Master Battery - Lights Out this morning?

So this morning the master battery 5.2 was dead to the world and the slave 2.6 all lights green. AC GIV-AC3.0 coupled no Solar (on separate inverter)

No obvious reason, no error messages, and house pulling load from the grid. - Agile rates 🙁

So Isolated the batteries - powered off and restarted - Master Battery and Slave all green lights and appear to now be working fine. Washing machine just drained a chunk of battery.

They show fully charged so the charge worked last night completing just before 5am.

What can I check to try and work out what went wrong?

Firmware 3015
Software D0.535-A0.535

I have e-mailed GivEnergy support to see if they can help - but I am guessing they will say upgrade firmware.

I have avoided doing that as the system has been going fine since I got up to 3015 which I think was around June.

I tend to discharge fully and recharge fully most days so the SOC drift has been OK.

Any suggestions on what can I check to try and work out what went wrong?

M_J This happened to me a couple of times a few months ago when the battery went over-voltage during a charge. These are hidden from the Portal log now so I can't see the actual date. I waited a few hours and it came back online itself (I think once the voltage dropped back into range). There seems to be no long lasting effect.

2x5.2 battery running 3015
3.0 inverter running D0.204-A0.204

I have seen the master battery go red status and then get sorted but never lose all lights and connections.

Will of course monitor it, hopefully just a one off...

Well ok this morning after a couple of 30 minutes charging but not quite up to 100% as Agile rates quite expensive.

Will monitor and hope it was a one off quirk. No response from GE support to my email. Yet...

Just got this from GE support...

This looks to have been caused by a cell misalignment causing the battery to shut off. We have a new firmware (3018) which deals with these misalignments a lot better getting them back in alignment. This update can be set off from my side however it will require a recalibration of the batteries as part of this process. This recalibration will fully discharge the batteries to the house/grid then fully charge up from the grid. Is this okay for me to set off for you?

I have told them Not to update as I am far from convinced that the new firmware will help especially after following the threads here...

If it happens again / more frequently I might update...

Just happened again early morning after charging. Both batteries at 100%. The master battery switched off so again I powered down the slave.

Isolated all power waited a while and then restored power.

Then power button on master - it came back up beep and lights.

Then power button on slave. Same as above.

Will monitor it. TBF both batteries have been pretty good without being out of step too much.

Maybe a coincidence but I did several small charges early morning to match cheaper rates. Not sure if the master was stressed by that...

I am regretting not just getting one big battery as I don't think BMS can deal with two very well.

Had happened to me twice in the last week. Email support. Twice. No reply.

MrChaz

Hi what set up do you have?

Did you power it back up ok?

Gen 1 3.6 inverter
2x 8kW batteries.

Happens whenever it's cold it seems, turning it back on appears to have no issues. Very annoying

I am on Go at the moment and touch wood not had it go down. I have noticed it go red early morning after being charged up, but by the time I get up it's green again.

I think when I have been doing the Agile tariff dance with lots of charge and discharge it doesn't like that.

As Agile rates have been pretty poor I have stayed on Go and it's more stable.

If I were you and it's happening on a regular basis I would register a support call.

Well the £uck Up Fairy was listening and the master battery was dead this morning. All lights out. The slave was on ok.

So power off the slave, isolate the inverter. Wait a while.

Power on the master - it came up all green. Power on the slave all good too.

Push up the isolator switch and all back on ok.

Checked the batteries both were 100% full and I mean 100% 81Ah and 47Ah which was likely the cause of the shut down.

Eco mode had not been restored by the inverter app so I switched that back on.

All seems to be working fine.

I had noticed earlier in week that the master had hit 81Ah and went Red Status but after a while the BMS brought it back on line switching to Green

I guess I should run a calibration but I don't want to really as a bit of a paff...

Usually the master and slave report 100% at lower Ah levels not the actual capacity so the BMS might need a reminder to set levels again.

How often do folk think we should recalibrate? If at all? What is the best practice?

Quick update Solar has charged the batteries to "100%" Which actually means 76 Ah out of 81 Ah and 46 Ah out of 47 Ah...

So BMS is back to the usual percentages

Happened again this morning and again eco back on message from inverter app didn't get applied.

I guess I am going to have to schedule a recalibrate to see if that stops the battery from being overcharged.

I used the schedule recalibrate option last night and it has said it has completed and time will tell if I get a more stable master battery.

The disturbing thing is that before the calibration my master battery had full 81 Ah and design of 102 Ah.

After the calibration it has new values!!!

