How closely do cells need capacity matched?

Diysolar123

Photon Sorcerer
Joined
Feb 28, 2021
Messages
662
I labeled the cells A through D so that I can clearly identify the cells independent of their position within the pack.

Due to having suspicions about what I was seeing, I took the time to order a battery capacity tester of the kind that Will Prowse recommends. With this tester, I was able to run synthetic "days" that mimic my real-world use of the battery while camping since the tester allows me to vary the load and it counts total energy dissipated which allows me to know when to stop discharging to mimic a day's worth of use. Additionally, I have run every cell individually from 3.65V down to 2.60V twice and have twice run the bank from 3.65V/cell down to the first cell hitting 2.50V/cell in order to measure total capacity.
this is the correct thing to do first...label and test...
so what are your results from this capacity test?
In addition there is one more test you need to perform, the self discharge test.
Charge battery up and let it sit with NOTHING connected for 3-4days; then charge it up again watching how much power you put in (which should not be much).
Make sure you are staying well within spec for both charging and discharging currents...

If your cells have met their advertised capacity during their capacity test, AND, do not have an internal discharge problem, then the problem is NOT the batteries.

if they are within a few percent of their capacity then they still should not drift so quickly UNLESS you are pulling a very heavy load...what is your load current?
If it is just a few percent, well, thats not really that big an issue. Batteries will naturally degrade that much within a year or if you have compressed your cells that will also cause an "Early" loss of capacity but extend the number of cycles you get over the long term.
Consider getting an active balancer, works great and will balance your pack to within milivolts; remember to turn off your bms balancer.
 

fafrd

Photon Sorcerer
Joined
Aug 11, 2020
Messages
2,445
OK, after extensive testing, I am back.
A lot to digest in your report, so I’ll take it a point at a time.

‘williaty’ said:
1) The bank cannot make its rated capacity. Discharged at the 20-hour rate with the cells being at 70F, the bank capacity is only 272Ah. I repeated the test twice to confirm this number.

From what you report below, ‘cell A’ is limiting pack capacity to 272Ah or 97% of 280Ah rating (with one other cell, ‘cell C’ on it’s heels).
2) The bulging problem got worse, not better, with deep discharges. Cells A and C have gotten so bad that I had to file a longer slot into the bus bars because they wouldn't reach between the studs.
Sounds as though you are not clamping in any way, is that correct?
3) I have been consistently torquing the nuts on the studs to 30in-lbs using a CDI Torque Products torque screwdriver. The driver was last calibrated in 2019 and has never required adjustment any of the other times I sent it out for calibration over the last 15 years. Even with being at the lowest torque setting I could find recommended in any of the datasheets for prismatic cells like this, I still ripped one of the studs out of Cell A.
The datasheets don’t include any torque specs for clamping busbars or lugs - the ‘anti-torsion’ spec of 8Nm they provide is the maximum torque that can be applied to the aluminum terminal before it will twist out of the cell. The tapped M6 aluminum threads put there by the reseller will strip long before that…

The only way you can avoid ripping out threads with solid busbars is to either clamp your cells with sufficient force or position them far enough apart for each cell to fully breath / expand during the charge/discharge cycle…
I am not confident this repair will last if I have to keep taking the bank apart to put it back into parallel and top balancing it every time I take a trip.
Your battery is not properly designed if you need to keep ‘taking it apart’.

There are ways to successfully recover a stripped thread, but you’ve got greater issues to address first. 3 threads of Helicoil is unlikely to hold.
4) The problems follow the cells regardless of which position they are in the pack. Cell A is the worst. It will always hit the high voltage cutoff first and always hit the low voltage cutoff first. It's clearly lower capacity than the other cells. However, Cell C is close on its heels. Cells A and C are never more than 10mV different when Cell A hits cutoff. Cells B and D are significantly larger capacity than A or C as they will be 250-350mV away from the low/high cutoff when Cell A triggers a shutdown.
Achieving a battery delivering 97% of rated initial capacity would be a good achievement, but you are a long way from that. What did you pay for these cells and who did you purchase them from?
5) The pack will slip out of balance in 4-6 synthetic "days" of testing. My real world use pulls 300-750Wh per day out of the pack.
If there is a high-degree of variability in self-discharge amongst your cells, that is a second factor that will limit capacity and dictate the ‘strength’ (current) of balancer you will need.

