is happening now. Portal reports correct firmware versions.

Yesterday, I upgraded my two 8.2 batteries (on AC3.0) from 3015 to 3017
Took about 15 min with half the time at the '90% verifying' stage. No calibration.
So similar to TimI and rjp

System has run smoothly since and I think a bit snappier.
In my givtcp log, I used to see some dropped slots and errors changing over from 1 day to the next (linked to MQTT?) . I also saw the odd 'assertion error'
At midnight last night, there was no error and just the code line:
2024-08-16 00:00:40,531 - Inv1 - read - [CRITICAL] - Midnight, so resetting Day/Night stats...

Getting to the end of an upgrade is always a relief, when it goes well. 🤞
I need to see how well the SOCs on the batteries are synced. They could wander by up to 15%.
Less than 1 day in though and things could go belly-up.
As geoffreycoan has concluded, the upgrade experience can be very different.

Rob

geoffreycoan I'm leaving it as I understand how 3015 behaves with the fast response f/w - which has specific "3015 support".

I don't understand how 3017 works and I don't have any confidence that G1 inverter f/w will be updated should 3017 turn out to be a basket case like 3012/3013.

My mate (gen 3) has updated to 3017 and he has a 9.5 and 2.6 battery.

“ 1. Fix the 9.5 mix with 2.6kWh issue of restricted current”

He’s seen an immediate increase in charge and discharge speeds. Now gets 3.6kwh (was 2.3) on charge and 3.4kwh (was 2.5) on discharge.

Vestas I have gen 1 5kw with so called fast response software and 3012 with no issues. Currently asking to update to 3017 and Givenergy are also recommending when I asked. Still not sure on if it aint broke don’t touch it rules

In the 18months of the system being installed I have updated 3 times. Only once has gone without a hitch and no manual intervention by myself and on the phone with Givenergy. The last time updating to 3015 battery and 912 inverter firmware I was on the phone most of the day and a stressful one at that.

It really has put me off updating unless it is necesary to fix an issue I can't live with. The system is stable and works as expected so I do not see a reason to update.

I woudl update if Givenergy say I need to but I do not want to volunteer to do so without a very good reason.

I updated to 3017 this morning, no issues apart from having to restart to inverter and the Wi-Fi access point to get the dongle to reconnect (it may have just sorted itself out but I was a bit impatient).
I'll see how the batteries behave now and will report anything adverse.

The D0.205-A0.205 inverter firmware is working really well. Put the kettle on and it's a matter of seconds before the battery responds to the increased load. I'm happy with that.

I received this from GivEnergy support, just in case anyone else experiences these updates failing as I did

Wasting paper again ...

Slartibartfast like facebook if you don’t like it move on and don’t comment!

Again lots of posts on this, and this is an official comment from GivEnergy support to “help others”! As not all will see all threads

Midlands_Solar
This isn't Facebook, and it is generally a friendly and helpful environment.

I also find your multiple identical postings unhelpful. No other user on this forum behaves in that way.

I upgraded to 3017 yesterday (nice and easy, but now when the battery reaches100% it doesn't sit at idle but discharges at around 15-20W.

This despite 3.7kW of solar being generated. Any thoughts?

PaDav6 are you hitting peak export?

I presume your solar is clipping and the inverter is ‘disappearing’ the excess.

This was yesterday for me.

No not clipping.

I have a Gen3 5kW Hybrid inverter with a 9.5kWh battery and 12x4.05kW panels facing SE, Highest gen I have seen was 5.3kW in May when it was cooler, but the lack of battery idle is since the 3017 upgrade. Ran at 15-20W for 11 hours yesterday from full until the solar dropped below the house load and the battery took over.

Oh thought you had a 3.6.

I have a single 5.2kw/hr battery which is set to charge, from the grid, to 50% during the Octopus Flux period, specifically 04.00-05.00. This has worked correctly since set that way in early May. On 16 Aug I updated the Battery Firmware to 3017 and since then the battery has charged, during the Flux period, to 100% despite still being set to Forced Charge to 50%. I have just changed the charge value to 45% to see if a change will trigger a change in behaviour. Not hopeful!

