IR threshold between cells that should not be paired anymore?

klrman

Newly Enlightened
Joined
Feb 25, 2009
Messages
181
Is there a limit between a batteries internal resistance compared to another that would warrant them not being paired anymore in parallel or series in a flashlight?

I have a neat little tester, the SM8124A, that shows the IR difference between some batteries I purchased in groups of two or four. With one pair of Protected 26650's the IR for the first battery was 30 and for the second it was 37 . Seems close enough but is it still ok to use them together in a flashlight since it is still a difference of more than 20%?
 
Last edited:

terjee

Enlightened
Joined
Jul 24, 2016
Messages
729
Location
Bergen, Norway
Being protected really helps you out here, so I definitively wouldn't worry too much. If you're worried, just doublecheck that the protection includes both overcurrent and undervoltage.

A little IR difference itself isn't what'd ruin a day, but rather things like overdischarging one of the cells, and the protection helps you out with that.

Still good to keep an eye on IR, it you see one at 30 and the other at 200, it's probably time to think about replacements.

If the cells are in parallel, it matters a bit more, but overcurrent protection should still have your back.
 

klrman

Newly Enlightened
Joined
Feb 25, 2009
Messages
181
Helpful info, thanks. All the batteries have overcurrent and undervoltage protection. Just a little surprised they came in pairs with one having much more resistance than the other battery as I thought coming pairs mean they should have been evenly matched.

BTW, nice to have this forum have no lag anymore like the past. Pages loads instantly now.
 

peter yetman

Flashlight Enthusiast
Joined
Mar 23, 2014
Messages
5,100
Location
North Norfolk UK
I was also worried about IR for pairs of cells in series a while back, until I realised you need to look at the capacity more. THe IR is a good indication of the cell's age, but capacity is much more important if you are matching cells.
P
 

klrman

Newly Enlightened
Joined
Feb 25, 2009
Messages
181
Good to know, thanks Peter. Never tested these two cells for capacity so that will be my project today.
 

terjee

Enlightened
Joined
Jul 24, 2016
Messages
729
Location
Bergen, Norway
Helpful info, thanks. All the batteries have overcurrent and undervoltage protection. Just a little surprised they came in pairs with one having much more resistance than the other battery as I thought coming pairs mean they should have been evenly matched.

It's a lot when you view it as percent, but not really if you view it as an absolute number. Also, there's a certain margin of error and noise here. A few microohms can be the result of something unrelated as well, such as how well each of the cells connected electrically to your tester that day, or even minute temperature variations. Not saying any of those were a factor here, just trying to put a scale to how small the difference is when not seen as percentage.
 

klrman

Newly Enlightened
Joined
Feb 25, 2009
Messages
181
Can't find the thread, but it was HKJ that put me right.
If I come across it I'll let you know.
P

That's ok as I remember reading one of HKJ's tests and he said +-5% of total capacity is the tolerance for using in series.
 

klrman

Newly Enlightened
Joined
Feb 25, 2009
Messages
181
It's a lot when you view it as percent, but not really if you view it as an absolute number. Also, there's a certain margin of error and noise here. A few microohms can be the result of something unrelated as well, such as how well each of the cells connected electrically to your tester that day, or even minute temperature variations. Not saying any of those were a factor here, just trying to put a scale to how small the difference is when not seen as percentage.

Now that you mentioned it, I tested the ir a few more times with my SM8124A tester and results were still the same. This 4 wire tester seems very consistent, but I see what you mean in the overall sense that the difference is small. Still glad I started this thread as I was not 100% sure and thought it couldn't hurt to ask just in case. Thanks for help.
 

Gauss163

Flashlight Enthusiast
Joined
Oct 20, 2013
Messages
1,604
Location
USA
Just as I mentioned in the other thread, AC (1kHz) impedance is not the correct test to be using here, since there is not necessarily any correlation between the AC impedance and low frequency DC impedance (e.g. they might be close on AC impedance but much further off on low-frequency DC - which is what matters for most applications here).
 

klrman

Newly Enlightened
Joined
Feb 25, 2009
Messages
181
Just as I mentioned in the other thread, AC (1kHz) impedance is not the correct test to be using here, since there is not necessarily any correlation between the AC impedance and low frequency DC impedance (e.g. they might be close on AC impedance but much further off on low-frequency DC - which is what matters for most applications here).

After reading the thread a little, it seems there is for and against for both types of measurements but that the AC(1kHz) is still a decent way to test.
 
Last edited:

Gauss163

Flashlight Enthusiast
Joined
Oct 20, 2013
Messages
1,604
Location
USA
After reading the thread a little, it seems there is for and against for both types of measurements but that the AC(1kHz) is still a decent way to test.

It seems that you did not read closely enough. Why do you believe that "AC(1kHz) is still a decent way to test"?
 

klrman

Newly Enlightened
Joined
Feb 25, 2009
Messages
181
It seems that you did not read closely enough. Why do you believe that "AC(1kHz) is still a decent way to test"?

Sorry about that. I thought when you did not reply to Jasonck's query that maybe it was a stalemate as to which is the better or more reliable way of testing. Give me time and I will go through it all link per link to understand better!
 

