Battery Tester for Linus (It's a charger, but hear me out)

[MrCupHolder]

I'm hoping an LMG staff member will see this and bring it to Linus' attention. Linus talked on WAN show about wanting to know the capacity of an old rechargeable battery. This thing will do that.
I'm not affiliated with this company. I'm just a very satisfied customer.
https://mahaenergy.com/mh-c9000/
I've been using this battery charger for more years than I can remember. It's since been replaced with a new model but it looks pretty similar with regards to functions.
New model: https://mahaenergy.com/mh-c9000pro/
This charger has a Refresh and Analyse mode that will tell you how many maH a battery will hold. It is super useful.
If you have 4x AAA batteries that are rated for 900 maH but one of the 4 only now holds 600 maH well yeah, once that one 600 maH is dead, the other 3 with 300 maH aren't going to be able to do squat.
 

[mariushm]

There are professional electronic loads that can log the capacity of a battery much better than a simple 8 bit microcontroller with a cheap uncalibrated ADC.

\

here's an example :

 

[Lurking]

You can get much better chargers with similar function that also handle all the common Li-ion batteries. At this time and age, no new fancy charger should be limited to NiMH. I've had a $50 Nitecore charger for maybe 10 years and that can do all the Li-ion batteries. So in 2022 there should be no promotion of a charger that excludes Li-ion. 

 

Yes, Li-ion flashlights have their own USB charge port. but you also can bring spare 18650 batteries etc. So there is a really good use for Li-ion for replaceable batteries. 

[ImorallySourcedElectrons]

3 hours ago, mariushm said:

There are professional electronic loads that can log the capacity of a battery much better than a simple 8 bit microcontroller with a cheap uncalibrated ADC.

\

here's an example :

 

ITech tends to have slightly better value: https://www.itech.sh/en/  They're the OEM that does/did the power supplies and loads for Keithley/Tektronix. They also seem to have SMUs now it seems, which are the obvious king if you're looking to characterize batteries.

[bakudan]

I made this account for that topic, also I have the feeling that some things are regional. The chargers with refresh, discharge and charge function are normal in Austria. Maybe not in North America? Especially the refresh function is neat, 5 times discharge and charge, after that you see on the display how much the last charge was.

[MrCupHolder]

Whilst I appreciate having started a discussion I feel some of you have missed the point. I'll admit I didn't mention the point myself.
The point is Linus mentioned on the WAN show wanting a battery charger or tester that would give him an indication of the health of a battery that didn't cost $400
If you follow the link you'll find the one I've suggested is ~$65 which is way under $400
 

3 hours ago, mariushm said:

There are professional electronic loads that can log the capacity of a battery much better than a simple 8 bit microcontroller with a cheap uncalibrated ADC.

\

here's an example :

 

So mariushm I think the multi thousand dollar piece of equipment you've suggested is not what he's looking for.

 

3 hours ago, Lurking said:

You can get much better chargers with similar function that also handle all the common Li-ion batteries. At this time and age, no new fancy charger should be limited to NiMH. I've had a $50 Nitecore charger for maybe 10 years and that can do all the Li-ion batteries. So in 2022 there should be no promotion of a charger that excludes Li-ion. 

 

Yes, Li-ion flashlights have their own USB charge port. but you also can bring spare 18650 batteries etc. So there is a really good use for Li-ion for replaceable batteries. 

Lurking - Telling me I shouldn't be promoting something doesn't come off particularly friendly. The whole point of posting here is to promote discussion and to bring something to Linus attention that I felt given the limited amount of info he provided on WAN Show he was previously unaware of.
Does your $50 charger provide a digital readout of the maH that a battery can charge to? If so perhaps you'd like to provide a link?
The only Nitecore charger I'm aware of doesn't have a digital readout with regards to the health of the battery and also has limited options with regards to discharge, recharge and break-in as compared to the Powerex that I've been using for likely more than 15 years now.
 

 

 

[mariushm]

You can discharge a ni-mh battery with a 1 cent resistor across the terminals. You don't need a 50$ charger to do that. You can discharge it manually, then put it in a slow charger for 10-12 hours, then discharge again ... and repeat 3 times and you've "reconditioned" your battery.

 

You don't need the charger to have a display to tell you the capacity. You get any regular charger and you tap into the wire going to the positive end of the battery and you measure the voltage and current simultaneously with 1 or 2 multimeters that can log and you will easily determine how much energy was pushed into the battery,

For proper benchmarks the accuracy of those readings is debatable, it's not good enough.

 

The capacity measured is kind of pointless anyway, because in reality how much capacity you actually have will vary on the load the battery will be subjected to. The battery will have a higher final capacity if discharged at 5-10mA (a clock, a co detector etc) compared to when discharged at higher currents like 100mA (a toy, a fm radio wit speakers) or 1-2A bursts (a digital camera for example)

 

[Lurking]

15 minutes ago, MrCupHolder said:

Whilst I appreciate having started a discussion I feel some of you have missed the point. I'll admit I didn't mention the point myself.
The point is Linus mentioned on the WAN show wanting a battery charger or tester that would give him an indication of the health of a battery that didn't cost $400
If you follow the link you'll find the one I've suggested is ~$65 which is way under $400
 

So mariushm I think the multi thousand dollar piece of equipment you've suggested is not what he's looking for.