Is this a known phenomena and can I do anything about it? Also, if I can't put this right can I roll back to 3015?

2 x 9.5 Gen 2 batts & 3.6kW Hy Gen Gen 2.
3017 installed from the Portal update button on the 16th Aug. It took around 20mins with no battery re-calibration.

On 3015 my batteries were not charging to 100%. The best max charge I was getting was 85-90% (indicated SOC) overnight from the grid & the batts only hit 100% when topped up by solar the next day & this delayed any export potential / income. I suspect battery charging behaviour was to stop the regular Over Voltage errors which I & many others had been experiencing - which have now been deleted from my notification history!

After the 3017 update the 1st overnight charge the batteries reached 96% so a bit better with some on/off inverter buffering around 90-95% & before the end of the 5.30am charge slot (I have my battery discharge paused between 11.30pm & 5.30 am to prevent the EV draining the batteries when it is plugged in).

Last night (2nd night on 3017) the batteries charged to an indicated 100% with less buffering & more stable operation so it seems to be learning. Fingers crossed it looks to be operating better.

Final night on 3015

1st night on 3017

2nd night on 3017

geoffreycoan
Our 9.5kWh battery is on 3009 and working OK, but nevertheless we plan to upgrade at the end of the week, (for speed of response and multiple charge slots)
Since we’ve not had serious SOC jumps I suspect our cell voltages will be similar. Can you see these from the portal or do you need to interrogate with HA (which we don’t have)?

Judith You can use the Inverter app, available free from Playstore to see cell voltages.
https://givenergymonitor.wixsite.com/tutorials

Judith You can’t get cell voltages from portal but, as you have guessed, that you can from HA.

Rob

Davegreen1 I would be surprised if the upgrade to 3017 has caused this. As far as I know, charging limits are controlled by the inverter, not batteries.
Is there any reason not to charge fully in the cheaper flux period? Any solar will then be exported rather than being used to fill up the battery. I think the cost is essentially neutral.

Rob

Judith Can you see these from the portal or do you need to interrogate with HA (which we don’t have)?

There's the android app as mentioned. It is possible to use modbus from a pc (or similar) without setting up the whole of HA, but that's a bit hacky at present. Not too hard if you've ever run any python code. If anyone's interested, I could try to write a step-by-step guide.

I think maybe GivTCP can run outside of HA? But perhaps I imagined that, and it probably still requires a certain amount of setup.

DD I think maybe GivTCP can run outside of HA? But perhaps I imagined that, and it probably still requires a certain amount of setup.

@Judith
Yes, GivTCP can be run standalone - it does not require HA. It even includes a web interface which can display cell voltages, I believe.

But yes, it does require some setup.

Rbor

Exactly what I thought Rob but it can't be a coincidence surely? I don't charge fully in the 'summer' because the solar finishes the job usually; the 50% covers me for poor weather days.

dave

Finally managed to get 3017 on my two 9.5Kw batteries after two attempts locally. Mind you it did not show up on the portal for an hour or so.

This is a summary of my two 8.2 kWh batteries from HA, upgraded last Thursday from 3015 to 3017.

Both batteries were made in 2nd half of 2022 and both can be charged and discharged at 3 kWh.
The right-hand battery was installed about 8 months after the left-hand battery and you can see the different 'number of cycles'.

Does anyone know what the 2 entries for 'Battery Design Capacity' means.
They both have a design capacity of 160
But the left-hand battery has greater actual capacity. Why?

And what does '160' mean? I cant make any sense of the entry.
There are no units.
Each cell has a voltage of ₃.2-3.3 V
With 16 cells, this fits with the total voltage ÷ 16.

In this screenshot, both batteries have the same SOC. With 3015, the secondary (on the right) usually lagged behind, sometimes by as much as 15%, but caught up near to the SOC extremities.

