LiFePo4 Balancing Question (fake imbalance?)

[rocwandrer]

If one cell reaches 3.65v before the others, how much imbalance can there be when, after settling out for a few hours to dissipate the surface charge, all the cells read 3.38v?

In other words, can multiple LiFePo4 cells have less than 10 mV of resting voltage imbalance fully charged, and still really need to be balanced?

In a 6s pack with 8800 mah rated capacity, balancing took 1000 mah, even though the charger had just fully charged the pack without balancing the cells, and the resting voltages were all the same within my measurement accuracy. If I include the energy dissipated from the one higher SoC cell, that was really like only putting 800 mah into the pack at a maximum, but that is still a lot.

Is it possible I am wasting my time "balancing" when really all that is going on is the surface charge dissipation is so fast that a slight imbalance in the dissipation rate affects the detection of 3.65v, at low charge rates? I noticed it takes more total ah to balance the two packs in parallel than it does total between the two separately.

I might try to balance charge right after balance charging tomorrow, to test my hypothesis, but I'm still looking for input.

 

[amberwolf]

It depends on the individual cells. With LiFePO4, voltage is not a good indicator of State of Charge, especially at partial-charge levels, and is one reason why a balancing BMS or manual balancing or individual cell charging can be a good idea, especially as a pack ages (because the cells get farther and farther apart in internal resistance and/or capacity).

Sometimes during charging the internal resistance of higher cells makes them appear to be at a higher voltage while current is applied, but htey drop back down as soon as current is removed. Unfortunately that doesnt' mean they're full yet.

On your pack, I don't know if the cells are really balanced--you would need to do an individual-cell discharge test for capacity, on each cell, one at a time, to verify their actual charge state. If all of them discharge the same amount down to the LVC point, then they are balanced even if their final charged resting voltages are different. If they don't each have the same amount of charge in them, they aren't balanced, even if their final resting voltages are the same.

But typically, as long as they all eventually reach the full HVC voltage during charge, they shoudl all be as balalnced as is possible for their various internal differences, without individual cell charging to a specific cell SoC.

 

[dnmun]

i tried already to explain that but he just wants to use the voltage as the measure of the state of charge. he decided already that capacity doesn't mean anything.

i even explained why the voltage drops after they are charged, but he thinks that if a cell goes to 3.65V first that it is defective.

 

[rocwandrer]

i tried already to explain that but he just wants to use the voltage as the measure of the state of charge. he decided already that capacity doesn't mean anything.

i even explained why the voltage drops after they are charged, but he thinks that if a cell goes to 3.65V first that it is defective.

Not one thing in this post seems like an accurate representation of what I think, or more accurately what i was asking about, which is quite frustrating.

In that other thread, I was arguing with the lipo user that voltage can't be used for SoC with LiFePo4.

I think that with one cell arriving at fully charged consistently ahead of the others, there must be some issue. I was asking if the one that gets to 3.65 v first is the only good cell, not if it is a bad cell, though I don't claim that the way I phrased my question makes me any less ignorant of battery characteristics.

 

[rocwandrer]

It depends on the individual cells. With LiFePO4, voltage is not a good indicator of State of Charge, especially at partial-charge levels, and is one reason why a balancing BMS or manual balancing or individual cell charging can be a good idea, especially as a pack ages (because the cells get farther and farther apart in internal resistance and/or capacity).

Sometimes during charging the internal resistance of higher cells makes them appear to be at a higher voltage while current is applied, but htey drop back down as soon as current is removed. Unfortunately that doesnt' mean they're full yet.

On your pack, I don't know if the cells are really balanced--you would need to do an individual-cell discharge test for capacity, on each cell, one at a time, to verify their actual charge state. If all of them discharge the same amount down to the LVC point, then they are balanced even if their final charged resting voltages are different. If they don't each have the same amount of charge in them, they aren't balanced, even if their final resting voltages are the same.

But typically, as long as they all eventually reach the full HVC voltage during charge, they shoudl all be as balalnced as is possible for their various internal differences, without individual cell charging to a specific cell SoC.

