Testing battery capacity

[OM222O]

Hello
I would like to test capacity of different batteries, so I got myself a huge collection of different brands, sizes, chemistry, etc (about 100 batteries all together) and made a data logger + a constant current load as my setup.

I know batteries will have "different capacity" based on the current draw. I log the data both in mAh and mWh.
I was wondering what causes this difference and couldn't think of anything except the internal resistance which consumes (I^2)*R power, so the more current draw, the less the usable capacity    I would like to run my test at about 100mA as it would take waaaaayy too long to drain all those batteries at something like 5 to 10mA ... weeks perhaps! but if my assumption is correct and that is the only reason for different capacities, I can measure the internal resistance, then drain the battery with the desired current, add the power consumed by the internal resistance, go back and using some algebra get the capacity at other current draw values    do you think my theory is correct? if not what other factors affect this difference? is there another method which I can use to calculate the capacity while maintaining the higher current draw?

[Conrad Hoffman]

AFAIK, it's a chemical thing and you have to run the test at the draw that you want to test.

[OM222O]

It can't be a purely chemical process ... the rate of those reactions only depends on temperature (and pressure which is definitely constant in this case) ... so if the battery isn't drained at say, 1A or something crazy like that and it doesn't end up heating up, the chemical process should be the same regardless of the load 

[jbb]

I’m no expert but I do know 3 things:
1) a lot of a battery’s ‘resistance’ is due to chemical behaviour (eg ion concentration etc.)
2) chemical activity changes very fast with temperature
3) ion diffusion rates can be quite important in sustained (dis)charge, so constant discharge can be limited by the rate at which ‘fresh’ ions diffuse into the active region.

This means that some cells have a current limit for thermal reasons - as we would expect - and others have a current limit for ion diffusion reasons.

If you discharge a coin cell too aggressively you may find it has much lower capacity than expected.

If you discharge other cells (eg LiIon, LiPo, LiSOCl) too fast they may overheat, burst, catch fire or even explode.

Sadly it sometimes just takes ages to test a cell.

[orbanp]

I have one of these:
https://www.ebay.ca/itm/18650-Li-ion-Lithium-Lead-acid-Battery-Capacity-Meter-Discharge-Tester-Analyzer/272563711537?hash=item3f76102e31:g:N1QAAOSw5IJWg~~~:rk:2:pf:0

Read the specs if it would do it for you.
Obviously, it is for rechargeable batteries.

Regards, Peter

[schmitt trigger]

Many E-loads include a function which accumulates the total amp-hours drawn from a battery.
You can also select the cutoff voltage.

[OM222O]

that's not my issue. I already have that setup

shameless self promotion (please watch the video) :



my main concern is the accuracy of the received capacity as there are different values for different discharge rates.
I'm trying to validate my testing methodology and correct for errors if I can.

[schmitt trigger]

Of course there are.
That is the reason battery manufacturers publish discharge curves measured at different rates.
As others have mentioned, there are many subtleties in an electrochemical reaction, which are affected by many things...among them temperature.
A heavily loaded battery self-heats significantly.

Thus, an accepted rate is C/10. I would start with that value. But since you have an automated setup, you could, in a straightforward manner, test them at other loading values