Thanks

Rob

Upgraded with no drama (AC-3.0 inverter and 8.2kWh battery). The new BMS firmware seems to have cured the issues I experienced after the D0.205-A0.205 inverter firmware update back in April, namely (1) repetitive minicharges when at 100% SOC and charge power was available, and (2) this leading to battery voltage ratcheting up until BMS overvoltage fault is triggered.

Before

After

Rbor And what does '160' mean?

It will be 160Ah

160Ah is the design capacity (multiply by a nominal voltage of 51.25V) gives your 8.2kWh capacity. The battery capacity (actual) is derived from the calibration delta low to high charge. When new it will be higher than the design as the battery ages it will decline.

My 9.5kWh battery has a design of 186Ah but actual is 200Ah.

Robgy Thanks.

I think my logic below is correct. Tell me if I am wrong.

So my left-hand battery would be A x V/1000 kWh
Design would be 160 x 53.125/1000 = 8.5 kWh.
My actual is 158.06 x 53.125 /1000 = 8.4 kWh.

The right-hand battery, which has done about 60% cycles as LH battery would be:
Design would be 160 x 53.327/1000 = 8.5 kWh.
My actual is 172.31 x 53.327/1000 = 9.2 kWh.

..... So is my right-hand battery really a 9.5 kWh battery?

Either way, one of the rules is that a number requires a unit!

I thought these batteries were manufactured greater than this so that they can be advertised as 8.2 kWh with DoD of 100% (i.e. 8.2 kWh really 10.25 kWh).

Rob

Rbor @PianSom @DD thanks for confirming my suspicions. Next silly question is there an Apple IOS compatible app?

Judith Yes - see https://community.givenergy.cloud/d/438-how-accurate-are-the-import-from-grid-metrics/16 - but unfortuantely it doesn't show the cell voltages. Source is available from https://www.bbcbasic.co.uk/inverter_source.html but I'd have no idea how to bundle it back up as an iOS app.

PaDav6 160Ah is the design capacity (multiply by a nominal voltage of 51.25V) gives your 8.2kWh capacity. The battery capacity (actual) is derived from the calibration delta low to high charge. When new it will be higher than the design as the battery ages it will decline.

My 9.5kWh battery has a design of 186Ah but actual is 200Ah.

Slightly worrying, both of mine the actual is below the design capacity.

My 9.5 has a design of 186Ah but actual is 182.1Ah, but sometimes has been briefly down to 181.8Ah over the last month. Installed in March its done 190 cycles.

My 5.2kW battery has a design of 102Ah and an actual of 96.85Ah. Installed in Jan 2023 its done 864 cycles (was working hard in the winter when it was the only battery I had and being used a lot with Agile import).

sunnyjoe Interesting.
Up until Thurs 17th, My AC3.0 inverter was also running D0.205-A0.205 with two 8.2 kWh batteries on 3015.
This is last Monday 12th Aug on 3015:

I upgraded from 3015 to 3017 on Thursday afternoon.

This was Friday 16th Aug on 3017:

The repetitive mini charges seem to have gone for me also with 3017. But early days .....

Rob

Just done 3017 update from 3015 mine went smoothly so far about 10-15min gen 1 3.6 and batterie's 9.5 and 5.2 . Was getting on 3015 BMS Overvoltage every time it went to 100% at least once a day but worked fine so will see if this stops it.

geoffreycoan Your 9.5 actual capacity is quite low given they start off at about 200Ah and would only be expected to "lose" about 10Ah of that during the first 6 months or so (SEI on anode, amongst other permanent chemical reactions).

FWIW my 9.5 went in mid-June 2023 and had 198Ah on install. This fell to 189Ah on the second calibration (Dec 23) after about 6 months and I suspect is nearer 186Ah now (occasional blip on GivTCP graphs to 186.16Ah). That's from 502 "battery cycles" as reported from GivTCP and 8.2MW battery throughput (again GivTCP).