Good stuff in there, thank you! What do you mean by manual balancing? Cell 5 in each pack does show higher voltage once it has been charging for a while, but it takes quite a while (and longer the longer I've had the batteries) to settle out to the same resting voltage as the other cells. Does that sound like higher internal resistance? The only way balancing is a waste is if cell 5 also has the least capacity, right?

You are the first person to explain, in a way that i think i understand, why I should find the total capacity of each cell. I think I've got 8 relatively gently charge/discharge cycles on these packs now. Something is changing, since the surface charge is fading much slower each cycle now. Should I wait longer before testing capacity? What cut off voltage per cell should I use for that test?

 

[dogman dan]

I would think they are broken in now. But in the first 4 cycles, ignore any differences you may have seen. Manual balancing would be charging one cell, or one group of paralell cells individualy. Easy to do if you have an RC charger. Just select 1s charge, and connect to just one cell.

As long as a cell has reached 3.5v, and holds it for a few min after you take the charger off, that cell is one I would consider fully charged. Same as another cell in the pack that went to 3.65, or even 3.8v. If the voltage dropped to 3.4v instantly when the charger was taken off, IMO that would indicate there is still some room in there for a few more wh.

Idealy new cells would not dissipate past 3.5v when the charger is taken off. But not all stuff is exactly ideal. For example, I just bought a small handfull of A123 ( or not?) cells that won't take a 4 amp discharge. They hold a charge great, but suck.

 

[docnjoj]

I'm running 12S A123 20 ah pack and have about 40 cycles. I had 1 cell that consistantly went low and shut off my LVC. I replaced that cell and balance charge every time and never go below 60%DOD. All works as it should now. It took about 10 cycles to determine there was a bad cell with Celllogs on board.
otherDoc

 

[rocwandrer]

So given: 1.) I'm not going to pull apart LiFePo4 flat packs to change out a bad cell, and 2.) it is cell 5 in each pack that reaches 3.65v before any of the others get above 3.45v.

I could charge the entire pack until cell 5 gets full, then charge the first part of the pack through the balance leads as a 4s, and then charge the last cell in the pack as 1s. Other than time, does this offer any advantage over using the iMAX B6 balance feature? Is repeatedly applying a tiny charge and a tiny discharge at the top of SoC every single charge cycle going to eventually have a negative impact on cell 5?

 

[scoot]

Not necessary to disassemble the pack in any way to top off a single cell. But you will probably need to make new lead/connection of some kind to attach to the individual cell(s) that need topping off. If you are unable to safely access the cell tabs/terminals with aligator clip style or similar connection, you could make the DIY lead with small enough pins to slip into the balance row connctor of the pack (using the corect pin assigments of course) to top off the cell via the balance wires that relate to each end of the cell. So with that approach, the charger should be set to 1s LIFE mode using 1 amp or less.

Or, you could wait perhaps several days or more for your imax charger (in balance mode) to finish the job... might even take longer than that depending on the offending cell's actual state of charge.

BTW, your other responders here really have been trying to give you correct/useful advice along the way. 8) Good luck.

 

[rocwandrer]

Or could I maybe do something with a resistor across cell 5's balancing taps, once I know about how many mah it needs to burn to balance the other cells up to it? That way i could just charge cell 5 a little slower by dumping some current around it? Obviously even if by some miracle this is a good idea, I wouldn't do it until I am comfortable with the pack consistency...

 

[rocwandrer]

Or, you could wait perhaps several days or more for your imax charger (in balance mode) to finish the job... might even take longer than that depending on the offending cell's actual state of charge.

The pack is only something less than the rated 4.4 ah, and the imbalance so far has only taken an hour or two per[10-hour] work day to correct. So it is not a really big deal to me. But I was wondering if this is a bad long term plan though. I'm asking if applying a lot of micro charge discharge cycles to one cell every day might be harmful to that cell?

Not necessary to disassemble the pack in any way to top off a single cell.

I know I was trying to say that I'm not planning to do soldering on the battery tabs to replace a bad or chronically imbalanced cell like otherDoc did.

So with that approach, the charger should be set to 1s LIFE mode using 1 amp or less.