The easiest way to characterize self-discharge is to do it from the ‘bottom’ up (charge needed to being each cell back to 3.65V from where it has settled), but that is tough without the correct equipment.

If your typical storage period will be overnight, the hard way to characterize self discharge is to charge each cell to 3.65V, let it settle for 12 hours, and then repeat a capacity test.

‘3% / month’ or ~0.1% per day is the level of self-discharge to expect. So ~140mAh over 12 hours is a realistic level of self-discharge and a cell that is self-discharging at 3-5 times that rate means an additional ~280-560mAh lost overnight that your balancer will need replace daily.

An active balancer can achieve that amount of balance charge in several hours easily but that degree of mismatch in self-discharge may go beyond what the typical passive balance integrated into most BMSs can compensate for over a typical charge cycle.
The pack gains 60-90mV of imbalance per "day" when doing this. Eventually, Cell A will trigger the cell overvoltage protection on charging, usually with the bank voltage around 13.5V resulting in an imbalance between the highest and lowest cells of around 350mV.
60-90mV means very different amounts of charge well up into the knee above 3.5V versus down in the flats of after the cells have settled overnight.

Above 3.5V, 0.1% per ~40mV or ~7mAh/mV is a realistic sway, so 60-90mV would correspond to ~420 to 630mAh (3-5 times the self-discharge you’d expect over 12 hours).

Below 3.375V, 0.1% per ~0.4mV or ~700mAh/mV is a realistic swag, so 60-90 would correspond to 42-63Ah or 15-22.5% of cell capacity. If your cell A is losing that much versus the other cells through a charge/discharge cycle it is hopelessly defective…

With those results, here are my current thoughts:

a) The battery bank as-is isn't functional.
Where ‘as is’ includes using cells without a clamping fixture and connected with solid busbars, you are correct.
b) The battery bank cannot be "fixed" by replacing a single cell. It appears A and C are pretty close to matched and B and D may be pretty close.
It’s not the capacity of your cells that is your issue, is it primarily the way your battery has been assembled and may possibly be one or two cells with defectively-high self-discharge levels.

Once you have more carefully characterized the difference in self-discharge, that will tell you whether you actually need to replace any cells or not.

One easy way to address the issue within the testing methodology you’ve been using is your add an active balancer to the system. A 60-90mV voltage delta will translate to a 600-900mA active balance current, so easily compensated in under an hour if the voltage mismatch you’re seeing is up in the knee. If you can achieve a repeatable voltage mismatch across multiple charge/discharge cycles, at least you know what level of usable capacity you can count on from these cells.
c) The bulging issue is a serious problem since these were listed as Grade A Matched cells.
I’m suspecting you purchased these cells for ‘cheap’ (compared to purchasing true Grade A cells direct from the manufacturer EVE) from an aftermarket reseller as most of us did. Pretty much no ‘Grade A Matched’ cells you purchase from resellers on Alibaba.com/AliExpress are either ‘Grade A’ or ‘Matched’ in the way the manufacturer EVE uses those terms.

With a clamping fixture (as EVE now recommends), your issue with cell bulging will disappear. If you don’t want to bother with a clamping fixture, you need to space your cells with a big enough gap that they can freely bulge without causing mechanical stress on your solid busbars.

The other solution is to invest in flexible busbars in which case you can allow the pack to bulge/move/expand as it needs to without stripping out terminals.

(For what it is worth, I have both a 300KgF clamping fixture as well as custom 2/0 lugged battery cables to absorb any movement without causing stress on my terminals).
OK, can you guys think of anything I'm missing here or any other way to patch the problems I've got going?
When you talk about connecting with ‘the lowest torque’ you saw specified, I’m also worried your terminal connections have higher contact resistance than they should and probably high variability as well.

So my advice would be one of two paths:

Easy Path: invest a bit in an active balancer to continue testing as you have been to understand whether excessive self-discharge of Cell A is an actual problem or not. If it is (and possibly cell C as well, you will probably need to replace those one or two cells before you can build anything usable.