76 Ah Full and design still at 102 Ah..

So over night it has had 5 Ah taken off the full position!!! Just over 6% removed by the process?

Cycle count is 1213.

Well I will expect it to fully go to 76 Ah when charging but I guess it won't...

M_J How long have you had the battery? I ask because if you lose another 5% then its warranty time.

Vestas

Feb 2022 but it's a 21 battery so three years now

CE2119

I was surprised that a recalibrate would change the value. I know I usually don't fill to 81 Ah but with the recent battery shut downs because it has filled up to 81 Ah has prompted me to try a recalibration.

I think the last one was July last year and it didn't change the Full value...

M_J I’m not surprised that a recalibration changed the max capacity level. I think that’s the experience of @hoggy in his long term stats.

The recalibration measures what voltage is considered full and empty, so as the batteries degrade then they can store less charge, and the calibration measures this.

M_J Recalibrations nearly always change the capacity downwards. The only exception to that is when you perform a calibration in warmer temperatures. ie a battery at 30C ambient will hold a good bit more than a battery at 10C.

If you have GivTCP/HA then you can get a fair idea what the next calibration is going to do. Look at the "Battery Capacity" graph and look for blips downwards (just a few Ah, not to zero) - that'll most likely be the next capacity value. eg mine is currently sitting at 186.74 but blips down to 184.32 when its overvoltage. That holds true for all the calibrations that have been done on this battery.

Your battery won't charge to 76Ah either - it'll charge to 97% of that value, so 73.72Ah will be your new "100%" value.

M_J This is pure guesswork but if your battery capacity has truly decreased then the problems you were previously having might have been the inverter trying to squeeze in 81Ah to a battery that could now only take 76Ah. Truly worrying that you have approached what is supposed to be a 10 year limit in 3 years.

PS Was your calibration long and slow? I found that the normal calibration found less capacity to the 2nd one done at a lot lower current. It is possible to edit the charge and discharge currents through th eportal and app whilst the calibration is happening.

Windy Miller It'll be interesting to see what happens when his battery hits 70% of original capacity - according to the datasheet for the cells being used (on the 9.5kWh certainly) GE MUST disable the battery once it goes below 70% as the cells are no longer safe to use.

Edit - the relevant section of the datasheet....

The battery life is limited. Customers should establish an effective tracking system to monitor and record the internal resistance and capacity of each battery during its life. The measurement method and calculation method of internal resistance and capacity need to be discussed and agreed between the customer and EVE Power Co., Ltd. When the internal resistance of the battery in use exceeds 150% of the initial internal resistance of the battery or the capacity is less than 70% of the nominal capacity, the battery should not to operate. Violation of this requirement will exempt EVE Power Co., Ltd. from its responsibility for product quality assurance in accordance with the product sales agreement and this specification.

Vestas I have 2x5.2KWhr batteries each dated 2246 (2+ years old installed May 23, cycle count just under 600). Mine show 97Ah (master) and 93Ah (slave) from the design 102Ah which is not on the same trajectory as M_J. There must be a reason for this. (I haven't recalibrated my system since last May but also don't get the problems M_J has had). What other factors other than age might come into play?

Windy Miller Overcharging. You know, the notifications that GE hid/deleted from the portal.

If its overvolting all the time then that will reduce the battery capacity very rapidly. Likewise deep discharge (undervolting) will do much the same.

I don't know whether there's an additional adverse factor coming into play if you have two batteries and one is only half the size of the other. I could see that causing more overvoltage/undervoltage conditions if the battery SoC's are nonsense (as they frequently are).

Edit - if it wasn't for the utterly crappy GE BMS then the ideal for extending lifespan would be to cycle between 20% and 80% capacity. That's not possible as you'll end up with nonsense SoC values VERY rapidly.

Vestas Should regular over and undervolting be addressed by a re-calibration then? M_J was reluctant to do this I think because of losing capacity but could it have been the need for a re-calibration that was actually accelerating the decline in the batteries?

I was going to do a once a year calibration but perhaps it should be based on whether I think I'm getting over/undervolting (difficult to be sure since they hid the notifications).

Windy Miller Depends on firmware etc. I had to write an automation to stop my G1 inverter overcharging a 9.5kWh battery on v3015 f/w every day.

The G1 inverter just charged at full rate all the way to 100% and beyond. I managed to get GE to put 191/193 f/w on the inverter and now it uses a 10-step decrease in charge rate from 90-100% & overvolts maybe once a month. The majority of people on G1 inverters will most likely still have batteries overvolting every day (in summer at least) but of course they don't see that now as GE hid the notification.

