Testing the 18650 battery

[CharlotteSwiss]

Hi, I'd like to test some 18650 batteries, which have the right voltage (3.7V), but I'd like to know if they're still holding a charge efficiently, or if I should replace them.
I saw a well-known Chinese store selling special testers, but I wanted to test these batteries myself. Any advice?
Thanks 

[petemate]

You need an electronic load. They are pretty simple and with all the Chinese manufacturers, there are a lot of options available.

The usual procedure is to use the constant-current mode of the electronic load and run it with an undervoltage turn-off limit. So your load will pull a constant current until the battery drops below a certain voltage. By measuring the time it runs, you can get the charge of the battery(because time * current = charge). Many electronics loads have built-in battery test functions, which is basically doing what I just described.

Its pretty easy to build an electronic load yourself: Its a FET, an opamp and a current sense resistor, plus some glue. Its a good project.

[thm_w]

Something like this will do the job well if you are specifically testing 18650's or AA's etc: https://lygte-info.dk/review/Review%20Charger%20LiitoKala%20Lii-500%20UK.html https://www.liito-kala.com/page92?product_id=6

Plenty of similar equivalent products around if you can't find that one.

If you wanted to do very basic testing yourself you can apply a load resistor and log or manually write down the voltage and current over time.

Then as mentioned above a constant current circuit would be a more advanced option.

[J-R]

I've had version 2.2 of the Opus BT-C3100 for many years now and have been very happy with it.  Being able to test the capacity of four cells at once is a great time saver versus the other methods, especially since you shouldn't ever discharge/charge Lithium-ion unattended.

Reviews:
https://lightsngear.com/reviewing-the-opus-bt-c3100-v2-2-battery-charger/
https://budgetlightforum.com/t/review-opus-bt-c3100-v2-2-nimh-nicd-li-ion-battery-charger/48488

[radiolistener]

You can test them yourself with a simple constant-load discharge setup.

Use a power resistor with a suitable value to draw the nominal current for that type of 18650 cell. Connect the resistor directly to the battery, then record the battery voltage over time using a voltmeter and a timer. Take measurements more frequently at the beginning and near the end of the discharge, where the voltage changes faster.

Once you reach the cutoff voltage, take your table of voltage vs time and integrate the delivered power in Octave. That will give you the effective energy in watt-hours, which is the practical capacity of the cell.

If the cells have been stored for a long time, the first 1–2 charge/discharge cycles may give a lower reading. It is better to measure capacity after two full cycles.

You can also use an electronic load set to constant-current mode. This simplifies the measurement: instead of tracking voltage over time, you only need to measure the discharge duration to get Amp-hours. Many electronic loads can even integrate the delivered energy automatically and report the result in watt-hours.

[pcprogrammer]

Whatever method you choose, do make sure to not let the battery drop below the recommended cut off voltage. A commonly used voltage here is 2.7V, although I have seen mentions as low as 2.4V. Going below that brings harm to your battery and can easily lead to a dead battery.

Also monitor the temperature of the battery under discharge (or charge) to not become more than 100 degrees Celsius.

Another point is to do not discharge with a current greater than the battery is rated for. For instance a 2Ah battery with a 10C rating should not be discharged with a current higher than 20A.

[BeBuLamar]

The capacity of the battery varies with the discharge current. Generally higher discharge current result in less capacity in the test. So for the 18650 how much discharge current should be tested at?

[wraper]

Something like this will do the job well if you are specifically testing 18650's or AA's etc: https://lygte-info.dk/review/Review%20Charger%20LiitoKala%20Lii-500%20UK.html https://www.liito-kala.com/page92?product_id=6

Plenty of similar equivalent products around if you can't find that one.

If you wanted to do very basic testing yourself you can apply a load resistor and log or manually write down the voltage and current over time.

Then as mentioned above a constant current circuit would be a more advanced option.

It can only test capacity, and won't discharge to 2.5V as specified in datasheets. Internal resistance measurement is total BS in the range of good li-ion cells. It can literally show higher ESR for high discharge current cells compared to regular cells.

[tunk]

As mentioned, there's (at least) three things to consider:
- capacity
- internal resistance
- rate of self-discharge
If you want to test self-discharge; store the cells somewhere
dry and at constant temperature, and measure the voltage over
some weeks or months. Preferably you should use at least a
6000 count multimeter (i.e. to see mV changes).
I have used one of these to test capacity (not the very best
accuracy): https://www.ebay.com/itm/201633345906

[CaptDon]

The constant current method is best but you must insure discharge cutoff at 2.4 to 2.5 volts. If you are lucky enough to be able to log the voltage during the discharge you can generate a file like a .csv knowing that the current was constant and perform a math function on the .csv and obtain very accurate results. It would be good to run several cycles of full charge / discharge to cutoff and look at the overall trend. A failing battery will show a declining trend. A battery worth keeping will show similar runs or even slightly increasing capacity with each run.

