Settings for b-grade cells. Having some challenges.

[Tyler702]

I have an 8S 200Ah DIY pack, and its become clear that I have 2 problem cells. 1 in particular reaches 3.65V while the others are still just a tad over 3.4V. Obviously the BMS charging switch turns off. Once the 1 cell, and the rest of the pack settles down, the charging switch turns back on, charging commences. This will happen several times in a 20 minute period, once I'm reaching full capacity. I've had to lower the charging limit on my epever CC to 28.5 in order to keep my run-aways from hitting over 3.65V. However, when I charge the same pack with my noco genius charger at the reduced rate of 10A, the runaway cells behave much better, and all 8 cells can reach near full capacity with little variation. In addition to the BMS balancing, I also have a Heltic8S (5A) active balancer, but both can't seem to keep up when the CC is putting 30-50 amps in at any given time. So my question is: How can I / or can I throttle the CC to reduce current once voltage has reached 28.5 (that seems to be the sweet spot to keep the run-aways from over voltage) then slowly bring the pack up to say 29V. I feel its somewhere in the charge limit / boost / equalize / float settings, but when I've tried to do what I think I need, I get parameter errors. Like float volts can't be above charge limit volts..etc..
Any advise what I can try?

 

[acdoctor]

Did you top balance these cells exactly according to the resources section here?

You don’t have to push those cells as hard as you are trying. 28 volts with some absorption time with get you to 95% SOC.

 

[Tyler702]

Did you top balance these cells exactly according to the resources section here?

You don’t have to push those cells as hard as you are trying. 28 volts with some absorption time with get you to 95% SOC.

I did top balance, yes. Had em offline for about 2 weeks on my bench and took my time. And yes, I'm not the kind to always charge to 100% every time. By 28V I guess they're close enough. I guess I was just trying to figure out how to lower the current once they start getting past 90% or so SOC. those runners really take off quickly at the top end. And like I said, the bms and active balancers just cant keep up.

 

[acdoctor]

As battery voltage climbs very close to charge voltage current with reduce.

 

[Substrate]

Sounds like a huge band-aid for two possibly bad / low capacity cells. When you made your 8S, you purchased at least one more cell as a spare right? (bad sarcasm, hang with me here...)

At this distance, a couple of diagnostic suggestions - if you swap the positions of the runners with some others, does the problem seem to follow the cell, or does it stay in the same mounting position? (I'm thinking high-resistance bus-bar / jumper contacts etc.

Even though you manually top-balanced, what happens if you charge with the Noco, and then *remove* the active balancer so you just have the bms in place?

Many variables - I'm thinking two dud cells, but would give these college-try diagnositics in the hope of gaining more info just in case ...

 

[Tyler702]

Sounds like a huge band-aid for two possibly bad / low capacity cells. When you made your 8S, you purchased at least one more cell as a spare right? (bad sarcasm, hang with me here...)

At this distance, a couple of diagnostic suggestions - if you swap the positions of the runners with some others, does the problem seem to follow the cell, or does it stay in the same mounting position? (I'm thinking high-resistance bus-bar / jumper contacts etc.

Even though you manually top-balanced, what happens if you charge with the Noco, and then *remove* the active balancer so you just have the bms in place?

Many variables - I'm thinking two dud cells, but would give these college-try diagnositics in the hope of gaining more info just in case ...

changed cell positions within the pack.. problem follows cell. So yup, I've got 2 dud cells. Good news is however, I broke out 4 of the good cells and added another 200Ah to my 12V milkcrate boondock camping setup. I think I'll use the 2 duds as target practice on my next outing.. Won't that be exciting...haha

 

[fafrd]

changed cell positions within the pack.. problem follows cell. So yup, I've got 2 dud cells. Good news is however, I broke out 4 of the good cells and added another 200Ah to my 12V milkcrate boondock camping setup. I think I'll use the 2 duds as target practice on my next outing.. Won't that be exciting...haha

You need to get an active balancer, retop-balance your pack, and then fully-discharge to assess capacity of your weakest cell.

