Looking for a cheap and simple bulk battery capacity test method

battery resistance testers such as this:
https://www.amazon.com/gp/aw/d/B07W3TCRYL/ref=psdcmw_14244451_t1_B0716ZL1SM#aw-udpv3-customer-reviews_feature_div
aren’t a good revealer of capacity or cycles left and I’m looking for an easy and cheap test rig I could assemble for multiple cells at once.

They’re lithium iron cells.
Can I charge them all to the same 3.6 full voltage, probably while in parallel to make it quick and easy, then connect them separately to an led (and maybe a transistor for greater voltage precision and maybe a resistor) and the bulb would turn off when the cell discharged to an exact voltage?




I read a transistor would have much better precision relating to voltage:
https://www.nutsvolts.com/magazine/article/build_a_01_accurate_voltage_reference

Capacity measurement ?..how accurate do you want ?
What is your reference, or are you just testing, comparing, and sorting cells.
What is wrong with a simple logging RC charger/discharge tester ?
Good luck trying to estimate the number of cycles left .!

I’d like to do like 40 cells at once. I Got 2000.

I want it accurate to within maybe 1-2%. Is that doable @hillhater.

I imagine would be a simple circuit and just requires precision current draw and voltage cut off. And a timer added of some sort would be ideal

Hummina Shadeeba said:

Can I charge them all to the same 3.6 full voltage, probably while in parallel to make it quick and easy, then connect them separately to an led (and maybe a transistor for greater voltage precision and maybe a resistor) and the bulb would turn off when the cell discharged to an exact voltage?

Click to expand...

All that would do is tell you that they reached a voltage, and that they aren't "dead", in that they had enough current delivery ability to drive your load for some amount of time (which you won't know unless you're sitting there watching them all continuously without blinking).

It won't tell you anything else.


If you want to actually test their characteristics, you'll need something that *measures* things, for each cell you are testing.

It must measure every characteristic that you want to know about, and you'll either need a separate one for every cell under test, or you will need something that can connect to and track as many cells as you want to test at once.


The only cheap thing I know of like that is a generic wattmeter that allows you to power it from a separate source (becuase a single cell is not enough voltage), though it doesn't have a cutoff so you'd still have to sit there the entire time so you don't overdischarge the cells.

You'd need one for each cell under test.

Then you need a load for each cell under test, as well. (lightbulb, resistor bank, whatever) It must draw at least as much current as your actual end-product will, or else you won't know how they cells behave in your application. (they could work fine at a small load, and flat out die under a higher one).

There are also various battery testers out there, depending on exactly what you want to characterize, and the time you ahve to spend, and your budget.

Less budget means more of your own time.


There are a number of "cell test" and "load test" type threads on the forum, that you should look up and read, to see some of what is required and what you might be able to do.

I looked up cel test and nothing related came up


I think ur over thinking it.
If possible to get a very accurate and consistent current, let’s say 1 amp continuously, (maybe hard to do as the cell voltage will be dropping) and measure the time discharging that current from an exact top voltage to an exact bottom voltage the time taken to get top to bottom will give a good representation of the cell’s capacity.
Yes adding a timer to each discharger would be much nicer and wonder what the cheapest way to implement that would be

There are many ways to get an accurate capacity SoH% measurement.

The cheaper ones are of course a bit less precise, but fundamentally all you need is a CC dummy load and a stopwatch.

Only measuring each individual cell is meaningful, I assume you don't plan on only testing pack capacity.

Since Peukert is so close to 1.0 for LI chemistries, the 20-hours per test that is the standard for lead (CC load set to 0.05C) is thankfully not required.

You could even use 1C, say 40-50min per cell but of course would not get near the cell **rated** capacity.

Doing multiple cells at a time is certainly possible, but doing five tests concurrently requires five independent setups, five time the cost.