More Difficult Path: characterize the self-discharge rate of Cell A, Cell C, and whichever cell you consider to be be of your ‘best’ cells at the individual cell level. To be clear, this means measuring either the charge needed to ‘top off’ each cel to 3.65V alter a 12 hour or 72 hour ‘rest’ or measuring remaining capacity using your capacity meter after a rest of at least 72 hours if not longer (since I doubt the accuracy of those cheap capacity testers is below 1% / 3Ah).

Once to understand whether you truly have any defective cells or not (meaning truly unusable) and which battery type you want to build (meaning fixture or not and connection type), there is a path to get you there.

(Again, for what it is worth, I caused cell bulging and stripped a cell terminal before I know what I was doing. I painfully learned the best technique to recover a stripped terminal, built a clamping fixture, ditched the bundled solid busbars for custom-built 2/0 lugged battery cables, added a Heltec Active Balancer and have been successfully using my 560Ah 24V LiFePO battery for several months now…).
 

fafrd

Photon Sorcerer
Joined
Aug 11, 2020
Messages
2,445
this is the correct thing to do first...label and test...
so what are your results from this capacity test?
In addition there is one more test you need to perform, the self discharge test.
Precisely.
Charge battery up and let it sit with NOTHING connected for 3-4days; then charge it up again watching how much power you put in (which should not be much).
He’s been disassembling to top-balance then reassembling has battery. Performing self-discharge at the full-battery level is tough since there is so much non-linearity above the knee, but he can charge individual cells within the battery without disassembly as long as he is exceedingly careful (watch / manage the top-off with a multimeter connected and ‘time’ the seconds of charge current to reach 3.65V or even 3.5V).

If he does not have the wherewithal or equipment to do that or is not comfortable with the risk topping-off, ‘remaining capacity’ test at the individual cell level after 72-96 hour rest is the more painful but safer way to achieve he same goal…

‘Diysolar123’ said:
Make sure you are staying well within spec for both charging and discharging currents...
Screwed up the quoting below and don’t know how to fix - apologies.
‘Diysolar123’ said:
If your cells have met their advertised capacity during their capacity test, AND, do not have an internal discharge problem, then the problem is NOT the batteries.

if they are within a few percent of their capacity then they still should not drift so quickly UNLESS you are pulling a very heavy load...what is your load current? If it is just a few percent, well, thats not really that big an issue.

‘Diysolar123’ said:
Batteries will naturally degrade that much within a year or if you have compressed your cells that will also cause an "Early" loss of capacity but extend the number of cycles you get over the long term.
Is there a thread / reference on capacity loss from compression somewhere? I’ve added a 300KgF fixture to my cell but have not recharacterized capacity since. If there are any members who have characterized loss of capacity from clamping, I’m interested to know what others have measured…

‘Diysolar123’ said:
Consider getting an active balancer, works great and will balance your pack to within milivolts; remember to turn off your bms balancer.
If the BMS balancer cannot be disabled, it’s really not a big deal. The active balancer has a much higher balance current and the passive balancer typically automatically disables high in the knee (3.55V in the case of my Heltec BMS).

I’m my case, I discharge to LVD nightly (while daily charge never reaches Float and depends on sun / season), so a bottom-balanced battery made more sense. An active balancer was the only way to maintain a bottom-balanced battery.

But in general, use of an active balancer is absolutely to best way to deal with self-discharge mismatch exceeding the passive balance capability of typical BMSs.
 

Bleedingblue

Solar Enthusiast
Joined
May 12, 2020
Messages
393
Correct, listed as Grad A Matched 280Ah cells and I'm struggling to get them to be useful.