As long as the sane cell is the first cell to reach fully-charged cutoff and first cell to reach fully-discharged cutoff, your getting as much capacity from your pack as you can.

A weak cell or two is not nearly as much of a problem as accelerated segregation of those cells. If you recheck capacity every 6 or 12 months and see that you are losing much higher % of rated capacity over 182 or 365 cycles than the datasheet specifies, your battery will quickly reach end of useful life.

Depending on how much capacity you are counting on, replacing any faster-degrading -cells by that point is probably a good idea…

I’ve also got 2 weak cells in my 8S pack and I’ve used an active balancer to confirm that the weak cells continue to limit overall capacity after one year of use, but they don’t seem to be degrading noticeably faster than the other cells, so I’m letting it ride…

 

[Substrate]

fafrd - he's got a Heltec active balancer already inline.

For me, an active balancer should only be used *once* to speed up the initial top balance on grossly out of balance cells, which he's already done with a Noco 10a as a low current source so as not to overwhelm the bleeders.

The possibility is that these are just too far gone as used / grade-b cells to be practical without reducing panel power itself.

Maybe put the best 4 together as a backup 12v system, and save up for better cells..

 

[Tyler702]

fafrd - he's got a Heltec active balancer already inline.

For me, an active balancer should only be used *once* to speed up the initial top balance on grossly out of balance cells, which he's already done with a Noco 10a as a low current source so as not to overwhelm the bleeders.

The possibility is that these are just too far gone as used / grade-b cells to be practical without reducing panel power itself.

Maybe put the best 4 together as a backup 12v system, and save up for better cells..

About that heltic active balancer... Not to get off topic, but I'm on my second one, and I have yet to meter anything over 900mA - 1 Amp during balancing, even when the cells are dramatically unbalanced. It's supposed to be a "5A" version, but I call bs on that. I've double checked terminal crimps and resistance on all wires from harness to ring terminals. Everything is in order, yet still nothing over 1A. I suppose I could've gotten 2 bad boards?? its possible. IDK but not too impressed.

 

[fafrd]

About that heltic active balancer... Not to get off topic, but I'm on my second one, and I have yet to meter anything over 900mA - 1 Amp during balancing, even when the cells are dramatically unbalanced. It's supposed to be a "5A" version, but I call bs on that. I've double checked terminal crimps and resistance on all wires from harness to ring terminals. Everything is in order, yet still nothing over 1A. I suppose I could've gotten 2 bad boards?? its possible. IDK but not too impressed.

I have that same Heltec Actve Balancer and am pretty happy with it. If I recall correctly, you only get 1A of active balance to the high cell when it is 100mA higher than the average of the pack, so getting the full 5A of balance current requires a 0.5V difference between the high cell and the average.

And I agree leaving it permanently connected is not the preferred approach.

The way I use mine is to occasionally do a ‘deep discharge’ to my lowest SOC of about 3.0V per cell or 24,0V for the full pack.

I then give the pack15-20 minutes to settle before checking for mismatch (my pack is bottom balanced).

If everything looks the same as the least check (meaning no more than 30mV of mismatch), I call it a day, but otherwise I’ll connect the active balancer overnight and remove it at first light.

I rebalanced after 6 months but I’m not even sure it was necessary.

I’ll reach my one-year anniversary in a week and will perform another check then.

I chose to bottom-balance my battery because I fully discharge to my low SOC limit nightly and only get anywhere close to fully-charging my battery on the very rare perfect day on the middle of summer.

If/when I ever conclude I have a problem with balance getting lost too quickly, I’d leave the Active Balancer always connected but control it through the relay input to only turn on after the battery reaches low voltage shutoff and the inverters shut down for the night.

As I stated earlier, a cell losing cycle capacity more quickly than all the others is the biggest problem to watch out for with defective cells. If they just have less capacity than the others but that more limited capacity is nit degrading much more quickly than it should, not a big deal.

If balance is getting lost so that the the runner start s running earlier and earlier while also no longer being the first cell to discharge, that’s a more serious problem than lesser but stable capacity, and that problem needs to be dealt with through daily rebalance (or replacing the cell).