[BILLPOD]

Good Morning CharlotteSwiss, I have one of these:   https://www.amazon.com/Multi-Chemistry-Charger-Universal-Rechargeable-Battery/dp/B0FNR1PYKH/ref=sr_1_14_sspa?  It does everything you want with cylindrical cells.   It is called 'SkyRC MC3000.  It's a bit pricey, but it is well worth it if you have a multitude of rechargeable batteries, (cells).

[Fryguy]

The capacity of the battery varies with the discharge current. Generally higher discharge current result in less capacity in the test. So for the 18650 how much discharge current should be tested at?

A common standard method is to discharge a cell @ C/10 to determine the nominal capacity . The charging current should not be too high either . C/10 or C/5 is a good choice and the cells won't get too hot . This method can be applied to all cell types .

If your cell is rated e.g. 2500 mAh - then charge it with 250-500mA and discharge it with 250mA .

[wraper]

The capacity of the battery varies with the discharge current. Generally higher discharge current result in less capacity in the test. So for the 18650 how much discharge current should be tested at?

A common standard method is to discharge a cell @ C/10 to determine the nominal capacity . The charging current should not be too high either . C/10 or C/5 is a good choice and the cells won't get too hot .

Most commonly it's 0.2C for rated capacity. But it's not that useful other than for detecting really bad cells. You need more like 2-10C discharge current to test them well depending on cell type.

[Fryguy]

The capacity of the battery varies with the discharge current. Generally higher discharge current result in less capacity in the test. So for the 18650 how much discharge current should be tested at?

A common standard method is to discharge a cell @ C/10 to determine the nominal capacity . The charging current should not be too high either . C/10 or C/5 is a good choice and the cells won't get too hot .

Most commonly it's 0.2C for rated capacity. But it's not that useful other than for detecting really bad cells. You need more like 2-10C discharge current to test them well depending on cell type.

What kind of result are you looking for with a discharge current of 2-10C ? Cheap 18650 cells are not even rated for more than 2C discharge current . You will never get a decent capacity reading with that much current and even the really expensive cells that can handle up to 30C discharge current are measured at C/5 or C/10 for rated capacity . 

[wraper]

That its capacity does not drop that much under load. Especially important when used for power tools or other high drain devices. For example my leaf blower can drain absolutely good 40V 4Ah battery flat in about 7 minutes with max airflow setting.


And it's not some fancy tabless cell. Just average high drain one.

[Fryguy]

Yes , that is a very nice powertool cell that can take that kind of beating - discharge up to 30A = 12C . . .
Look at the first item under chapter 3 - specifications . Nominal discharge capacity 2500mAh @0,5A (C/5) / 2,5VDC cut off .
If you take used cells from e.g. defective powerbanks or laptop batteries , you get the cheap ones which will not take that much load . . .

[CharlotteSwiss]

Thanks everyone for the advice. I thought it would be easier to check if the 18650 batteries are still working, or better to replace them.
I'll add some information about why I asked this question here on the forum: my dad has a cordless multitool (it's a Parkside, let's say a Dremel clone). The tool is about 7 years old. He doesn't use it frequently, but when he does, he recharges it. However, when he needs it (maybe after a month or two), every time he starts it, it doesn't turn on due to low battery. In short, the batteries don't hold a charge for long, even when the tool isn't used.
So I took it apart and had the idea of ​​checking if this battery pack (3 x 18650) is still working. Sure, I could take three batteries, unsolder the current ones, and replace them. But for educational purposes, I wanted to understand when it would be time to throw the battery away. I'm adding a photo of the device. According to the specifications, the operating voltage is 12V, and they state 1300mAh.
Thanks again.

[wraper]

Yes , that is a very nice powertool cell that can take that kind of beating - discharge up to 30A = 12C . . .
Look at the first item under chapter 3 - specifications . Nominal discharge capacity 2500mAh @0,5A (C/5) / 2,5VDC cut off .
If you take used cells from e.g. defective powerbanks or laptop batteries , you get the cheap ones which will not take that much load . . .

It's a quite cheap cell actually. Just a tiny bit more expensive than "laptop cells". Also that's why I wrote:

You need more like 2-10C discharge current to test them well depending on cell type.

The problem is that almost totally worn out cell may still show a decent capacity when tested at 0.2C

[wraper]

But for educational purposes, I wanted to understand when it would be time to throw the battery away. I'm adding a photo of the device. According to the specifications, the operating voltage is 12V, and they state 1300mAh.

IMHO do not bother if you still have a decent run time. You could replace them with 2.5-3Ah high drain cells if you want longer tool run time.
EDIT: you can easily measure if cells are reasonably balanced (individual cell voltage matches between the cells).

