Running parallel Overkill BMS and having issues

JaVid

Solar Enthusiast
Joined
Aug 5, 2020
Messages
71
Hi All!

I finally got my system installed and its been running for a few days, but I have been noticing an issue. Specs are : 24V Off Grid. Multiplus 24/3000/50, Victron Smart Solar 150/70 CC, (2) 272AH 24 V battery packs each with Overkill BMS.

I am noticing that my battery packs are not working equally. When I have a large load, pack #1 is supplying around 70% of needed watts. When I am charging, pack #1 takes around 70% of the amps and charges up quicker. This seems like a problem. All my BMS settings are the same and wire length, etc is the same for each.

Some background - I built these packs in spring and did capacity test on them both (they were fine) - then I charged each of them up to about 30-40% SOC because I knew it would be a bit till I installed them. I was not exact at all to what SOC I brought them up to before letting them sit. Then I installed these packs in my system last week without any further charging/balancing/etc. I thought they would equalize within a few days, but that doesn't seem to be happening. Right now I have my Charge controller set to 27.1 for absorb and float - I had them both set to 27.4V, but then I noticed pack #1 had a few cells taking off and starting to diverge as they got above 3.4V and pack #2 was behind and I wanted to play it safe and not send any cells running high...

Any ideas what is happening and ways to figure out the issue? I know (think) its not a big issue for the time being as I am not running huge loads, but it just doesn't seem right and I am not smart enough to figure out what the issue could be....

Thanks for any ideas and feedback!
 

Bob142

Build more, learn more.
Joined
Oct 31, 2019
Messages
1,313
Location
Rhode Island, USA
Do you have all your charge sources and loads connected to one battery pack? You should have your positives connected to one battery pack and negatives connected to the other in order to have them share the load equally.
 

JaVid

Solar Enthusiast
Joined
Aug 5, 2020
Messages
71
I am not sure I understand your question. If I had all positives to one 8s pack and negatives to other 8s pack, wouldn't they be wired in series? Mine is parallel. Maybe I am not understanding correctly.

I have all my charge sources and loads connected to Victron Lynx Power In busbar. I have positives from each 8S battery going to this busbar with equal length wire. I have both negatives (coming from BMS's) going to Victron Smart Shunt (same length wires), which is connected to the negative bus of Lynx.
 

Bob B

Photon Sorcerer
Joined
Sep 21, 2019
Messages
3,342
How hard would it be to reverse the 2 packs and see if the higher current remains in the same location .... or moves with the pack?
 

JaVid

Solar Enthusiast
Joined
Aug 5, 2020
Messages
71
How hard would it be to reverse the 2 packs and see if the higher current remains in the same location .... or moves with the pack?
This is a good idea, but would require some effort to take off balance lead connector from each BMS before switching the positive wires between packs - not super easy the way I have set up unfortunately. The only things between the packs and the Lynx busbar is (2) MIdnite MNEDC 175A breakers - one for each pack - seems unlikely the issue is with those?

Or maybe you mean switching the cells to the different BMS's?
 

Bob B

Photon Sorcerer
Joined
Sep 21, 2019
Messages
3,342
I was just thinking of reversing the whole pack ... if it stays with the pack .... then reverse the BMS. It might even go away when reversing if there is a borderline connection.
 

JaVid

Solar Enthusiast
Joined
Aug 5, 2020
Messages
71
Well, I am enough of an amateur, I can't tell if this means anything, but I just tested resistance between each positive battery to the Lynx connection. Very different! Pack #1 shows .5 OHM wheras pack #2 shows 11 OHM! I need to investigate! Unfortunately I have to run an errand - will report back what I find...
 

Bob B

Photon Sorcerer
Joined
Sep 21, 2019
Messages
3,342
Well, I am enough of an amateur, I can't tell if this means anything, but I just tested resistance between each positive battery to the Lynx connection. Very different! Pack #1 shows .5 OHM wheras pack #2 shows 11 OHM! I need to investigate! Unfortunately I have to run an errand - will report back what I find...
Good find ..... High likelihood that what is causing that is your problem.
 

