Batteries out of sync - calibration needed?

26 comments started 2023-06-01 last 2023-06-03
GivEnergy ProductsBattery
T
#1 Tim Adams

Inverter SA2053G084
3 x 8.2kw batteries still seem to be significantly out of "sync" and have significantly different charge and discharge percentages even though I have updated the firmware.

I made a forced discharge that saw:
Primary 0%
Batt1 21%
Batt2 7%

and idled at
Primary 1%
Batt1 17%
Batt2 11%

@TheDragon (GivEnergy) I am wondering if a further calibration would help? Would an inverter firmware update help?
or any other thoughts?

In recent time the batteries have not fully discharged due to good weather & solar output so have mostly discharged to say 70% and back to full. When filling they typically showed:
Primary 98%
Batt1 100%
Batt2 92%
which is part of the reason I did a forced discharge. I'm not sure how close we should expect the 3 batteries to be?

#2 TheDragon (GivEnergy)

Id be interested to see the batteries behaviour at idle. as they should self balance via the OCV map

T
#3 Tim Adams

TheDragon (GivEnergy) many thanks!

ok... i will leave modes as:
Charge: Solar
Discharge: Dynamic

unless I should change?

with current good weather and low usage I would expect batteries to fill over next day or so.

Currently they're slowly filling at 1.15kwh - not idle (early and cloudy. sun expected in next hour or so).
Primary 3%
Batt1 18%
Batt2 11%

M
#4 matt-drummer

Mine are even `worse'

Primary 26%
Battery 1 17%
Battery 2 71%
Battery 3 96%

The first thing I will never understand is why primary, 1, 2 and 3, surely it should be primary, 2, 3 and 4. Even my 9 year old son thinks it's strange numbering!

I have been using lithium batteries in model aircraft for 25 years and we never judged the state of charge by a percentage, we always look at pack and individual cell voltage.

Do these percentages even mean anything?

The voltages on my batteries are all with 0.2 volts of each other, that must be how they are balanced with each other and how the percentage is calculated?

I have four 9.50 kWh batteries. The reason I have so many is to run an ASHP in the winter from solar and E7 electricity.

This time of year we barely touch the batteries and every day they have been fully charged by 09.00, they just sit there at 100% for months as we get enough solar until 20.00 to run the house entirely from solar. We use very little from the evening until the sun comes up the next morning.

I just let them run down to around 50% and then set a charge level of around 300w from solar just to keep the overall state of charge at around 50%. It works in total but each battery shows very different percentages.

Maybe I should just leave them at 100%?

I thought I was being smart, we always stored our model batteries at around 80%, a storage voltage of around 3.7 volts per cell. The chargers even had a routine for this exact purpose. Storing the batteries at 100% was seen as a bad thing and damaged the packs.

I assume the GE system has better management but I wish I could just tell it to keep each pack at a certain state of charge.

T
#5 Tim Adams

matt-drummer i'll let @TheDragon (GivEnergy) comment on the technicals - my guess is that you might need a calibration or forced discharge to get back into sync? from what ive seen the self-balance only happens at full discharge?

as separate thing... we also have ashp (16kw) and do not use E7 or other tariffs as the batteries only charge/discharge at 2.6kwh with our Gen1 Hy-5.0. so would still need significant grid during the day.

So we use Tracker import
currently Unit rate: 16.23 p/kWh
Standing charge: 42.77 p/day
Tracker has been well under gov caps all year. Not checked recent averages but probably close to 20p as a day/night average from 1 Jan to 31 May. and May average under 18p i expect

Locked in fixed export 1 year at 15p (I'm expecting export rates to be cut)

Not sure what E7 rates look like? but Tracker might be worth looking at

M
#6 matt-drummer

Like I said, I don't really know whether the percentages mean anything?

Presumably GE could explain it, I did ask them once and I got an answer that they should be within about 10% of each other.