Predbat is going to accelerate "aging" (resistance growth) in the cells, no way around that when a lot of the time its running full rate charge/discharge and potentially alternating them every 30 minutes.

I suppose you could argue that you're only "excercising your battery" and it should last longer 😉

Vestas it kind of is what it is. As you say, Predbat is exercising the battery, but that’s what it’s there for.

Maybe something for the future and a warranty claim if it diminishes too much, but for now not something I can do much about

geoffreycoan I was being somewhat facetious in terms of "excercising the battery" 😉

I assume you didn't notice GE's latest attempt (video) to minimise overvoltages?

I still think your 9.5 isn't right - its a big drop for 190 cycles.

I can't remember the history but didn't you have another and it failed?

What do the cell voltages look like at max charge/discharge rates?

Just had a look at my battery cycles and somehow yesterday I used 2 battery cycles.

Is this correct?

Hook I think GE define a "cycle" as enough energy to charge the battery. It can overlap days and probably that will cause the cycles to be out of sync one way or another so you get a sudden readjustment on the integer cycle count.

It's not far off the mark - I have 502 cycles over about 420 days and the rest are basically Agile "winter work" 🙂

My 9.5 is showing full 175Ah, design 186, remaining at 95% 166, voltage 53.1, 400 cycles which is a bit under one a day. Cell voltage between 3.303 and 3.317. This is on Rubikcubes app. It’s always said 186Ah.

SimonClarke Try getting them to run a calibration on it if the last one was done when temps were a lot colder around the battery.

Anyone know how long the rollout will take as it’s staged? Asking as no sign off 3017 for me yet, and it’s been a week now since I’ve seen posts about it…

Vestas its in a cupboard indoors so the temperature is probably 20 min and 30 max. It occasionally gets down to 4% and has then gone up a bit like 8% before running out again then charging to 100%. The numbers don’t seem to have varied much since we got it April 23.

Vestas I still think your 9.5 isn't right - its a big drop for 190 cycles.

I can't remember the history but didn't you have another and it failed?

What do the cell voltages look like at max charge/discharge rates?

I’ll keep an eye on it and see what happens over the next year or so.

Yes this is my second 9.5 battery, the failure of the first one was entirely of my making and I haven’t fessed up here as to what happened. It is a good tale so I do need to share it.

The 9.5 battery is 98% full with cell voltages between 3.343 and 3.346V, the 5.2 is 100% with voltages between 3.361 and 3.366V. I think the 9.5 was recalibrated the most recently when the inverter failed in April, the 5.2 was probably March.

gwynnej3
The new firmware D0.205-A0.205 and 3017 have been working well - no SOC or temperature drops and the inverter response very quick. So much so that the Zappi now doesn't get a look in until the batteries are full.

HOWEVER yesterday I allowed the batteries to charge to 100% from the grid as they were over 80% after sunny weather. And throughout the day they stayed at 100%. Turning on a load (kettle) any required excess came from the grid. In the evening when the sun went down there was no battery discharge, they stayed firmly at 100 percent (although the individual packs reported 97%).

I have reported this to GE support, waiting for a response. In the meantime I decided to power-cycle the inverter and this restored apparently normal operation.

I am wondering what will happen the next time the batteries get to 100% however?? That could be today as the sun is shining!

Battery traces from just now show the absence of dropouts, but the two packs are 11 percent apart.

I was under the impression that 3017 was not created for the SOC and Temp dips. my gen 2 system with 2 x 9.5Kw batteries is still getting the Temp dips with 3017, However as I can only see the secondary battery (Android App error) I can't confirm if the SOC dips are also happening.

