took long enough...

32 minutes ago, WannabeSolarSparky said:

took long enough...

Hope it sticks around for a while... I always get cheesed off when we have cloud cover... but fixing that by adding more panels... :-) and eventually re-doing the inverter end as well...

still a bit early, but spot the inconsistency.... what's not visible here is...

also a bit early (in the shade) but this (4 X 550W Jinko) added 10kWh to the mix yesterday... :-)

Oops, sunsynk integration stopped talking due to mbusd going south-ish... now, a 20-odd minutes later and the sun having moved on, things are looking betterer and more correct, since the Sunsynk data is now also up to date again...

and the Jinko's...

I hate winter, cold, rain, cloudy, etc, etc, etc🙂🙂

31 minutes ago, Kalahari Meerkat said:

Hope it sticks around for a while...

Yeah, me too... been painful having to part with some cash going to eskom.

6 minutes ago, WannabeSolarSparky said:

cash going to eskom.

that, luckily is not an option here anymore, for a few years now... with the panels added and to be added and more LiFePo4 cells under way... we should be able to be wasteful in the future, at least when the sun shines :-)

Also enjoying the sun

Loosing power on PV 2 because I positioned that string like an idiot and now a shadow is casting on the bottom of 2 panels

On 2025/07/08 at 9:04 AM, WannabeSolarSparky said:

took long enough...

That home assistant? Very nice dashboard

30 minutes ago, sunset1 said:

That home assistant? Very nice dashboard

Yes this is latest Home Assistant on RPi5 with nvme drive to make it nice and quick and to handle the sheer volume of live data, this is the latest version of my layouts with multiple tabs to access all the data and settings for the batteries and inverter all on a single dashboard.
No more running to the garage to check and adjust settings :)

Battery Screen.


Power Charts


Battery Settings - not all added yet.

Changed the main screen a bit to use the full power flows, looks better IMO 😄

Short video of the main screen in action to show off the almost real-time data.
No more worrying about the sometimes online sunsynk portal.

Edited July 9, 2025Jul 9 by WannabeSolarSparky

Been updating my one dashboard as well that is used on my tablet to get a quick overall view of the most important info. Added the individual battery info because of an issue I ran into requiring me to get 2 batteries warranty repaired. This way I get a visual indication in conjunction with my new rules I put in place to prevent a similar occurrence from happening again.

A day of (at least) 75% sunshine sees 4 x 5100wH batteries fully charged by 15h00

Testing a scheme where the house runs from Solar and Battery only from 16h00 to 08h00.

I want to avoid running the batteries below 70% SOC

12 hours ago, WannabeSolarSparky said:

Yes this is latest Home Assistant on RPi5 with nvme drive to make it nice and quick and to handle the sheer volume of live data, this is the latest version of my layouts with multiple tabs to access all the data and settings for the batteries and inverter all on a single dashboard.
No more running to the garage to check and adjust settings :)

Battery Screen.

Very cool and nice graphics. I must get a dashboard like that one day ;)

Im using my partner's discarded laptop with broken screen as my home assistant server

Been working a treat frigate and some sonoff devices. Have solat assistant connected to my axpert inverter

How are you pulling data from your jk bms and inverter into HS?

19 hours ago, -cK- said:

Been updating my one dashboard as well that is used on my tablet to get a quick overall view of the most important info. Added the individual battery info because of an issue I ran into requiring me to get 2 batteries warranty repaired. This way I get a visual indication in conjunction with my new rules I put in place to prevent a similar occurrence from happening again.

May I ask what the issue was you had with the batteries and what it shows you now that avoids future problems?

5 minutes ago, sunset1 said:

How are you pulling data from your jk bms and inverter into HS?

esp32 with a rs485|ttl converter and running the code from https://github.com/txubelaxu/esphome-jk-bms/tree/main

2 minutes ago, sunset1 said:

May I ask what the issue was you had with the batteries and what it shows you now that avoids future problems?

I had a few weak cells 4 to be exact (1 on one battery and 3 on the second battery) replaced the weak cells with new ones and added a 16th cell to each battery.

When you have graphs running which shows the cell voltages throughout the full charge/discharge cycle you can quickly spot the weak cells when the batteries are almost empty on the one end and when they are almost full on the other end.
The weak cells will jump out at you, you cannot miss them.
They are usually the ones that deviate a lot away from the good cells.
I will see if i can find my old posts where I had pics of that happening.

I never had sight of the cells before changing over to the JK BMS's so never even knew I had issues, I just thought my 2 batteries lasting 12 hours was normal.
They now last almost 24 hours if the sun behaves.

Cells get damaged quickly if your bms has the wrong settings. Most of the batteries factory bms settings are WRONG.... You need to do whatever you can to get into the bms to set them correctly to help ensure your cells last long :)
See the new cells indicated, here you can see they are stronger than the rest. Opposite will happen with weak cells.

Edited July 10, 2025Jul 10 by WannabeSolarSparky

I set my Sunsynk inverter to discharge down to 20% during the evenings however I did not know nor was I informed before hand the SOC you get on the inverter is only the avg SOC accross my batteries. You would expect the BMS to keep the battery levels in check accross all 4 but that is not the case it would seem. With it being winter I don't always get to 100% each day to "sync up" the SOC's accross the batteries, so every day you only partially charge and discharge the variance between the batteries gets bigger and bigger.