As my pack voltages are close I would assume that their true state of charge is actually much closer than the percentages suggest.

I would assume it is a bit more complicated than this but are the percentages calculated by measuring how much energy is put into and leaves the pack? And then a calibration just takes them down to a set low voltage and tops them out at a set high voltage? The energy put back in is measured and then used to calculate the soc as a percentage.

With regards to my ASHP, my E7 rate is currently 15p (I'm on a fix) and my export is 15p, also fixed. As long as my E7 rate and my export rate are close it makes no difference to me. What I export for nine months pays for what I will import for the other three.

We don't use so much electricity and we have got quite good at managing the house load before we had batteries. I probably won't be able to avoid some day rate imports but I will do my best to manage it.

I want to add a second AC coupled charger which would solve my problem completely. I spoke with my installer this week and have decided to wait until the GE EMS is ready. Of course, it may never be ready.

I have tried to find anybody running two ac coupled chargers in parallel but haven't succeeded as yet.

T
#7 Tim Adams

please keep me updated if you get further with adding EMS and/or inverter as I think I will try to do that too... at some point!

I'm hoping that I am close to matching pv generation with usage over a year - around 8000kw. Using weather compensation on my ashp has helped a lot.
But of course hopelessly miss-matched summer vs winter! I reckon from around mid Apr to mid Oct I'm a net exporter

T
#8 Tim Adams

@TheDragon (GivEnergy) fyi solar charging currently showing 20%. should be sunny remainder of day
Primary 20%
Batt1 30%
Batt2 22%

i don't know if that's expected behaviour?

M
#9 matt-drummer

Tim Adams

My installer is aware of the EMS and I am on `pole position' to be informed when a solution is available.

I will import nothing for nine months of the year (apart from the small amounts get with batteries, around 4 kWh in May)

I generate more than I use annually so I am defenately a net exporter.

T
#10 Tim Adams

minor update:
aggregate 41%
Primary 44%
Batt1 47%
Batt2 36%

my imports in May were 113kwh of which 58kwh was to home and 55kwh to batteries
exports in May 381kwh of which 341kwh direct from generation and 40kwh from batteries
the filling and emptying of batteries with grid was probably calibration and attempts by me to empty/fill to get the self-balance to kick-in

M
#11 matt-drummer

My aggregate now is 57%, made up of

33%
21%
72%
100%

But the voltages are between 53.70 and 53.90

Temperatures are between 16.80 and 18.40 c

So, I'm not sure I really care about the reported percentages, the one at 21% is at 53.9 volts and the one at 100% is at 53.8 volts.

I think the percentage reported is untrue and unrepresentative of the real state of charge.

T
#12 Tim Adams

reminder to self... will add voltages!

aggregate 60%
Primary 63% 54.1v 34.4C
Batt1 64% 54.0v 33.0C
Batt2 52% 53.4v 29.2C

T
#13 Tim Adams

not sure it means anything but I see from the 3rd party Givenergy Monitor app that my batteries show significantly different Full Capacity

Primary 173Ah
Batt1 170Ah
Batt2 81Ah


T
#14 Tim Adams

for the record:
aggregate 69%
Primary 70% 54.3v 34.7C
Batt1 72% 54.3v 31.7C
Batt2 62% 53.7v 29.4C

Givenergy Support kicked off a calibration at about 16.10 - should ensure I don't lose out too much on solar production!
lets see how it looks in am

M
#15 matt-drummer

It's interesting that my batteries seem to be much cooler than yours, mine are all under 20c

Mine are in my garage that is attached to the house.

Where are yours?

What is the GivEnergy monitor app?

I don't have this and I have no idea what my individual cell voltages or the capacities are.

M
#17 matt-drummer

Tim Adams

I will check it out.

I am really curious to see individual cell data, it'll take me back to my model flying days!

Thank you.