Hardware designed to automate these test is not cheap, and independently doing 40!! at a time would certainly cost thousands.

https://endless-sphere.com/forums/viewtopic.php?p=1488274#p1488274

https://www.rcgroups.com/forums/showthread.php?1107072-How-to-build-a-constant-current-discharger

https://www.rcgroups.com/forums/showthread.php?2406395-An-effective-variable-load-automatic-LiPo-discharger/page32#post40969831

https://hackaday.com/2020/02/11/build-your-own-active-load
https://hackaday.com/2018/09/06/zpb30a1-electronic-load-gets-an-open-firmware
https://hackaday.com/2017/02/28/beefy-100-amp-electronic-load-uses-two-mosfets
https://hackaday.com/2013/10/28/building-a-dc-constant-currentpower-electric-load

https://youtu.be/8xX2SVcItOA

https://youtu.be/htY2mHTx3kQ

https://www.rcgroups.com/forums/showthread.php?3490399-High-Power-LiPo-Discharger-Project

https://www.rcgroups.com/forums/showthread.php?p=44052829

https://www.rcgroups.com/forums/showthread.php?p=43950349

Thanks for all the info and I haven’t gone through it all even yet but wanted to get your thoughts on possibly using just a simple resistance as the dummy load. Ideal would be a load that kept the current consistent but in the interest of keeping it simple and cheap I figured I could time a cell I know to have full capacity on this variable load and whatever time it took to become depleted I could set as the standard. I can do the same with a cell I knew to be maybe 95% of rated capacity and similarly get a time till discharged and can compare these times with the cells I test.

Assuming this varying load method can work that leaves me still needing a timer of some sort and an accurate voltage at which the timer and discharging of the cell would stop. What you recommend for these parts? I’ll look through all ur links now.

Resistance will vary the current according to many variables, so not enough consistency.

There are little 3-bulb testers, put a pot/dimmer on it watching an ammeter might get pretty close.

Some Duo channel chargers you can put a resistor or depleted battery on the other channel and it will automate the start / stop and even data log the curves to a PC.

If you're not spending any money then you need to sit and watch I think.

john61ct said:

Resistance will vary the current according to many variables, so not enough consistency.

There are little 3-bulb testers, put a pot/dimmer on it watching an ammeter might get pretty close.

Click to expand...

With just using a resistor to discharge the cell the current will be dropping as the cell voltage drops of course but assuming the cells all had the same starting voltage and resistance and the circuit was a consistent temp what else would I need to consider that would effect the current drawn?

And then I’d just need a very accurate voltage the discharging would stop and what would be the simplest timer?

Hummina Shadeeba said:

assuming the cells all had the same starting voltage and resistance

Click to expand...

If you're going to assume the same resistance, then you should also assume all the cells are the same in every other property.

It's unlikely that they will all be the same in any property, otherwise you have no need to test them at all.

If they are different capacities, the likelihood is they are also different resistances.

This also means they will discharge at different rates based on a simple fixed-resistance load.

I meant if hypothetically the cells’ resistances and temp were all the same what other variable would alter the current drawn when connected to a simple resistor thereby ruining a capacity comparison.

State of Health is what varies, might seem a nebulous concept, but

the actual capacity is the primary indicator, but just like ESIR, the best way is to be able to compare to when first put into use

or really, after capacity has reached its peak, usually within the first dozen cycles or so if commissioned / broken in properly.

All of the variable need to be eliminated across time and between cells as much as possible, for accuracy is important.

There is inevitably going to be some variability / inaccuracy, but allowing any on purpose will IMO make going to all that trouble more or less wasted effort.

What u mean broken in properly. I’ve never heard of that for lithium batteries.

And what u mean esir


what would be a variable in determining the capacity of two cells in relation to each other if both are of the same model, temp, and resistance and both are put to the same resistance to discharge until empty.

> What u mean broken in properly.

Commissioning, break-in period, battery pack’s capacity will be activated to its full capacity after X cycles. Best to keep C-rates low before then. See SEI formation.

> And what u mean esir

internal resistance, GIYF

> what would be a variable in determining the capacity of two cells in relation to each other if both are of the same model, temp, and resistance

No two cells are ever the same anything!

If you set up an accurate test rig / protocol, then compare to what you're considering, you will see.

john61ct said:

> What u mean broken in properly.

Commissioning, break-in period, battery pack’s capacity will be activated to its full capacity after X cycles. Best to keep C-rates low before then. See SEI formation.

> And what u mean esir

internal resistance, GIYF

> what would be a variable in determining the capacity of two cells in relation to each other if both are of the same model, temp, and resistance

No two cells are ever the same anything!

If you set up an accurate test rig / protocol, then compare to what you're considering, you will see.