272 sounds pretty good to me. Struggling for them to be useful when you getting like 98% of capacity out of em. Can't expect too much
 

time2roll

Photon Sorcerer
Joined
Mar 20, 2021
Messages
2,418
In common use I would expect to use 40% to 70% of capacity. Needing to go 90%+ sounds like a second parallel battery is needed.
 

toms

Photon Sorcerer
Joined
Jul 24, 2020
Messages
807
a: Only ever assemble your pack once.

b: Skip parallel top balancing, it is a waste of time and energy, individually adjust cell capacity if your pack capacity requires urgent attention. (a properly set up BMS will maximise useable capacity over time)

c: Ensure your BMS has a greater balancing capacity than the difference in your cell self discharge.

d: Use an active balancer temporarily (or permanently if c isn’t achieved)

e: remember that individual cells in series (when the pack as a whole is isolated) can be treated as if the rest of the pack doesn’t exist.

f: it is possible to get 4 280ah cells and have a series pack capacity of less than 280ah.

The “parallel top balance before you assess the pack condition” is surely responsible for more LiFePO4 cell failures than any other failure mode by a very long stretch.
 

Short_Shot

Photon Sorcerer
Joined
Jul 13, 2021
Messages
1,820
a: Only ever assemble your pack once.

b: Skip parallel top balancing, it is a waste of time and energy, individually adjust cell capacity if your pack capacity requires urgent attention. (a properly set up BMS will maximise useable capacity over time)

c: Ensure your BMS has a greater balancing capacity than the difference in your cell self discharge.

d: Use an active balancer temporarily (or permanently if c isn’t achieved)

e: remember that individual cells in series (when the pack as a whole is isolated) can be treated as if the rest of the pack doesn’t exist.

f: it is possible to get 4 280ah cells and have a series pack capacity of less than 280ah.

The “parallel top balance before you assess the pack condition” is surely responsible for more LiFePO4 cell failures than any other failure mode by a very long stretch.

You're not technically wrong, if you can wait several days/weeks with large packs for that balance to be achieved and don't care about getting rated capacity from it until then that's great. Go nuts.

Many people can do this and will be just fine doing so.

You should however preface these statements with "this is my opinion and much of the rest of the world disagrees with me".

There are as many downsides to your suggestion as there are valid reasons for your suggestions.

All of the cons of parallel top balancing are avoided by using a bms and a relay, and you get the benefit of not having to wait ages and/or buy another component that can fail just as easily as the power supply could.

You say top balancing is a waste of time and energy (and presumably money in buying a power supply).

You then immediately suggests multiple ways to top balance things that cost time (arguably more time to achieve the end result) and energy setting up even more leads, and money in buying an active balancer. So clearly balance is important.

You also ignore the fact that most bms units with passive balance will take ages to do this, some may generate a lot of heat, and many active balancers can fail into a dangerous state and the bms can't do anything about it. That alone is enough reason to avoid an auxiliary balancer in my mind, and since likely you place importance on the idea that a power supply can fail or creep during top balance why wouldn't this logic apply to any other piece of equipment?

Have you any statistics on your last statement or are you just making the assumption because we "often" see people screw up the top balancing process and inflate their cells based purely on anecdotal evidence? You do know it can be done safely using the bms right? Thus negating the power supply concern.


If balance is important enough to go out of your way in the ways you suggested then it's important enough to do it at the start. Plenty of reasons why it might be preferable to do it up front.

Not the least of which is the fact that you cannot get an accurate capacity test out of your pack if you just slap it together and call it a day, nor can you even get full capacity out of it for some time until the pack becomes balanced.

If you're buying cells then you usually have a limited window to perform these capacity tests and waiting for passive balancing or even active balancing with a device may take too long.


I do agree with point F though.
 

Diysolar123

Photon Sorcerer
Joined
Feb 28, 2021
Messages
662
Precisely.

He’s been disassembling to top-balance then reassembling has battery. Performing self-discharge at the full-battery level is tough since there is so much non-linearity above the knee, but he can charge individual cells within the battery without disassembly as long as he is exceedingly careful (watch / manage the top-off with a multimeter connected and ‘time’ the seconds of charge current to reach 3.65V or even 3.5V).

