Issues balancing with Batrium BMS

[sunshine_eggo]

I'm a long time Batrium user.
Reading thru the thread, I don't recall reading that the cells were ever in balance so I ask.... Were the cells in balance when you 1st built the battery or did this imbalance occur after some time period?

I use Auto Level on Lithium-ion 260ah packs and it can take 4-6 days to 'tighten the balance' by 20mv and you've got 90mv max difference in the 1st post. Patience is good in these situations - especially if things are not getting worse

FYI - the overall battery voltage does not mean the individual cells (in series) will have a specific voltage in relation to each other. They can vary widely and is one of the reasons BMS's are important so you know what's happening.

Do you know the settings that reset the bypass mAh values in cell telemetry? I'm trying to disable it to get a good feel for what is actually being burned off.

 

[OffGridInTheCity]

Do you know the settings that reset the bypass mAh values in cell telemetry? I'm trying to disable it to get a good feel for what is actually being burned off.

I don't - I've found the data to be all over the place after a while. I'm WM4 + Longmons and it could be a bit different.

For example the trailer Batrium shows this:

with Bypass never used and AutoLevel completely off. I did an Auto Level 6 months ago but the "Initial BypDate" is all 2023... weird.


The Home powerwall is even weirder. No Bypass and last AutoLeve was several months ago. No data at all. But I recently power cycled Batrium.

 

[Zwy]

How can I increase the balance power of my k9's? This seems to be hardcoded at .72A from batrium

You can add fans to increase the on time for balancing. If you don't have fans, when the K9 hits the temp limit, it will cease balancing.

View attachment 132592

 

[hummmingbear]

There are a lot of cells that are getting about 500mAh of bleed off, so that's encouraging; however, the "Initial Byp date" has likely been reset again. I can't believe your cells have only balanced for less than 1 hour on average.

I wonder if it resets when I update settings? I don't understand...I set my absorption time to 3hrs so it is holding the cells at 55.2/3.45 for that amount of time. It's definitely longer than that.

I can't see all the cells, but those displayed look very consistent - all very near 3.46V.

Yes totally, the majority are in balance but there are about 3-4 outliers between 3.37 and 3.40

Can you sort by Min V with the lowest displayed first and post? Did you change the main Bypass Voltage to 3.50?

I did change bypass to 3.50. I'll have to post the chart tomorrow sorted by minV when it goes back into absorption.

Can you uncheck these?

Sure, what does unchecking bypass initial, final and session do here? I'll post again tomorrow, we should have enough sunshine for a full charge again.

 

[hummmingbear]

I'm a long time Batrium user.
Reading thru the thread, I don't recall reading that the cells were ever in balance so I ask.... Were the cells in balance when you 1st built the battery or did this imbalance occur after some time period? If they started out in balance and then dropped you might have some weak cells which would make balancing more difficult.

Great question, and yes these were all initially top-balanced. I slowly brought them up to 3.64v and used and active balancer. They were all in balance. They also were in balance at the beginning of december when I would get to a full charge. We had a very cloudy late december and january so these didn't get a full charge for some time until recent.

Patience is good in these situations - especially if things are not getting worse

This is what I'm hoping and betting on, the watchmon doesn't have a lot of balancing power...so I think these just need time. I should have left my active balancers installed in these but didn't want to sink more money into buying one for each pack.

FYI - the overall battery voltage does not mean the individual cells (in series) will have a specific voltage in relation to each other. They can vary widely and is one of the reasons BMS's are important so you know what's happening.

Yes, I think my first issue was realizing that the voltage measured at the shunt doesn't represent the voltage of each pack. I wish batrium did a better job of displaying data per-pack instead of the entire system.
I needed to change my charge controller to charge to 55.45 (measured at the shunt) which turns out to be 55.2 (my target voltage) measured at each battery. Voltage drop, resistance...etc

Appreciate you chiming in here, thank you

 

[hummmingbear]

You can add fans to increase the on time for balancing. If you don't have fans, when the K9 hits the temp limit, it will cease balancing.

It's been about 50F in my garage where they are stored so I've been ok. They are just getting to 60-65C when they have been balancing, so about 10C below where they shut off. I will probably install some fans before summer comes. I forget, do the fans plug directly into the k9 or do I need to wire them to the main board?

 

[Zwy]

With 4 packs, you would need an expansion board. Only one relay on the CORE and nothing on the K9. You could use something like this https://www.amazon.com/LM-YN-Thermo.../B076YLNHTH/ref=psdc_5006566011_t3_B076YD8H19 instead, I have a few of these in my truck camper system to control fans and heating pads for the battery, they work well and are cheap. I'm not certain if a 48V model is available but you will probably have to use a 48V to 24V converter for fan power anyway.

 

[hummmingbear]

With 4 packs, you would need an expansion board. Only one relay on the CORE and nothing on the K9. You could use something like this https://www.amazon.com/LM-YN-Thermo.../B076YLNHTH/ref=psdc_5006566011_t3_B076YD8H19 instead, I have a few of these in my truck camper system to control fans and heating pads for the battery, they work well and are cheap. I'm not certain if a 48V model is available but you will probably have to use a 48V to 24V converter for fan power anyway.