Click to expand...

That’s why I say hypothetically. The point is to discover what the other variables would be. IF running two identical resistance and temp cells through the same simple resistance till discharged what would other variables be that effect the time till totally discharged. Assuming there are other variables. I don’t know what they would be

Esir doesn’t show up in google but guess just means resistance.

Hummina Shadeeba said:

That’s why I say hypothetically. The point is to discover what the other variables would be. IF running two identical resistance and temp cells through the same simple resistance till discharged what would other variables be that world effect the time till total discharged. Assuming there are other variables. I don’t know what they would be

Click to expand...

We're trying to help you by *telling* you what the other variables are, so you don't have to waste your time and money and resources figuring it out.

You don't have to discover them, they've been known for years (decades), if you just read what others have already posted about testing batteries.

Reduced capacity and increasing IR are just **indicators** of EoL approaching

the only variables accessible without electron microscopes other lab gear worth tens of thousands.

You need to as accurately as possible measure capacity in order to infer actual SoH

Doing so "well enough" to make go/no-go decisions with a pile of used cells

does not have to cost more than IMO $500

For your standard, maybe under $200.

Trying to do it for under $50 would be "pioneer" territory, as I said, try it out lets us know how you go.

Does this seem expensive to you?

https://www.ebay.com/itm/383223420420

amberwolf said:

Hummina Shadeeba said:

That’s why I say hypothetically. The point is to discover what the other variables would be. IF running two identical resistance and temp cells through the same simple resistance till discharged what would other variables be that world effect the time till total discharged. Assuming there are other variables. I don’t know what they would be

Click to expand...

We're trying to help you by *telling* you what the other variables are, so you don't have to waste your time and money and resources figuring it out.

You don't have to discover them, they've been known for years (decades), if you just read what others have already posted about testing batteries.

Click to expand...

I don’t know what ur *telling* me. I read no other variables in the scenario I just posted. If u have a concise link that answers my particular question that would be good to hear. If u don’t have an answer as it seems u don’t best to roll along and hopefully someone who has an answer can answer. It’s a question related to batteries obviously, and when two identical cells in temperature and resistance are put to a simple resistance what would make them possibly show different capacity

John I’ve just been using a typical lipo charger discharging and charging. It’s likely accurate enough for me but I’m wondering what the variables would be in the scenario I presented to better understand what’s what

Hummina Shadeeba said:

If u don’t have an answer as it seems u don’t best to roll along and hopefully someone who has an answer can answer.

Click to expand...

No need to be so obnoxious to people that are spending their time trying to help you.

"What if the sky were yellow?" is just as relevant a hypothetical question.

No two cells will **ever** have the same resistance.

No two cells will **ever** have the same capacity.

And the same goes for their internal physical / chemical condition internally, for which State of Health we use capacity as a proxy.

Please do spend the time money and energy to see if your theory is correct, and let us know.

i didn't post a theory. i posted a THEORETICAL question. If you where to ask what would happen if the sky were yellow....that sounds like an interesting question to me as well.

No one is "trying to help me by "telling" me what the other variables are" as no variables were presented. Am i wrong? if people are going to not answer but pretend they have answered the question and roll their eyes, i'll ask them to roll on.

Well I am coming from a practical POV, have no desire to try to detect dendrite formations in my garage with under $20 worth of gear.

So will just roll on. . .

amberwolf said:

Hummina Shadeeba said:

That’s why I say hypothetically. The point is to discover what the other variables would be. IF running two identical resistance and temp cells through the same simple resistance till discharged what would other variables be that world effect the time till total discharged. Assuming there are other variables. I don’t know what they would be

Click to expand...

We're trying to help you by *telling* you what the other variables are, so you don't have to waste your time and money and resources figuring it out.

You don't have to discover them, they've been known for years (decades), if you just read what others have already posted about testing batteries.

Click to expand...

what are they?

im no looking for dendrites. im not going to explain it again.

from my looking ...it doesnt seem there are any variables im missing but i could be wrong. battery capacity testers recording amp hours can be gotten for 10$.

And you did post a theory

that just using a static resistor to load test without even attempting to hold the load constant will give comparable results.

In fact now that I think about it, you can do just that, use wattmeters to count coulombs

but you won't be accurate enough IMO with cells under say 20Ah