R
#18 Rubikcube

To see battery data on the Android app, you can tap the battery bar at the bottom of the screen. If you have the Collect Stats setting turned on, there will also be a battery icon on the menu bar.

M
#19 matt-drummer

Rubikcube

Is that your app?

The link I was given didn't work and I can't find anything on google playstore n my phone

R
#20 Rubikcube

The app requires Android 5 or above. You won't find it if you have an old Android 4 phone. Also you need to be in the UK or Ireland. Try searching the playstore for givenergy and scroll down past the official app.

The link that Tim gave is correct.

T
#21 Tim Adams

Givenergy have run calibration overnight. just over 50% full now as it recharges. A few observations... I have some screenshots if useful that I can post. Perhaps @TheDragon (GivEnergy) could kindly check

From @Rubikcube Monitor:

  1. Batt 3 shows a Full capacity of 81Ah. Other 2 are 170Ah & 173Ah. is that ok?
  2. Batt1* shows cell 10 at 2.794V when all others around 3.11 to 3.15. is that ok?
    *Batt1 in Givenergy aka Batt2 in Rubik

At calibration kickoff
Primary 70% 54.3v 34.7C
Batt1 72% 54.3v 31.7C
Batt2 62% 53.7v 29.4C

at batteries emptied & idle calibration
aggregate 0%
Primary 0% 45.6v 38.1C
Batt1 0% 49.3v 30.6C is this ok?
Batt2 0% 45.3v 32.8C

Batt1* shows cell 10 at 2.413V when all others around 3.10 to 3.15. is that ok?
*Batt1 in Givenergy aka Batt2 in Rubik

During calibration
aggregate 50%
Primary 61% 53.8v 35.9C
Batt1 49% 53.8v 31.3C is this ok?
Batt2 43% 53.2v 28C

Cell 10 showing similar levels as all other cells 3.358 (others 3.358 - 3.366)

#22 TheDragon (GivEnergy)

Tim Adams cell 10 looks lazy, it dropped to the lower limit before the others had, so the battery didnt fully discharge. Leaving battery 1 at 49.3v instead of 45-46v so there will be some capacity loss, maybe the read on the Ah ready 81Ah instead of 102 if a 5.3.

so what will happen, battey 1 will spend its life higher SoC than battery 2. Have support check cell 2 with an engineer, as this may need to be investigated.
All cells should be together, apart from the very top and bottom, due to slightly differing characteristics.

Im busy today doing ENA approval work, so wont be on here much today, and this weekend ive private stuff on

T
#23 Tim Adams

TheDragon (GivEnergy) thanks for that.

I will ask about Cell10

any thoughts on why batt3 (Batt2 in Givenergy) shows Full capacity as 81Ah vs other 2 showing 170 & 173?

M
#24 matt-drummer

My batteries were at 50% this morning overall.

They are now at 100% and I have only added 5 kWh according to the app.

38 kWh of batteries should have taken 19 kWh.

I think the percentages displayed are completely untrue, after a long time of not fully charging or discharging it just loses track of the actual state of charge.

Battery voltage was a much better indication of reality.

I am going to download the app and look at individual cells.

M
#25 matt-drummer

Rubikcube

I downloaded it thanks.

My battery data all looks good, no duff cells and all four batteries are over their design capacities.

They all appear to be fully charged confirming that the percentages displayed in the GE app are completely untrue.

The voltages and temperatures shown in the GE app agree with Rubikcubes app

If I can work out how to get the data from my phone to post here I will, I don't know how to do that.

T
#26 Tim Adams

hmmmmmm! calibration started on Thurs 1 Jun at about 16.15
Discharge completed normally by around midnight same day.

Charging then took place and I believe reached "100%" SOC by noon 2 Jun and then followed final self balancing.

I have not received any notification that calibration is finished. The batteries appear to be discharging/charging in normal dynamic way.

No idea if calibration is finished successfully? or will need starting again (preferably start at end of day to maximise solar etc)