Ortwin

I got a email saying they are aware and they are working on the app to fix the missing primary battery issue. It’s also not displaying on the IOS app too ! (??) 🤔

I don't get the impression there is any coherent software development and release strategy at GE, it seems to be totally random what gets developed, how thoroughly it's tested, what they break trying to fix something else, what they tell customers. The whole system consists of GE software/firmware running on GE hardware, yet they struggle to get it working reliably. Is any one person in charge?
My system is working quite well at the moment but after the stuck at 100% incident I don't fully trust it.

gwynnej3
Are the software/firmware writers UK based?
Do UK GivEnergy staff have any "control" over the writers, or is the UK really just a retail outlet of the products

SteveCook Can we see an organisation chart for the company is one place to start...

Hmmm. I now have the upgrade option appearing in my web portal. Wondering if I should click yes or not. Are the problems people are noting thus far confined to gen 2 batteries (ie 9.5 etc) or gen 1?

Smk82 No real issues with my 2 x Gen2 9.5's.

Smk82 No issues with my Gen 1 5.2

Finally updated, I’ve had no notifications from the battery about voltage for days so way happier.

Primary battery is now showing again in the iOS app

matttheotter You won't. They got "filtered out" a couple of weeks ago, so now nobody gets any overvoltage notifications..

Vestas I was still getting alerts just nothing to actually read.

They stopped when I updated to the beta firmware and 3017.

I updated to 3017 a few days ago on advice of GE support - was getting random SoC drops before. No battery recalibration needed which was welcome on a cloudy day.

Since though I'm still finding things acting oddly. Right now, for example, we're generating strongly and at 98% battery but everything's being exported rather than bringing the batteries up to 100%:

30 mins ago I stuck a few appliances on and it drew from the grid instead of from the batteries.

I'm not sure this firmware fixes everything - I suspect it'll be quickly followed by 3018. It's frustrating that the system never really acts exactly like it's supposed to. I guess the answer is to be like my parents who haven't logged in to their inverter once since having it installed 3 years ago and wouldn't notice any oddities

Dabz Check the Portal to see if any random changes have been made e.g. Eco mode turned off.

Have been on 3017 for a week or so, "nothing to report", which is good. No data drop outs (which I did use to get), no SoC drops (so far), batteries staying in sync, so all good.
Ge seems to be responsive to "problems in the wild", which ALWAYS occur with software of any ilk, no matter how thoroughly it is bench or beta tested. For example - Apple iOS/vehicle [insert marque of choice here] software/ Windows OS - I could go on.
We should be thankful that Ge take their problems on and do their best to fix them, even improve them and give us OTA updates FOC (speed of inverter response for example, that is really good now) - there are many manufacturers in this field about which the same could not be said.
Are they perfect? No. Are any suppliers? Ditto - no. Is it easy to criticise Ge? - apparently so.

Updated my 9.5 gen1 a week ago and I have started getting battery voltage high errors. Happens at 98% and won’t go higher. Plan is to give a ring on Tuesday and get them to recalibrate

Annoyingly no errors were reported on 3015

Gruff 57, 58v is quite high. Higher than it should be. Sounds like it needs a recalibration, yes

My 2 9.5s are often 15% apart. So as the fitst one hits 4%, battery discharge stops. I tend to fully discharge each day at 15p and recharge at night at 7p. Can't say 3017 has impressed me.

Gruff which inverter firmware do you have?

Hook I’ve got the Hy5.0 running D0.450-A0.451

Gruff may be worth requesting the beta firmware 190/193.

That and 3017 stopped my over voltage errors.

Since updating to 3017 the battery SOC values for the primary and secondary batteries have improved. Now the SOC variance is only around 3 to 8% rather than 20%, with the voltages within 0.5v. Now GE just need to work out the SOC and temperature dips.

Think I'm going to wait a while longer 3015 really upset my 9.5 battery and panels, GE sorted once I got through, presently everything is ok and working as it should, so unless I have too, will hold off for now.

AC Coupled 2 x 8.2kWh. Inverter on FW 205 battery on FW3015.

This is becoming silly now. I updated the inverter in June to 205, it really caused me multiple issues, eventually completely failed and ended with the inverter having to be replaced. An 8 week nightmare

Yesterday for the 3rd time in a week a the SOC dropped: went from 46% to 24% whilst charging from solar with little load. Co incidentally, or not, I'd finished a 2h 30m export about an hour before.