If he does not have the wherewithal or equipment to do that or is not comfortable with the risk topping-off, ‘remaining capacity’ test at the individual cell level after 72-96 hour rest is the more painful but safer way to achieve he same goal…


Screwed up the quoting below and don’t know how to fix - apologies.
@fafrd
"Is there a thread / reference on capacity loss from compression somewhere? I’ve added a 300KgF fixture to my cell but have not recharacterized capacity since. If there are any members who have characterized loss of capacity from clamping, I’m interested to know what others have measured…"

here is where it looks like there is a "slight" initial loss of capacity.
This is from the eve 280 lifepo4 spec sheet page 9/11; looking at the graphs (and they change scales between the two so you need to take that into account) it looks like retention of 94% at just over 750 cycles without fixtures and 93% at just under 750 cycles with fixturing...
then again maybe its just the quality of the graphs...

1636454869123.png
 

toms

Photon Sorcerer
Joined
Jul 24, 2020
Messages
807
You say top balancing is a waste of time and energy (and presumably money in buying a power supply).

You really need to learn how to read. Too busy arguing to listen.

I said PARALLEL top balancing is a waste of time and energy.

With decent cells any top balancing is generally not required.

With crappy cells, any top balance that is applied will rapidly be undone by differing cell self discharge.

This notion that parallel top balancing is a “normal” thing is ludicrous. A few youtubers that have next to no idea what they are doing thought they were smart by coming up with a risky process to solve a problem that doesn’t exist.

I’d wager that less than 1% of LiFePO4 packs that have been assembled have been parallel top balanced, yet we have a couple of threads/week on this forum where cells have been damaged by the process.

I bet you can’t give one advantage parallel top balancing has over series charging followed by cell top up if required.
 

Sojourner1

Itinerant
Joined
Sep 30, 2019
Messages
603
Location
USA
My GBS 500ah 4s5p were never top balanced and put into operation 5.5 years ago. Used everyday 40% +/- DOD at various PSOC with over 1,800 cycles so far. The sense boards take care of the balancing at 3.55 vpc. 14.1v has been pretty much the absorb (6 minute event) setting since new, "float" 13.6v. ;)
 

Short_Shot

Photon Sorcerer
Joined
Jul 13, 2021
Messages
1,820
You really need to learn how to read. Too busy arguing to listen.

I said PARALLEL top balancing is a waste of time and energy.

With decent cells any top balancing is generally not required.

With crappy cells, any top balance that is applied will rapidly be undone by differing cell self discharge.

This notion that parallel top balancing is a “normal” thing is ludicrous. A few youtubers that have next to no idea what they are doing thought they were smart by coming up with a risky process to solve a problem that doesn’t exist.

I’d wager that less than 1% of LiFePO4 packs that have been assembled have been parallel top balanced, yet we have a couple of threads/week on this forum where cells have been damaged by the process.

I bet you can’t give one advantage parallel top balancing has over series charging followed by cell top up if required.
I can read just fine. Clearly you can't or are going out of your way to insist so, since you deleted all the context where I used the exact phrase "parallel".

Spare everyone the nonsense and try reading it again.

Screenshot_20211109-152651_Chrome.jpg

All of my crappy cells have maintained their top balance just fine, fyi, after *parallel* top balancing. Since you can't function without reading the word apparently.

If your statement that they will "rapidly" come out of balance was true then all of your statements about any method of top balance being done or its importance would be irrelevant, and yet the worse the cells the more important it actually is to do so.
 

fafrd

Photon Sorcerer
Joined
Aug 11, 2020
Messages
2,445
I can read just fine. Clearly you can't or are going out of your way to insist so, since you deleted all the context where I used the exact phrase "parallel".

Spare everyone the nonsense and try reading it again.

View attachment 71842

All of my crappy cells have maintained their top balance just fine, fyi, after *parallel* top balancing. Since you can't function without reading the word apparently.

If your statement that they will "rapidly" come out of balance was true then all of your statements about any method of top balance being done or its importance would be irrelevant, and yet the worse the cells the more important it actually is to do so.
I’m going to support Sojourner1 here.

If you’ve got high-quality well-matched cells and they arrived within a reasonable delta of each other, any half-decent BMS can get them balanced and keep them there through normal usage cycles.

If you’ve got poorly-matched cells, perfect top balancing is not going to do anything to stop the ‘stray’ that evolves from mismatched self-discharge and if the passive balance of your BMS is sufficient to handle that, it could have gotten a bit-too-far-unbalanced pack balanced to start with (do no gain from having top balanced).