Calibrations make no difference in the scenario I had.

Edit - oh and that snippet from the datasheet above is actually for the cells used in the 5.2kWh battery, not the 9.5kWh (although that says the same).

Windy Miller PS Was your calibration long and slow? I

I set it to 25 rather than 50 so slower but not super long as I wanted it done between the 12:30 to 5:30 Go Rates.

A bit worried if I have been inadvertently stressing the batteries. I think adding the 2.6 battery has not helped.

I seriously wouldn't advise anyone to add a second battery based upon these discussions about the BMS and it's ability to manage multiple batteries.

I will keep this thread updated as it is an effective audit log unless they kill this forum at some stage...

M_J Going through the thread again I am wondering if your previous calibration last year might have been a bit near the over volt limit. As GE hid the notifications you wouldn't have been aware of any overvolting until the cold weather reduced the battery capacity a bit causing the battery to reach its shutdown point. It seems to have reached this point several times and I wonder if all these were on particular cold nights?

Alarming though it may be perhaps the new calibration reducing the battery capacity will avoid the battery shutting down and hopefully avoid what appears to be an acceleration in the battery life cycle.

Also is your firmware one that is a bit more gentle with charge current at the top end as Vestas reports (I have 3015 which does this).

Windy Miller

D0.535-A0.535 and 3015 refused to move up to later firmware.

Battery is inside house and doesn't get cold as house kept to 15% minimum.

Hiding messages from the users is a crap move and another nail that says in future choose an open system you can monitor.

Ohhh well I have an escape ladder from my bedroom...

M_J Same battery firmware as me. My inverter is D0.204-A0.204 but I have an AC3.0 inverter.

There may be two approaches you could try...

1. Calibrate now when the house is coldest (overnight?) at the full discharge/charge current. Will likely return a reduced capacity but less likely to shutdown (a supposition).

or

2. Re-calibrate on a hot summers days at 10th current (will take 48+ hours like mine did) returning the maximum capacity but with the possibility of more shutdowns and accelerating battery ageing.

Premature failure should be addressed by GE but that in itself is a risk.

Just a few more years before your EV becomes the house battery - outside the house.

M_J Ohhh well I have an escape ladder from my bedroom...

LiFePO4 cells mainly failsafe unlike any lithium battery with cobalt in it so worst case you'll get some smoke (which should be avoided) but no "jet engine" battery fires like you've no doubt seen online with the likes of Tesla cars.

I'm just curious to see whether GE do actually shut it down at 69% original capacity in accordance with the cell maunfacturers instructions or not. I'm betting on NOT 😉

Edit - sorry but I thought we'd have to wait a LOT longer to find someone with a battery heading for that level.

Vestas The battery life is limited. Customers should establish an effective tracking system to monitor and record the internal resistance and capacity of each battery during its life. The measurement method and calculation method of internal resistance and capacity need to be discussed and agreed between the customer and EVE Power Co., Ltd

These are crazy words in the datasheet.

We are individually as customers to put in place a tracking mechanism to measure resistance and capacity. What I am to disconnect the battery every say 2 weeks, put my multi meter on the power connectors and measure the resistance?
Or do I need to take the front cover off, disconnect the bus bar between the BMS from the battery cells and measure the cell resistance directly?
And make sure we agree the measurement and tracking mechanism with EVE Power Co Ltd (presumably the Chinese manufacturer I have no relationship with) ?

Mad

geoffreycoan The "customer" in this instance is GE. This comes from the datasheet of the company which manufactures the individual cells (EVE Power in this instance) so they sell to GE 🙂

Edit - to clarify further (if it needs doing) then its GE's responsibility to disable the battery pack when it hits 69%.

Edit2 - Had a browse of various other Chinese LiFePO4 battery manufacturers sites and and it seems like any cells over 50Ah have a clause in the datasheet where the manufacturer REQUIRES the customer to disable the battery at 70% of initial capacity/150% of original resistance. Under 50Ah goes down to 60% initial capacity in some cases. LiFePO4 batteries have some very weird aging characteristics in terms of what SoC you cycle them at - for example if you cycle between 55-65% then that apparently ages the cell more than doing a full charge/discharge.

Vestas Your battery won't charge to 76Ah either - it'll charge to 97% of that value, so 73.72Ah will be your new "100%" value

Yep today with slow solar top up..

New 100% is not 100%

73 Ah and 45 Ah...

Will update thread if the battery shuts off again, but last night it was ok.

Well the recalibrate has not stopped the battery from being overcharged.