[pcprogrammer]

Thanks everyone for the advice. I thought it would be easier to check if the 18650 batteries are still working, or better to replace them.
I'll add some information about why I asked this question here on the forum: my dad has a cordless multitool (it's a Parkside, let's say a Dremel clone). The tool is about 7 years old. He doesn't use it frequently, but when he does, he recharges it. However, when he needs it (maybe after a month or two), every time he starts it, it doesn't turn on due to low battery. In short, the batteries don't hold a charge for long, even when the tool isn't used.
So I took it apart and had the idea of ​​checking if this battery pack (3 x 18650) is still working. Sure, I could take three batteries, unsolder the current ones, and replace them. But for educational purposes, I wanted to understand when it would be time to throw the battery away. I'm adding a photo of the device. According to the specifications, the operating voltage is 12V, and they state 1300mAh.
Thanks again.

Even older batteries should not loose the charge that quick when fully unloaded, so it makes me wonder if there is something in the tool that slowly discharges the batteries when switched off.

Be aware that soldering 18650 batteries directly on the battery terminals is not advisable. There is a risk of injecting too much heat when doing so and damage the battery. When you look for new batteries make sure to get some that already have spot welded tabs on them, to which it is possible to solder.

The lifetime of a 18650 battery is based on several aspects, of which old age is one of the least likely causes. They do degrade when left fully charged or almost discharged for a long time, but this can be avoided by dropping or raising the charge to between 40 and 80% before putting them in storage.

What degrades them the most is full charge discharge cycles. They can be rated anywhere from 500 to 1500 cycles before the need of replacing them.

[Fryguy]

You need more like 2-10C discharge current to test them well depending on cell type.
The problem is that almost totally worn out cell may still show a decent capacity when tested at 0.2C

That is indeed possible , if the internal resistance of the cell is much too high . A good charger can measure the cell's internal resistance and some of them are not that expensive .

And yes , it sounds like something in this tool eats the batteries . . . 

7 years and 1300mAh ? Maybe they are just dead and should be replaced anyway . 

[CharlotteSwiss]

IMHO do not bother if you still have a decent run time. You could replace them with 2.5-3Ah high drain cells if you want longer tool run time.
EDIT: you can easily measure if cells are reasonably balanced (individual cell voltage matches between the cells).

The problem is that my dad charges them, but then when he needs them, the tool won't work and requires a battery charger. He uses it for small jobs away from his bench, but otherwise he has a Dremel with a power cord. Replacing them with batteries larger than 1300mAh seems reasonable. I can't check the individual voltages right now; they're connected by cables. But as soon as I can, I'll unsolder them and check.

Even older batteries should not loose the charge that quick when fully unloaded, so it makes me wonder if there is something in the tool that slowly discharges the batteries when switched off.

Be aware that soldering 18650 batteries directly on the battery terminals is not advisable. There is a risk of injecting too much heat when doing so and damage the battery. When you look for new batteries make sure to get some that already have spot welded tabs on them, to which it is possible to solder.

The lifetime of a 18650 battery is based on several aspects, of which old age is one of the least likely causes. They do degrade when left fully charged or almost discharged for a long time, but this can be avoided by dropping or raising the charge to between 40 and 80% before putting them in storage.

What degrades them the most is full charge discharge cycles. They can be rated anywhere from 500 to 1500 cycles before the need of replacing them.

It may not be obvious from the photo, but the three batteries have the classic soldering lugs and are connected by red cables.
As for battery life, the fact that my father didn't use this Parkside much probably affected their lifespan.

--------------------

I also thought that there's power consumption even when it's off. Now I'm trying to figure out how to check. Thanks.

[wraper]

The problem is that my dad charges them, but then when he needs them, the tool won't work and requires a battery charger.

Then certainly cells are bad as it has a mechanical switch. They have a very high self-discharge as minimum.

I can't check the individual voltages right now; they're connected by cables.

So what? You still can measure them by probing terminals of individual cells with multimeter.

[wraper]

You need more like 2-10C discharge current to test them well depending on cell type.
The problem is that almost totally worn out cell may still show a decent capacity when tested at 0.2C

That is indeed possible , if the internal resistance of the cell is much too high . A good charger can measure the cell's internal resistance and some of them are not that expensive

There in no inexpensive charger that can measure ESR properly, it requires 4 wire measurement or extremely good 2 wire contact to get proper readings in < 50 mOhm range. Also you'd be surprised but some cells still may have a decent internal resistance when measured at 1kHz as specified in datasheets but somehow fall on their face when loaded with high current.

[Fryguy]

You need more like 2-10C discharge current to test them well depending on cell type.
The problem is that almost totally worn out cell may still show a decent capacity when tested at 0.2C

That is indeed possible , if the internal resistance of the cell is much too high . A good charger can measure the cell's internal resistance and some of them are not that expensive

There in no inexpensive charger that can measure ESR properly, it requires 4 wire measurement or extremely good 2 wire contact to get proper readings in < 50 mOhm range. Also you'd be surprised but some cells still may have a decent internal resistance when measured at 1kHz as specified in datasheets but somehow fall on their face when loaded with high current.

Interresting - did that occur ? I'd honestly like to know how that works .
A decent charger costs below 100 bucks - and a cheap but real 4-wire esr-meter is available around 50 -70 bucks - yes , not nearly professional level , yet good enough for the job and quite affordable for hobbyists . Or is that still too expensive ?