If you’ve got poorly-matched cells that have differences in self-discharge big enough to overwhelm the passive balance of your BMS, you’re going to need an active balancer in any case, so you are back to the case above (your active balancer will get an somewhat-unbalanced pack balanced and will keep it there.

So bottom-line, only in the case that you have cells so poorly matched in terms of initial state of charge that you can’t make a usable battery out of them is there any advantage at all to top balancing at all. And even in that case, just directly charging the cell or cells that need to be boosted up into the same SOC ballpark as the average cells is a quicker and more effective way to get the battery up and running.

If I had a nickel for every thread I’ve seen asking whether overcharged and/or bloated cells can be recovered…. Top balancing a full pack takes so long and is always done without BMS protection so that the risk of it charging the pack too far outweighs any benefit (in most cases).

Charge individual cells up to 3.65V (attended) for a capacity test down to 2.5V then charge them back up for a specific number of Ah into the ‘knee’ (such as 20% or 30%, controlled by using a 10A power source for a timed number of hours.

That’s good enough to build a bottom-balanced battery which you can then assemble into a battery with BMS to charge up and see how close you are in terms of top balance.

If any cells are low, you can either let the balancer take care of it through cycling or, if there is a low-capacity cell you’ve decided to live with, you can drain it to match the others with a power resistor (which means that cell will always hit LVD before the others if you discharge far enough).

I wish I’d learned all these lessons before I blindly followed all of the conventional wisdom here on the Forum at the time and spent over a month balancing my 16 280Ah cells (stripping one thread in the process).
 

Short_Shot

Photon Sorcerer
Joined
Jul 13, 2021
Messages
1,820
I’m going to support Sojourner1 here.

If you’ve got high-quality well-matched cells and they arrived within a reasonable delta of each other, any half-decent BMS can get them balanced and keep them there through normal usage cycles.

If you’ve got poorly-matched cells, perfect top balancing is not going to do anything to stop the ‘stray’ that evolves from mismatched self-discharge and if the passive balance of your BMS is sufficient to handle that, it could have gotten a bit-too-far-unbalanced pack balanced to start with (do no gain from having top balanced).

If you’ve got poorly-matched cells that have differences in self-discharge big enough to overwhelm the passive balance of your BMS, you’re going to need an active balancer in any case, so you are back to the case above (your active balancer will get an somewhat-unbalanced pack balanced and will keep it there.

So bottom-line, only in the case that you have cells so poorly matched in terms of initial state of charge that you can’t make a usable battery out of them is there any advantage at all to top balancing at all. And even in that case, just directly charging the cell or cells that need to be boosted up into the same SOC ballpark as the average cells is a quicker and more effective way to get the battery up and running.

If I had a nickel for every thread I’ve seen asking whether overcharged and/or bloated cells can be recovered…. Top balancing a full pack takes so long and is always done without BMS protection so that the risk of it charging the pack too far outweighs any benefit (in most cases).

Charge individual cells up to 3.65V (attended) for a capacity test down to 2.5V then charge them back up for a specific number of Ah into the ‘knee’ (such as 20% or 30%, controlled by using a 10A power source for a timed number of hours.

That’s good enough to build a bottom-balanced battery which you can then assemble into a battery with BMS to charge up and see how close you are in terms of top balance.

If any cells are low, you can either let the balancer take care of it through cycling or, if there is a low-capacity cell you’ve decided to live with, you can drain it to match the others with a power resistor (which means that cell will always hit LVD before the others if you discharge far enough).

I wish I’d learned all these lessons before I blindly followed all of the conventional wisdom here on the Forum at the time and spent over a month balancing my 16 280Ah cells (stripping one thread in the process).
Which as I said is true if you don't care about getting an accurate capacity test right out of the gate, which is necessary for all the shitty aliexpress sellers so many people are dealing with.

You two are both making the assumption that people should and can only have great quality matched cells and then simply entirely discounting the possibility of used cells as irrelevant with the hand wave that they'll just immediately go out of balance.

It's definitely not true that cells that are "so bad that it matters" cannot be assembled into a usable pack. The simple fact is that perfectly viable used cells can come with a double digit percentage of soc imbalance right out of the box and yet don't fall out of balance after the top balancing is achieved, through whatever method.