 

Lurking - Telling me I shouldn't be promoting something doesn't come off particularly friendly. The whole point of posting here is to promote discussion and to bring something to Linus attention that I felt given the limited amount of info he provided on WAN Show he was previously unaware of.
Does your $50 charger provide a digital readout of the maH that a battery can charge to? If so perhaps you'd like to provide a link?
The only Nitecore charger I'm aware of doesn't have a digital readout with regards to the health of the battery and also has limited options with regards to discharge, recharge and break-in as compared to the Powerex that I've been using for likely more than 15 years now.

My charger is maybe 10 years old and likely doesn't exist anymore. But no, it doesn't have a detailed readout. It will tell if a battery is broken, though. I just wanted to demonstrate that even an old charger already had Li-ion standard. Recommending a charger that doesn't charge modern batteries seems to be bad advice. I suspect 99% of the good chargers can charge Li-ion. A charger is something you can keep for decades. You always want the best charger that treats the batteries the best. Batteries are wear items (and wear more with cheap chargers used). If just battery testing is your main concern, what is the point if it doesn't work with modern batteries? Did Linus specifically say he never wants to use Li-ion batteries and only uses Ni-MH batteries? I didn't watch the video (and won't). But for some reason I doubt that someone whose life is based on having the newest tech is not using batteries that were already common 10 years ago. 

 

I recently bought a cheap 12V car battery charger. Even that works with 12V Li-ion batteries, even when those are very very rare to almost non-existent. 

 

I don't know how the charger would determine if the remaining capacity. I assume it measures current it can push through at a specific voltage. But this still will be more an estimate than actually measured. I mean, good for you to bringing up this topic. Because if I ever buy a new charger, I may look for more detailed battery health features. 

[Lurking]

21 minutes ago, mariushm said:

You can discharge a ni-mh battery with a 1 cent resistor across the terminals. You don't need a 50$ charger to do that. You can discharge it manually, then put it in a slow charger for 10-12 hours, then discharge again ... and repeat 3 times and you've "reconditioned" your battery.

 

You don't need the charger to have a display to tell you the capacity. You get any regular charger and you tap into the wire going to the positive end of the battery and you measure the voltage and current simultaneously with 1 or 2 multimeters that can log and you will easily determine how much energy was pushed into the battery,

For proper benchmarks the accuracy of those readings is debatable, it's not good enough.

 

The capacity measured is kind of pointless anyway, because in reality how much capacity you actually have will vary on the load the battery will be subjected to. The battery will have a higher final capacity if discharged at 5-10mA (a clock, a co detector etc) compared to when discharged at higher currents like 100mA (a toy, a fm radio wit speakers) or 1-2A bursts (a digital camera for example)

 

Good points, but a built-in feature still would be more convenient. Battery capacity is a bit voodoo anyway. Because as you discharge, you reduce voltage. But every device needs a minimum voltage to function and a current still available at that voltage. So assuming 1.4V per cell this battery would give more energy to a device that requires 1.2V and 1A. but if you have a device that requires 1.3V and 2A minimum, that battery will appear "empty" sooner. I just made those numbers up to demonstrate, but hopefully shows how useless the measuring really is. Like the batteries that "died" in my toothbrush still can operate my remote. If you have an EV, the device using is very well known and they can determine the usable capacity much better (and much more safety is required).

 

My DIY way to find old batteries is to put them in my toothbrushes or some other higher wattage user. If they last 3 weeks before charging, they still are good. If they only last a week, they are not. 

[ImorallySourcedElectrons]

Well, if you want to make an actual simple device to characterize batteries, as he suggested, you could go all the way and implement something along the lines of this:

 

Should do the push-pull stage slightly different to get better low current performance/linearity (important for coin cells), and definitely don't use that opamp. But it's kind of simple to make a low current source-sink circuit that can do constant voltage or current and keep track of the other unit.

[Lurking]

6 hours ago, ImorallySourcedElectrons said:

Well, if you want to make an actual simple device to characterize batteries, as he suggested, you could go all the way and implement something along the lines of this:

 

Should do the push-pull stage slightly different to get better low current performance/linearity (important for coin cells), and definitely don't use that opamp. But it's kind of simple to make a low current source-sink circuit that can do constant voltage or current and keep track of the other unit.

You can measure how much energy you push into the battery. But you don't know how much of that energy is stored, and how much is turned into heat. So if you put in 100Wh, the battery capacity may only be 60 Wh if 40Wh turned into heat. 

 

I suspect as batteries age, they become less efficient and produce more heat. So you can't just use a standard efficiency. But this is just an assumption. I'm sure battery chemistry is much more complex. 

[ImorallySourcedElectrons]

1 hour ago, Lurking said:

You can measure how much energy you push into the battery. But you don't know how much of that energy is stored, and how much is turned into heat. So if you put in 100Wh, the battery capacity may only be 60 Wh if 40Wh turned into heat. 