This morning it's gone the other way. The inverter set to charge in IOG 23:30 - 05:30 slot and was charging at full power until 03:20 then dropped to 500w SOC at 84%. It continued at this rate until 05:30 when the battery was at SOC 91%.

It then, whilst discharging, climbed to 100% in 15 minutes. After seeing multiple issues on here with FW 3017 I am reluctant to update to for now. I see 3018 is on the horizon so I'll hang on unless it gets any worse.

I can't explain the drop for you but the screenshot of the charging you show I believe is the FW protecting the battery. It charges at full rate until 84% but at that point the battery voltage is high (can see 55.6v at 05:18 snapshot) so it slows charging to try to prevent an overvoltage, the reduced rate is about 500W, it continues this until it hits the battery high voltage alarm at 57v (or a single cell hits 3.7v) which given you have 56.7v recorded at 5:28 happens almost exactly at the point your timed charge ends. Hitting this voltage alarm causes the battery to set itself to 100% SOC (because it knows for definite it is full) though it may display that as a graduated process, in this case over 10-15 minutes from 91% up.

There are various different experiences from people on 3017, for what it's worth my BMS overvoltage issues are much improved having installed it but obviously it's an individual's consideration as to what's best.

3019 was installed on my batteries for over voltage issues. And that has gone away now. Apart from those 3017 was ok for me.
What is this talk about 3018? Is it usual to go backwards with FW numbering.

Before updating my two 9.5Kw batteries to 3017, the primary battery always had a higher SOC than the secondary. However since the update both have been running a lot closer until today. My secondary battery is now showing a higher than the primary. Not sure if that is a coincidence but the primary still has the occasional high temperatur 60° on occasions when charging (outside location at 23°)

Just updated my two 9.5kW batteries and asked support to check both updated ok (since it doesn't list both separately).

Portal has a notification in the My Inverter section which doesn't seem to want to clear but otherwise seems ok.

Was glad to not have to do a full calibration in summer.

The weirdness continues. Following call with support, they requested leaving the battery to carry out its normal function, eco day, import at night. Still getting warnings at 98% voltage is at 57/58V

As it’s been a sunny day there’s been no need to draw on the battery until heat pump came on. After 5 mins of use the battery still hadn’t kicked in; drawing from grid. This is not normal.

Restarted inverter and bingo, battery started working.

One thing I have noticed with 3017 is that the batteries "leak" to grid when at 100%. i.e. there is always a small ( less that 50W) discharge to grid at 100%.

And again this morning. 54% SOC down to 28% whilst charging off the solar.

Seems to happen after a battery export session and when I turn off Eco to allow the solar to export.

mjkelly5a
Did you resolve this we have a 70-30% drop at 7am (no load or discharge shown).
I posted about it on the 3017 verdict thread btw
Cheers, Judith

Judith I did a battery calibration 2 weeks ago and not had the same issue since.

We’re away at present but when we return and the heating goes on we’ll no doubt go down to the lower limit.
If that does sort it a recal might be called for.

Judith Recals are a little drastic to "fix" this, and are likely only to be a temporary fix. The real answer to to exercise your battery. The technical issue is that the way SOC is calculated is by measuring the voltage of the battery. GE batteries have a very flat voltage curve. The difference between nearly full and nearly empty is two-thirds of a gnat's whisker.
GE has tried to work round this with various different firmware versions but have yet to arrive at a universally accepted good solution.
Your SOC drop is from high to low. The likely cause is that the battery has been at a high state of charge for a long time and the software lost track of the state of charge.
Around 08:45 (just as you turned the kettle on?) you asked for more power, the firmware realised there wasn't as much juice in the battery as it thought and reset the SOC to be much lower.
If you are able to regularly (at least once a week?) run your battery down to a very low SOC (10%?) you'll find the calculated SOC to be a lot more stable.
From firmware 3012 onwards you can force this recalculation by running the SOC to 20% or lower then using very little power for at least 30 minutes. See bullet 10 under BMS(3012) 05/05/2023 here: https://kb.givenergy.cloud/article.php?id=22

BobB Only up to 3015 (I think). 3017 doesn't work the same.