You're just totally ignoring that middle ground case in favor of an absolute argument.

They do not do that, and not everyone has good matched cells right out of the gate.

My obviously used cells are maintaining balance just fine but there's no way in hell they would have come remotely close to achieving their maximum capacity without being balanced, and doing so saved weeks of slow BMS balancing.

Also as I've stated a few times now, parallel top balancing can be done with the bms.


Is there some slightly increased risk of mechanical damage by assembling a pack more than once? Yes.

Is there a risk of overvolting them? Yes.

Is there a way to prevent overvolting the? Yes.

Is that user error when the process fails? Also yes. Because it can be avoided.
 

Bleedingblue

Solar Enthusiast
Joined
May 12, 2020
Messages
393
You really need to learn how to read. Too busy arguing to listen.

I said PARALLEL top balancing is a waste of time and energy.

With decent cells any top balancing is generally not required.

With crappy cells, any top balance that is applied will rapidly be undone by differing cell self discharge.

This notion that parallel top balancing is a “normal” thing is ludicrous. A few youtubers that have next to no idea what they are doing thought they were smart by coming up with a risky process to solve a problem that doesn’t exist.

I’d wager that less than 1% of LiFePO4 packs that have been assembled have been parallel top balanced, yet we have a couple of threads/week on this forum where cells have been damaged by the process.

I bet you can’t give one advantage parallel top balancing has over series charging followed by cell top up if required.


I'm not going to parallel top balance my cells. I have 96 of them in all and it would take way too long to do so and I wasn't gonna risk blowing up my cells like a toad.
 

Short_Shot

Photon Sorcerer
Joined
Jul 13, 2021
Messages
1,820
I'm not going to parallel top balance my cells. I have 96 of them in all and it would take way too long to do so and I wasn't gonna risk blowing up my cells like a toad.
This is an acceptable choice as long as you understand and are OK with the results.

Not a big deal.

And as they have stated repeatedly is largely irrelevant with good matched cells anyways.
 

time2roll

Photon Sorcerer
Joined
Mar 20, 2021
Messages
2,418
Parallel top balancing will not fix low capacity or high internal resistance cells. Still stuck with a poorly performing battery.
The battery will be what it is regardless.
 

Short_Shot

Photon Sorcerer
Joined
Jul 13, 2021
Messages
1,820
Parallel top balancing will not fix low capacity or high internal resistance cells. Still stuck with a poorly performing battery.
The battery will be what it is regardless.
It will not fix higher internal resistance. It will not fix low capacity.

It will allow the pack to be driven solely by the lowest capacity cell however.

Top balancing will correct the issue where cell A reached 100% first on charge and cell B reached 0% first on discharge and the fact that the delta could be much less than either cell's full capacity.

Parallel top balancing achieves this out of the gate rather than waiting for the bms or active balancer. This means your first cycle in service will achieve the maximum capacity you can get from those cells.

The battery will not "be what it is regardless" until its balanced at one end or the cells. Doing it up front is important if you're potentially going to be chasing a refund from a supplier for low capacity cells.
 

fafrd

Photon Sorcerer
Joined
Aug 11, 2020
Messages
2,445
Which as I said is true if you don't care about getting an accurate capacity test right out of the gate, which is necessary for all the shitty aliexpress sellers so many people are dealing with.

You two are both making the assumption that people should and can only have great quality matched cells and then simply entirely discounting the possibility of used cells as irrelevant with the hand wave that they'll just immediately go out of balance.

It's definitely not true that cells that are "so bad that it matters" cannot be assembled into a usable pack. The simple fact is that perfectly viable used cells can come with a double digit percentage of soc imbalance right out of the box and yet don't fall out of balance after the top balancing is achieved, through whatever method.

You're just totally ignoring that middle ground case in favor of an absolute argument.

They do not do that, and not everyone has good matched cells right out of the gate.

My obviously used cells are maintaining balance just fine but there's no way in hell they would have come remotely close to achieving their maximum capacity without being balanced, and doing so saved weeks of slow BMS balancing.