 

I suspect as batteries age, they become less efficient and produce more heat. So you can't just use a standard efficiency. But this is just an assumption. I'm sure battery chemistry is much more complex. 

Yeah, that's why you need a push-pull stage, so you can drain it as well. You need to run it through multiple full cycles, preferably with different profiles, to get a good idea of the battery's performance. Ideally you'd use a full-fledged source-measure unit and imitate common load and charging profiles, but this circuit would be a poor man's version of that.

 

Temperature measurement is usually not that effective to determine efficiency, especially if there's an internal battery management chip, since a lot of power will be lost in that one in some instances.

[Lurking]

31 minutes ago, ImorallySourcedElectrons said:

Yeah, that's why you need a push-pull stage, so you can drain it as well. You need to run it through multiple full cycles, preferably with different profiles, to get a good idea of the battery's performance. Ideally you'd use a full-fledged source-measure unit and imitate common load and charging profiles, but this circuit would be a poor man's version of that.

 

Temperature measurement is usually not that effective to determine efficiency, especially if there's an internal battery management chip, since a lot of power will be lost in that one in some instances.

Thanks for explaining. That makes sense

To get an accurate capacity number it seems to be a lot of effort in time and also wear of the battery if it takes a few charges and discharges to measure the capacity. it seems a device that just "knows" the capacity by a quick measurement wouldn't be accurate. 

[ImorallySourcedElectrons]

1 minute ago, Lurking said:

Thanks for explaining. That makes sense

To get an accurate capacity number it seems to be a lot of effort in time and also wear of the battery if it takes a few charges and discharges to measure the capacity. it seems a device that just "knows" the capacity by a quick measurement wouldn't be accurate. 

You could monitor the battery while it's being ran inside a device, but that's not always practical, though it's relatively easy to cobble something together for it that will fit in a small space.

I made some flex PCBs for this purpose, but sadly can't post pictures without self-doxing. Basically I had this sort of setup going:

The reason to use a current shunt instead of a regular current meter was to get a smaller series resistance for small currents. But you need a fairly good volt meter to measure those ultra-low voltages. You could make a small flex PCB with the necessary circuits to integrate everything nicely though.

[Kilrah]

On 12/17/2022 at 4:26 PM, mariushm said:

There are professional electronic loads that can log the capacity of a battery much better than a simple 8 bit microcontroller with a cheap uncalibrated ADC.

  

On 12/17/2022 at 8:45 PM, mariushm said:

You don't need the charger to have a display to tell you the capacity. You get any regular charger and you tap into the wire going to the positive end of the battery and you measure the voltage and current simultaneously with 1 or 2 multimeters that can log and you will easily determine how much energy was pushed into the battery,

You're somehow mentioning both ends of "extremely accurate" and "super basic and impractical" while completely missing the point that was right in the middle with "a convenient device you can just plop your battery in and get a sufficiently good idea".

 

For that there are already things in the $100 range:

 

If you're only after AA/AAA

https://www.aliexpress.com/item/1005004559965233.html

 

If you want basically every round cell that exists including 18650s, Li-Fe, NiZn,...

https://www.aliexpress.com/item/1005002228184485.html

 

Have had the latter for several years.

 

[Heliian]

On 12/17/2022 at 3:05 PM, Lurking said:

don't know how the charger would determine if the remaining capacity

It's simple, a battery rated for 600mah is at 100% capacity when it maintains 600ma for 1 hour. 

Our fancy load testers have an adjustable current setting and it times the drain, stopping at the minimum operating  voltage of the battery.

 

 

[Lurking]

1 hour ago, Heliian said:

It's simple, a battery rated for 600mah is at 100% capacity when it maintains 600ma for 1 hour. 

Our fancy load testers have an adjustable current setting and it times the drain, stopping at the minimum operating  voltage of the battery.

 

 

I meant to measure without draining it. What you propose will end up with an empty battery. That is highly impractical and will wear out the battery by needlessly adding charge and recharge cycles.

[Kilrah]

11 hours ago, Lurking said:

I meant to measure without draining it.

You can estimate, which pretty much every BMS in common devices does, but to be sure and actually measure there's no choice but to do an actual cycle.

 

11 hours ago, Lurking said:

That is highly impractical and will wear out the battery by needlessly adding charge and recharge cycles.

You don't need to do that frequently, wear's negligible. If you want to ensure capacity, say once every 6 months you do couple of cycles, that's it.

 

For consumer use it doesn't matter. For sensitive use what's important is to know the condition, who cares about wear. 

[Heliian]

13 hours ago, Lurking said:

will end up with an empty battery

No, the battery isn't empty, its discharged to a minimum operating voltage.    And like @Kilrah said, once every 6 months or a year, only one cycle is necessary to determine if the battery can meet its rated capacity.

 

 

 

 

[ImorallySourcedElectrons]

4 hours ago, Kilrah said:

You can estimate, which pretty much every BMS in common devices does, but to be sure and actually measure there's no choice but to do an actual cycle.

Indeed, and that requires you to have an aging model of the battery if you wish to have any chance at accuracy. Then you can usually derive it from battery parameters that are relatively easy to measure (impedance during charging/discharging, voltage drop as function of mAh discharged, ...)