One would expect that if the inverter is set to 20% the BMS would prevent the individual batteries going below 20% but that was not the case and it would run things down until the avg SOC reach 20% irrespective if one or more of your batteries are already below 20%. So without knowing it 2 of my batteries was running down to the point the the batteries BMS would actually put the battery in a protected shutdown state on while I thought everything was cool as the inverter showed I was above 20% things was far from it. I only noticed it when I saw my SOC graph the one day drop from 20 to 18 instantly. Needless to say cells got damaged and is being replaced hence the 2 offline batteries.

Below you will see the massive gaps between the 4 SOC's that would creep in after 2 or more days not charging up to 100%. This would register as 36% on the inverter.

See how the SOC starts to deviate between the 4 batteries (this was after the damage was done but this shows what I'm looking for and monitoring now).

I have an automation set to adjust my SOC on my Slot 1 to match the avg SOC as soon any one of the batteries reach 20% or any one of the 64 cells reach 3volts or if any battery reaches 49V.

And if the variance between the max SOC and Min SOC is more than 10% it would set slot 5 to 100% and enable grid charge to force a top up of the batteries now.

A bit overkill but once bitten, twice shy.

Never let things go this far:

Edited July 10, 2025Jul 10 by -cK-

48 minutes ago, -cK- said:

I set my Sunsynk inverter to discharge down to 20% during the evenings however I did not know nor was I informed before hand the SOC you get on the inverter is only the avg SOC accross my batteries. You would expect the BMS to keep the battery levels in check accross all 4 but that is not the case it would seem. With it being winter I don't always get to 100% each day to "sync up" the SOC's accross the batteries, so every day you only partially charge and discharge the variance between the batteries gets bigger and bigger.

One would expect that if the inverter is set to 20% the BMS would prevent the individual batteries going below 20% but that was not the case and it would run things down until the avg SOC reach 20% irrespective if one or more of your batteries are already below 20%. So without knowing it 2 of my batteries was running down to the point the the batteries BMS would actually put the battery in a protected shutdown state on while I thought everything was cool as the inverter showed I was above 20% things was far from it. I only noticed it when I saw my SOC graph the one day drop from 20 to 18 instantly. Needless to say cells got damaged and is being replaced hence the 2 offline batteries.

Below you will see the massive gaps between the 4 SOC's that would creep in after 2 or more days not charging up to 100%. This would register as 36% on the inverter.

See how the SOC starts to deviate between the 4 batteries (this was after the damage was done but this shows what I'm looking for and monitoring now).

I have an automation set to adjust my SOC on my Slot 1 to match the avg SOC as soon any one of the batteries reach 20% or any one of the 64 cells reach 3volts or if any battery reaches 49V.

And if the variance between the max SOC and Min SOC is more than 10% it would set slot 5 to 100% and enable grid charge to force a top up of the batteries now.

A bit overkill but once bitten, twice shy.

Never let things go this far:

Interesting thanks for the explanation and good to know.

I sort of took it for granted that top tier inverters or most with communication to batteries would take care of this sort of thing..

Are there any inverter battery combos that take care of this automatically?

May I ask what batteries or bms you use?

Edited July 10, 2025Jul 10 by sunset1

1 hour ago, WannabeSolarSparky said:

I had a few weak cells 4 to be exact (1 on one battery and 3 on the second battery) replaced the weak cells with new ones and added a 16th cell to each battery.

When you have graphs running which shows the cell voltages throughout the full charge/discharge cycle you can quickly spot the weak cells when the batteries are almost empty on the one end and when they are almost full on the other end.
The weak cells will jump out at you, you cannot miss them.
They are usually the ones that deviate a lot away from the good cells.
I will see if i can find my old posts where I had pics of that happening.

I never had sight of the cells before changing over to the JK BMS's so never even knew I had issues, I just thought my 2 batteries lasting 12 hours was normal.
They now last almost 24 hours if the sun behaves.

Cells get damaged quickly if your bms has the wrong settings. Most of the batteries factory bms settings are WRONG.... You need to do whatever you can to get into the bms to set them correctly to help ensure your cells last long :)
See the new cells indicated, here you can see they are stronger than the rest. Opposite will happen with weak cells.

12hr to 24hrs is quite a change wow..well done for spotting that and setting up the graphs.

Makes me wonder how many people have under performing batteries.

Did you retro fit the jk bms to a seplos or similar?

42 minutes ago, sunset1 said:

Interesting thanks for the explanation and good to know.

I sort of took it for granted that top tier inverters or most with communication to batteries would take care of this sort of thing..

Are there any inverter battery combos that take care of this automatically?

May I ask what batteries or bms you use?

Inverter = Sunsynk

Batteries = uses Pace BMS

11 hours ago, -cK- said:

That is exactly what I was telling @sunset1 about how you will see a weak or bad cell when you have some nice graphs :)

On 2025/07/09 at 8:43 PM, WannabeSolarSparky said:

Changed the main screen a bit to use the full power flows, looks better IMO 😄

Using the Andy scale , very cool 😀

1 hour ago, MrBeauvedere said:

Using the Andy scale , very cool 😀

I could not help myself 🤣
He probably runs the most useful channel I know of when it comes to lifepo4 batteries and bms's
Definitely the 1st stop anyone should have when doing any sort of research about batteries :)

For anyone who has not yet seen Andy's channel... Here is the link.
https://www.youtube.com/@OffGridGarageAustralia

I have the same issue with one of my battery cells, it's been like that since day one. Cell number two. However, it has not deteriorated over time.

I can still charge it every day to 102 A/H, rated 105 A/H

Battery is almost 4 years old and has 1927 Cycles.

Make LBSA.