Also as I've stated a few times now, parallel top balancing can be done with the bms.


Is there some slightly increased risk of mechanical damage by assembling a pack more than once? Yes.

Is there a risk of overvolting them? Yes.

Is there a way to prevent overvolting the? Yes.

Is that user error when the process fails? Also yes. Because it can be avoided.
I think you missed the point of what I was trying to express.

First, it was directed to the average bargain LiFePO4 cell customer off of AliExpress - not those rare members paying a huge premium to get true factory-direct Grade-A cells cells that have been properly matched and come with a capacity test report. Not those who got ripped off and received one or more severely under capacity cells. Used cell owners like you are also a perfect match for the members I’ve aimed my message at.

Second, it sounds as though you and I are in agreement that a proper capacity test of cells is the important first step to assess the quality of aftermarket or used sells from grey market resellers off of AliExpress. I didn’t say anything about how to get the cells each up to 3.65V and if you want to speed that up by charging in series through a BMS and then parallel charging all cells to 3.65V, fine.

My main point was that a properly-done capacity test from 3.65V to 2.5V on each cell can result in a perfectly fine bottom-balanced pack as long as attention is put into controlling recharge Ah so it is equal for all cells after discharge.

And the other key poin I was trying to get across is that having a bottom-balanced pack is as effective as having a top-balanced pack for the purposes of assembling the final battery - performing a top-balance again after discharging all cells to 2.5V is a total waste of time.

Assemble the series battery with BMS and bottom-balanced cells then charge it up in series and you are good to go (the BMS can take it from there, or the Active Balancer if one is needed due to self-discharge mismatch or worst-case, you can help things out manually by using a power resistor to manually drain the lowest-capacity cell until it match the others).

And when you state that ‘there is a way to prevent overcharge while top-balancing’, if that means anything other than speeding the charge process into the knee up by charging in series through a BMS or never charging in parallel with a voltage source set over 3.65V and waiting until charge current drops below some threshold, I’m all ears…
 

Short_Shot

Photon Sorcerer
Joined
Jul 13, 2021
Messages
1,820
You can use the bms to cut the negative side of the power going to a relay if it detects anything over 3.65v on the cell side.

Then your 3.65v supply has to run through the relay first, which the bms is able to shut off.

There's some screwing around to be done with how it's wired though.

There are also voltage sensing units on the market which can detect over voltage and shut off an output, if you don't mind a slight error of 0.1v they're very cheap.

Going to 3.75v for a bit before shutoff won't kill these cells, but you can also set the thing to cut off at 3.55 which should still be far enough into the knee that the residual capacity can easily be covered by the bms balancing and it prevents the inaccuracy of the device from allowing it over 3.65 in the end. It will also be a fairly negligible impact on your capacity test.


I absolutely suggest charging in series first to speed it up dramatically, then finishing with parallel. It's toms up there who is on a crusade against it.

It's the parallel part that he's so hot against doing, and while I understand the reasons why one would want to avoid it and the reasons why one CAN avoid it in certain situations the simple fact remains that some scenarios still require one to achieve top balance as quickly as possible before putting it into service.

His insistence that he personally doesn't need it and therefore nobody else should either is asinine.
 

Short_Shot

Photon Sorcerer
Joined
Jul 13, 2021
Messages
1,820
Here's one example of an adjustable voltage cutoff you could use.

LM YN Digital Voltage Comparator 0-100V Voltage Measurement Control Voltage Measuring Charge Discharge Protection, Overvoltage Undervoltage, Overload Protection (12V) https://www.amazon.com/dp/B01M1JEH7T/ref=cm_sw_r_fm_apa_glt_i_X8AWPP2ZW8G4VCR4WNK8

The catch is you will need to set up a circuit that stays off when the overvoltage is triggered so it doesn't detect the voltage drop (when it shuts off) then turn right back on again.

I'm half asleep at the moment but I believe a normally open auxiliary contact on the relay would do the job, so you would need a double pole relay then run the coil power through it.

When the cutoff is triggered the coil is powered down then the NO contact breaks the coil circuit so it can't turn back on.

You'd then just manually activate the relay to start the process.
 
Top