I don't think this would be hard to do. I have a 6x6s parallel adapter, and I could just cut off a leg and use the appropriate taps. But that gets into a lot more daily maintenance than just sitting the battery on my desk and turning on the charger...

BTW, your other responders here really have been trying to give you correct/useful advice along the way. 8) Good luck.

I know. It is still quite frustrating that what I think I said is so frequently not what other read. And to the point that dnmun feels like others are wasting their time trying to help me. I think I need to write less, so it is harder to misunderstand what I do say, instead of trying to cover everything all at once.

 

[chvidgov.bc.ca]

Dumb newby question.... what exactly is "balanced", and what does that look like.

What I'd like it to mean is "all cells reach LVC" at the same time - I think.

So why is that very likely when all cells are brought to the same " state of charge", usually represented by a minimal spread of final voltages. Couldn't there be a considerable spread of voltages, with an apparently widely variant "state of charge", and yet the different discharge rates of the cells still lead to reaching LVC at the same time....what is the bms trying to accomplish, and why wouldn't balancing be better if you could put the same number of watt hours into each cell, even if they ended up at different voltages at the end. Or does that not make any sense? So, does aiming for top balance also maximize the chances of simultaneous LVC on the low end.

What relationship does the spread of top voltages have to the spread at the end? If you single-cell charge a bunch of cells, and then run them in series, will they all reach LVC at the same time? I find that hard to believe - so shouldn't the bms try to do a better job than that (purely single cell charging)?

 

[DAND214]

Reading all this and I don't think I read that you get all the cells balanced.

Let them charge till the charger shuts off just like others do when they balance thier LiPo's. I do, when I balance them. Which is not very often.

Dan

 

[SamTexas]

For example, I just bought a small handfull of A123 ( or not?) cells that won't take a 4 amp discharge. They hold a charge great, but suck.

What? What happens when you exceed 4 amps? Are you discharging through the BMS? Are these the same cells you bought from a member here a few weeks ago who bought from another member who claimed that the cells are genuine and new?

 

[rocwandrer]

Reading all this and I don't think I read that you get all the cells balanced.

Let them charge till the charger shuts off just like others do when they balance thier LiPo's. I do, when I balance them. Which is not very often.

Dan

Is that directed at me? If so, I have let the charger balance the cells to its satisfaction every charge save the last one. And then only because it hadn't finished and I was headed out to a meeting.

 

[scoot]

One of my 8s packs has a cell (a paralleled pair actually) that was damaged from overheating during some misguided discharge tests LOL. It exhibits about 90% total capacity compared to the other cells in the pack. I keep that cell in the pack because i never got around to swapping it out and i still have all the range i need anyway. The consequence when I let that pack sit for long periods (i.e., I do not keep the pack topped off on a daily basis), is that cell self discharges significantly and requires me to top off that cell if I want to ensure full usable range on any given ride. kind of annoying but apparently not enough for me to do anything about it at this point. The fact that my charger and pack's balance wires and have home made anderson power pole connectors installed on the ends makes things much more convenient for me when playing pack doctor.

So if you are really achieving "balance" (3.45 volts or higher for every cell) on most charges and still running into trouble may indicate that you have an under achiever cell to contend with. BTW, not sure if this has been mentioned to you yet, but avoiding deep discharges like more that 75% or more is important towards getting repeatable easily balanced charges even with some healthy packs.

 

[rocwandrer]

So if you are really achieving "balance" (3.45 volts or higher for every cell) on most charges and still running into trouble may indicate that you have an under achiever cell to contend with. BTW, not sure if this has been mentioned to you yet, but avoiding deep discharges like more that 75% or more is important towards getting repeatable easily balanced charges even with some healthy packs.

It does seem fair to assume 3.45v is full-ish, doesn't it? In that case, it only takes 200 mah to balance a 6s "8800 mah" pack after 5400 mah of discharge (2715 mah in 12s config). I think I haven't seen any problems with discharge yet because I haven't gotten a very deep discharge yet.

75% of rated capacity, or of the weakest cell's capacity? I had been using 80% as my thumb rule, but basing it on the rated capacity.