JaVid

Solar Enthusiast
Joined
Aug 5, 2020
Messages
71
Well, after messing around with those circuit breakers and making sure connections were all clean, etc, I realized by turning them on and off and measuring OHMs that the reason I was seeing such a difference in OHMs was that I was near the top of charging and one pack was still getting charged (#2) and the other one wasn't (#1). Nothing changed when switching breakers between packs. I didn't realize load/charging would change the resistance - showing I am amateur I guess.

I guess the next step is to take each pack apart and check all connections? Pack #2 is staying really close together voltage-wise, but lagging in charging and not being used as much when there is load. Pack #1 is diverging a lot more, but getting charged quicker and being used more when there is load. When I top balanced these packs, they were both pretty good. Which pack would I look for problems in? Any other ideas to figure this out?
 

Bob B

Photon Sorcerer
Joined
Sep 21, 2019
Messages
3,342
I would systematically go thru all connection points, clean them, and apply a dielectric grease to prevent oxidation.
I use noalox on my battery .... don't remember the name of the other stuff I use on other contact points.
 

RCinFLA

Photon Sorcerer
Joined
Jun 21, 2020
Messages
1,025
Do some voltage drop measurements with loads.

1) Total battery series stack voltage at battery terminals for same (similar) current load.
2) Input to output of each BMS for same current load.
3) battery cabling voltage drop for same current load.
misc. breakers, shunts, terminal lugs.

27.1v absorb setting is not going to get you much of any balancing.

A YR1035+ meter is great for checking cabling, bus bar connections, breakers, and other very low resistances. Fairly accurate to below 0.1 milliohm. My 50mv/500A shunt reads 0.10 milliohms. My 75mv/500A shunt reads 0.15 millohms on meter. They are my YR1035+ calibration checks.
 

JaVid

Solar Enthusiast
Joined
Aug 5, 2020
Messages
71
I would systematically go thru all connection points, clean them, and apply a dielectric grease to prevent oxidation.
I use noalox on my battery .... don't remember the name of the other stuff I use on other contact points.
Thanks for response. When I assembled, I did the full cleaning, smoothing, and used thin part of OXguard on each terminal. I guess I should do that again just to be sure...
 

JaVid

Solar Enthusiast
Joined
Aug 5, 2020
Messages
71
Do some voltage drop measurements with loads.

1) Total battery series stack voltage at battery terminals for same (similar) current load.
2) Input to output of each BMS for same current load.
3) battery cabling voltage drop for same current load.
misc. breakers, shunts, terminal lugs.

27.1v absorb setting is not going to get you much of any balancing.

A YR1035+ meter is great for checking cabling, bus bar connections, breakers, and other very low resistances. Fairly accurate to below 0.1 milliohm. My 50mv/500A shunt reads 0.10 milliohms. My 75mv/500A shunt reads 0.15 millohms on meter. They are my YR1035+ calibration checks.

Thanks for update. I have 27.1v absorb just until I figure out where my issue is at...

I will do the voltage drop measurements. Not sure what #3 means exactly - does it mean to take voltage after each lug, breaker, shunt until I get to Lynx Bus Bar?

I will look into getting one of those YR1035+ meters....Thanks!
 

wholybee

Solar Addict
Joined
Feb 11, 2021
Messages
364
This can happen if the cells (all of them with both packs) are not matched. They may be tested to similar capacities, but if not also matched to the same internal resistance, one will charge faster than the other. Unless you need to draw close to the max rating of the pair of BMS's, it doesn't really matter. As the cells draw down well into the flat portion of the voltage curve, the other battery will take more load, and if you draw down into the lower knee, the "slower" battery will catch up and take most of the load until both batteries fully discharge at the same time. They can be way out of whack in the flat portion, but should come to the same SOC in the knees. At least that is the way it is expected to work.

You should do some capacity tests to completely discharge both batteries to confirm that. If you don't like the results, you could (after a top balance) mix up the cells, swapping some in one battery with some in the other. The goal would be to make both batteries have the same internal resistance. You might be able to find a configuration that works better than what you have.
 
Top