Gauss163

Flashlight Enthusiast
Joined
Oct 20, 2013
Messages
1,604
Location
USA
Sorry about that. I thought when you did not reply to Jasonck's query that maybe it was a stalemate as to which is the better or more reliable way of testing. Give me time and I will go through it all link per link to understand better!

I'm not sure what you refer to. In any case, to be much more explicit I quote below from a paper published this year in Nature's Scientific Reports on precisely this subject (my emphasis)

Anup Barai et al. said:
The (AC) 1 kHz resistance test, although useful for a quick check in a manufacturing environment e.g. quality check, not much value is derived from it. Furthermore, whether the measured 1 kHz resistance value lies in the inductive or conductive region is highly dependent on the battery sample, therefore the single frequency (e.g. 1 kHz) needs to be chosen depending on the sample when employed for quality check. [...]

Measuring battery resistance with a 1 kHz AC signal (or similar single frequency signal), is common practice in industry, especially for measuring lead-acid battery resistance. It is a relatively fast (in the order of seconds), low power consuming and low cost technique, using handheld equipment. Usually a low current sinusoidal signal of 1 kHz is applied to the battery and the voltage response is measured. Although this technique is time-efficient, a single value of resistance is not sufficient to characterise the battery's performance. This is because charge transfer through multilayer surface films and kinetic and diffusional processes in the solid and liquid phases of the battery lead to a frequency dependent resistance

If you peruse that paper (or related papers) you will learn that such AC impedance tests only measure one part of the impedance (mostly Ohmic). In particular they omit the very important charge transfer resistance. This is the component that typically grows most as the cell ages so it is what you want to be tracking for battery health. Since AC impedance tests completely miss that, they are of little value for tracking battery health.
 
Last edited:

klrman

Newly Enlightened
Joined
Feb 25, 2009
Messages
181
Interesting info Gauss, thanks very much for the help. Is there any reasonably priced meter that could accurately test for aging cell health then that you think would be good enough to do the job?
 

Gauss163

Flashlight Enthusiast
Joined
Oct 20, 2013
Messages
1,604
Location
USA
Interesting info Gauss, thanks very much for the help. Is there any reasonably priced meter that could accurately test for aging cell health then that you think would be good enough to do the job?

Any charger that competently implements an IR test using a DC pulsed load method should serve well for most hobbyist applications. This includes some popular analyzing chargers, e.g. Opus BT-C3100, SkyRC MC3000, etc. Check reviews to see if the charger yields repeatable accurate results.
 

klrman

Newly Enlightened
Joined
Feb 25, 2009
Messages
181
Any charger that competently implements an IR test using a DC pulsed load method should serve well for most hobbyist applications. This includes some popular analyzing chargers, e.g. Opus BT-C3100, SkyRC MC3000, etc. Check reviews to see if the charger yields repeatable accurate results.

Thanks. I have the latest MC3000 but got frustrated with the changing IR values every time I rotated a cell or switched bays, but I did read yesterday that HKJ mentioned to generally take the lowest reading as the most accurate, so I will go with that more or less and use my SM8124A as a tool to detect if I receive any fake cells from sellers.
 
Last edited:

Gauss163

Flashlight Enthusiast
Joined
Oct 20, 2013
Messages
1,604
Location
USA
Thanks. I have the latest MC3000 but got frustrated with the changing IR values every time I rotated a cell or switched bays, but I did read yesterday that HKJ mentioned to generally take the lowest reading as the most accurate, so I will go with that more or less and use my SM8124A as a tool to detect if I receive any fake cells from sellers.

What sort of variations are you seeing? Iirc reviews of prior versions reported highly consistent IR values. If yours are not then maybe your unit is faulty, or maybe QC has gone downhill in later versions.
 

klrman

Newly Enlightened
Joined
Feb 25, 2009
Messages
181
What sort of variations are you seeing? Iirc reviews of prior versions reported highly consistent IR values. If yours are not then maybe your unit is faulty, or maybe QC has gone downhill in later versions.

Just by rotating a protected 18650 cell, numbers can jump from 48 to 130 in the same slot as well as the same type of variations between slots with the same cells. I have noticed that my protected 26650 readings in the MC3000 seem to be much more stable but never figured out exactly why.

On another note, for the protected 26650 batteries I use in series in one of my flashlights, the MC3000 reads them both at 48 for ir and the SM8124B reads them both at 31ir. After you guys helped me out to understand that both are using different methods of measuring resistance, it makes much more sense now.
 

Gauss163

Flashlight Enthusiast
Joined
Oct 20, 2013
Messages
1,604
Location
USA
Just by rotating a protected 18650 cell, numbers can jump from 48 to 130 in the same slot as well as the same type of variations between slots with the same cells. I have noticed that my protected 26650 readings in the MC3000 seem to be much more stable but never figured out exactly why [...]

That's not normal. Typically you should see only a few milliohms variation whether rotated in-slot or between slots. The MC3000 should serve better for tracking health because - unlike the AC 1kHz test - it will incorporate some of the resistance components with longer time constants (but not all of them since the test duration is too short).

It would be interesting to see how the AC test compares to the MC3K pulse test on a wide variety of cells. For example, what's the largest difference between the two that you've encountered?
 
Last edited:
Top