BobB This topic was covered here by GE

I was on 3012 for ages with no issues and foolishly updated a little while ago to 3015 and since then had loads of undervoltage notifications and SOC drops.

Updated to 3017 last week as I felt I had nothing to lose and it seems better, not perfect, but better at least for now....

Ra Is your experience of 3017 better or worse than your experience of 3012? I'm still on 3012 and I think I'm going to end up waiting for the rapid response inverter update and do battery and inverter at the same time (probably next year so the sun can help with the calibration charge).

BobB I guess you are on the old Gen 1 firmware?

I have two Gen 1 hybrids running the beta 191/193 fast response firmware and BMS 3015 on both of my batteries.

For me this has been very stable, I saw maybe one over voltage notification every few weeks (before the portal suppressed them) and I see zero evidence of sudden SoC drops, but I do cycle my batteries a lot.

Quite happy with what I have and don’t see a need to upgrade to BMS 3017

I upgraded to 3017 last night as I thought that because it was a battery firmware update, it would have to drain the battery and charge up to 100% for calibration. I did it at that time so it would charge during my off peak rates but once the update was done, it just resumed charging.

Anyway, seems to be behaving so far. Anything in particular I should keep an eye out for?

BobB Too early to tell at the moment. In hindsight, I shouldn't have moved from 3012 at all as I didn't have any issues with it. If it ain't broke, don't fix it.....how I wish I should have listened to myself when I was hovering over the upgrade button 🙁

geoffreycoan Hi Geoffrey. Yes. I'm on the old Gen 1 firmware (D0.450-A0.451) I'm an avid follower of the fast response threads; the Gen 1 fast response firmware still feels a while away from public release. I can imagine benefits from the fast response SW and, in my head, there's merit in updating the battery at the same time (presumably most testing is performed with the latest versions of everything).
If I recall correctly, the beta 191/193 fast response firmware arrived before 3017, so my "keep everything in line" thought doesn't apply to you.
@Ra - I feel your pain. I've been stuck on the "if it ain't broke, don't fix it" wagon for a year now. I do like the sound of the faster response inverter SW though...

Giv recently updated me to 3017 and I was already on 191/193. I was on 3012 with no problems and suspect they didn’t really have any particular reason to do it as the problem I was having was the app but so far I can’t tell the difference. Touches wood…

BobB If I recall correctly, the beta 191/193 fast response firmware arrived before 3017, so my "keep everything in line" thought doesn't apply to you

Yes 3015 was out at the time that 191/193 was released. I think I was one of the first if not the first to go onto the fast response firmware and I started the thread on it to share my experiences. It did contain fixes to resolve some of the problems in 3015.

Must be getting increasingly complicated for GivEnergy to test everything with all the different inverter and BMS versions now out there

BobB Just to add my two penny worth! 3017 seems to work fine with 191/193, but according to the Inverter app by Rubikcube, I've lost a bit of usable battery amp hours.
According to the app, my battery capacity by design is 186Ah and full is 194Ah.
After I was upgraded to 191/193 and on 3015, charging my 9.5Ah battery to full (100%) reached 188 Ah-194Ah each time. Now, on 3017, I only get 187Ah.
Although in the summer months I really reached the 4% at the bottom end (usually around 10%), I now have hit the bottom a few times, and the battery stops at around 6% SOC, and gradually creaps back till the Flux 2-5 slot kicks in.
It's not really an issue as the battery has done 644 cycles, but the loss of battery capacity is a slight annoyance, and 3017 has made it worse.

BobB the Gen 1 fast response firmware still feels a while away from public release

Is 3017 the Firmware reference or the planned release date?