I do have the expansion board so I think those relays would work, just more wiring to get over to where it is. Thank ya

 

[sunshine_eggo]

I do have the expansion board so I think those relays would work, just more wiring to get over to where it is. Thank ya

Yep. I have a small 5V fan precariously perched blowing through the side of the cooling fins and powered by the expansion board 5V source 24/7. Eventually, I'll wire it properly.

 

[OffGridInTheCity]

Yep. I have a small 5V fan precariously perched blowing through the side of the cooling fins and powered by the expansion board 5V source 24/7. Eventually, I'll wire it properly.

One of the advantages of mons (e..g longmons in my case) is you can set the max current + max temp - and it works. It does slow down balancing - e.g. the 20mv over 4-6days on 260ah packs. However, these settings don't do anything with the K9 design.

 

[hummmingbear]

Sun is out and charging up again, more details to follow! Thanks to everyone for chiming in and helping me troubleshoot. My OCD will feel resolve once I regain balance

My watchmons are in the battery case I have so I can't blow a fan on them from the outside, I think I'll use the 25mm(?) fan that can mount over the heatsync.

 

[OffGridInTheCity]

My biggest concern (for you) is you may have some weak (or self-discharging) cells that will need to be replaced - e.g. 90mv is a lot. One should not have to constantly balance healthy cells so when you see this, it can be a warning. Time will tell.

 

[jtvt]

So I'm getting little to no progress with balancing here. For instance cell 48 was initially at 3.40. But it and the other low cells drop to 3.38 (edit, now 3.37).

Interesting. I see essentially the same voltage drop scenario here

Strange cell behavior

I've got a 280ah cell with some weird behavior. - When charging, it's voltage leads the other cells. I have absorb set to ~56.1 and balancing set at 3.51v. It reaches this before the other cells which do range with lowest at ~3.46v at that point. - When it hits the absorb/balancing voltage...

diysolarforum.com

I've never noticed this behavior before. In my case, the cell reaches set point voltage. But drops like your cells as the whole bank stays in CV and balances. I have other cells that are similar, but not to the same degree. It makes me want to look at busbar connections, but I'm fairly confident that is not the issue. I think it may be more of a balancing configuration problem perhaps combined with conditions that exacerbate cell behavior differences that lead to imbalance. I just came off an extended low sun period with SoC bouncing 20% <->60% and then full sun conditions with max amp charging. I'm speculating that in my situation, getting back into a more regular balancing schedule will be the fix.

 

[hummmingbear]

Interesting. I see essentially the same voltage drop scenario here

Strange cell behavior

I've got a 280ah cell with some weird behavior. - When charging, it's voltage leads the other cells. I have absorb set to ~56.1 and balancing set at 3.51v. It reaches this before the other cells which do range with lowest at ~3.46v at that point. - When it hits the absorb/balancing voltage...

diysolarforum.com

I've never noticed this behavior before. In my case, the cell reaches set point voltage. But drops like your cells as the whole bank stays in CV and balances. I have other cells that are similar, but not to the same degree. It makes me want to look at busbar connections, but I'm fairly confident that is not the issue. I think it may be more of a balancing configuration problem perhaps combined with conditions that exacerbate cell behavior differences that lead to imbalance. I just came off an extended low sun period with SoC bouncing 20% <->60% and then full sun conditions with max amp charging. I'm speculating that in my situation, getting back into a more regular balancing schedule will be the fix.

I think this is normal, that the shunt voltage is quite different than the voltage right at the pack +/-. I'm summing the values of each cell in a pack/battery (i get data from udp and send it to node-red) to get a more accurate reading of what the actual pack voltage is. The shunt is about .2v higher than the actual.

 

[hummmingbear]

My biggest concern (for you) is you may have some weak (or self-discharging) cells that will need to be replaced - e.g. 90mv is a lot. One should not have to constantly balance healthy cells so when you see this, it can be a warning. Time will tell.

We might be out of the hole here...20min into absorption and already looking better. Though the low cells then do go down lower later in absorption, each day is looking better...I think patience has been the key here

 

[sunshine_eggo]

Cell telemetry?

 

[jtvt]

Though the low cells then do go down lower later in absorption

That's what I'd like to understand better.

Later in absorption means lower amperage. Either the cells aren't seeing full amperage or the those cells need higher amperage to maintain the voltage.

Could be a red herring inherent to having cells that are not perfectly matched.

 

[hummmingbear]

Cell telemetry?

Sorted by mAh


Sorted by minV

 

[hummmingbear]

What I don't understand is why do the lower voltage cells have higher bypass mAh? Wouldn't the higher value cells be the ones burning off mAh? Or do I have it backwards?

 

[sunshine_eggo]

What I don't understand is why do the lower voltage cells have higher bypass mAh? Wouldn't the higher value cells be the ones burning off mAh? Or do I have it backwards?

Your intuition is correct. Given that the mAh values are resetting themselves, it's hard to compare them.

You should see more intuition-consistent behavior if you reduce the bypass banding to 0.01, it should force the unit to just work on the highest 0.01V cells.