Lights out on the master battery this morning and it had charged it to 100%. 76Ah

If it happens again I will have to see if a BMS version update makes any difference.

The 3019 latest test version sounds promising from the few people reporting on it here.

M_J What a nuisance for you. I had hoped a calibration would sort it but perhaps the latest firmware is the answer. Nothing to lose really now. GE will have to do it for you whilst it is on test.

Windy Miller

Yep, I will see when it next switches off.

Hopefully by then the help desk team are back on track and I can load the firmware ok.

The last few BMS updates I have had to manually load them on to the USB which is a paff..

Happened again this morning.

I had a thought that if it's switching off due to overcharging perhaps I could stop that by setting battery maximum charge level.

So I have set it to 97% to see if it helps.

Of course I should not have to do that but I will see if it works.

The other alternative is to get it all to the latest firmware and if it happens with that then to try and get GE support to deal with it...

M_J this should really be on Giv to do, but I'd be intrigued by your BMS local log (not the portal stuff, the BMS records every single error since its birth) & SOH values of your battery. Sadly I'm not quite there yet (if at all) with getting this remote via the modbus side of things.

hoggy

Happy to provide info to you if I can...

Yes I should be on to GE but I have lost trust with them and I will be told to update to latest firmware.

Considering they are overloaded at the moment I would rather avoid dealing with them. The staff are friendly enough but each firmware update has not gone smoothly and it's draining trying to deal with it...

I will see if the % limit works...

Well despite setting max recharge to 97% the master battery was switched off on the morning of the 10th March.

So changed max recharge to 95% to see if that helps.

What I had noticed during the week of sunshine was that both batteries were being charged to full, so hitting 47 Ah and 76Ah. The stated Full capacity as displayed on Rubiks App.

So I am guessing that I have lost some more capacity in the Master, it has tried to over charge it and said stop - shut down the master...

I guess I could do another recalibration and lose more capacity, but that has not helped with the last time.

I will monitor more and see when it shuts down next, or if the 95% self imposed limit helps..

Ultimately I will need to upgrade firmware, but am holding off that as it is a PITA if I have to do the USB file shuffle like the last few times...

Well another wonderful sunny day here and despite having the setting that I will only charge to 95% for the batteries they have now filled to the brim...

76Ah and 47Ah

It would seem that setting 95% is not stopping them getting to 100%

I wonder when the next time it will kill the master battery...

M_J It would seem that setting 95% is not stopping them getting to 100%

what did you set to try to limit the battery charge level?

Certainly on my inverter there isn’t a max charge level, only a max AC charge level which is used when force charging from the grid. When in Eco mode the battery will always charge to full. You can reduce the charge rate which might be beneficial, but not AFAIK stop this from happening

geoffreycoan

On Rubik's control it's called charging limit to and a percentage level that I set to 95%

I could have misunderstood it's effect as I am in Eco mode.

Hmmm trying to upload image is not working?

Will try on PC later...

M_J On Rubik's control it's called charging limit to and a percentage level that I set to 95%

I’m sure this will be what’s called AC Charge Limit in the GivEnergy Portal - you can check and see if its set to 95%

Timed Charge under the portal set to 95% and seems to imply that only when operating under the timed feature it will stop at 95%

TBF I have not used the GE app since getting the far superior Rubik app. Oops just realised you meant looking at the Web Portal. BRB

Sorry had a meeting that I needed to rush out to.

Checking the portal it is called AC Charge Upper % and set at 95% I can see it only relates to specific charge windows.

So it won't stop the batteries from being fully charged unless it's a specific charging window of time.

I can confirm the charging limit in my app is just the "AC Charge Upper % limit" in the portal. The setting may function slightly differently depending on firmware version or if the system is Hybrid or AC coupled, I don't know exactly, that's up to GE.

On the subject of battery firmware, I'm running 3017 very successfully. Never tried 3015, upgraded 3013 to 3017 last year. I avoided 3015 because some people reported issues with over charging.

Rubikcube

Thanks for the advice on the firmware I will undertake an update in April when I have the bandwidth to deal with any fall out.

The last few updates have required me to do the USB dance which is a bit of a pita.

The release notes really didn't lead me to think it would solve the issue, but if you have noted issues from others on 3015 overcharging maybe it will solve it?

M_J I’ve remained on 3015 and haven’t noticed any adverse effects on mine, of course the lack of over voltage notifications in the portal doesn’t help

I stayed off 3017 due to the temperature related throttling introduced in that version, but then I see it happening on my 5.2 so must be down to inverter firmware as well