To evaluate performance of used 18650 cells...

This is the place for any magazine-related discussions that don't fit in any of the column discussion boards below.
User avatar
MrAl
Posts: 3856
Joined: Fri Jan 11, 2002 1:01 am
Location: NewJersey
Contact:

Re: To evaluate performance of used 18650 cells...

Post by MrAl » Tue Jun 19, 2018 4:06 pm

haklesup wrote:
Mon Jun 18, 2018 3:30 pm
I noticed that Max discharge current is the other primary specification used for selection. Naturally you would want a high discharge current for say a super flashlight but lower discharge and a little more capacity for something like a low power toy. How do you suppose one would measure that spec. I guess you could turn up the current on the load until you had too much voltage drop. If you were selecting for an actual application then you could test at the actual expected load current. Have you ever repeated a discharge test several times in a row using different load currents and plotted on the same graph. I wonder what that would tell you. Is capacity a constant over a wide operating range?
Hello there,

The criterion for max discharge is usually based on max cell temperature. I think it was 80 deg C in most cases but some may go higher. With any load if the cell reaches that temperature before it is discharged or before the temperature starts to drop (because it is running out of steam) then the current would be considered too high.

The load current affects the apparent cell capacity quite a bit. A cell rated for 2000mAHr at C/20 may only be 1500mAHr at C/5 for example. There is a scientific measurement for this behavior called the Peukert Law or simply "P Factor" that describes the decrease in apparent capacity as load current increases, but strangely i dont see manufactuers using this for any spec. What i do see sometimes is a few different capacities at different load currents, or with different load resistances.
Radio Shack published a good book on this back in the 1990's but it only covers cells like Carbon Zinc and NiCd and Alkalines, and i dont think they even sell those particular cells anymore :-)
LEDs vs Bulbs, LEDs are winning.

User avatar
evahle
Posts: 186
Joined: Mon Aug 13, 2007 3:48 am
Contact:

Re: To evaluate performance of used 18650 cells...

Post by evahle » Tue Jun 19, 2018 7:55 pm

That would be fun to build too, I think.

User avatar
MrAl
Posts: 3856
Joined: Fri Jan 11, 2002 1:01 am
Location: NewJersey
Contact:

Re: To evaluate performance of used 18650 cells...

Post by MrAl » Wed Jun 20, 2018 5:04 am

evahle wrote:
Tue Jun 19, 2018 7:55 pm
That would be fun to build too, I think.
Hi,

What would be fun to build?
LEDs vs Bulbs, LEDs are winning.

User avatar
evahle
Posts: 186
Joined: Mon Aug 13, 2007 3:48 am
Contact:

Re: To evaluate performance of used 18650 cells...

Post by evahle » Wed Jun 20, 2018 11:19 am

It was from your earlier reply MrAl:
Hi there,

Yes if you use a constant current source then you have to log the time to 0.9v or whatever cutoff you choose. The OPUS i have goes down to 0.9v for NiMH and Alkalines.

That's not too hard to do really though, because if you watch it maybe very 30 minutes you will see it start to decline faster so you know it is getting close to the end. But you can also hook up a simple comparator circuit to light up an LED or set off a buzzer when one is done. The LM339 would be good for this.

User avatar
haklesup
Posts: 2850
Joined: Thu Aug 01, 2002 1:01 am
Location: San Jose CA
Contact:

Re: To evaluate performance of used 18650 cells...

Post by haklesup » Thu Jun 21, 2018 12:19 pm

If you can find a clock that you can start and stop with a control line, then it would not be too hard to make a 0.9V level detector with an op amp and stop the clock when the threshold is reached. Some MultiMeters had DataAcquisition ports and software, you can use like a datalogger.

this was the cheapest I could find and resembles the TEK meters
https://www.ebay.com/itm/SainSmart-DMT1 ... SwYIxX9GtY

User avatar
evahle
Posts: 186
Joined: Mon Aug 13, 2007 3:48 am
Contact:

Re: To evaluate performance of used 18650 cells...

Post by evahle » Thu Jun 21, 2018 3:39 pm

I watched a youtube video from bigclive. He used a wall clock to time the battery discharge. I'm sure there's a better way in there some where, like maybe a 60hz up counter and op amp?

User avatar
MrAl
Posts: 3856
Joined: Fri Jan 11, 2002 1:01 am
Location: NewJersey
Contact:

Re: To evaluate performance of used 18650 cells...

Post by MrAl » Sat Jun 23, 2018 8:13 pm

evahle wrote:
Thu Jun 21, 2018 3:39 pm
I watched a youtube video from bigclive. He used a wall clock to time the battery discharge. I'm sure there's a better way in there some where, like maybe a 60hz up counter and op amp?
Hi,

Keep in mind you dont have to be super accurate with this measurement. Even if you use increments of 1 minute you'll still get good results. Your 'clock' only has to count minutes and if you use constant current once you reach 0.9v you stop the clock. That will give you good results.
Using a wall clock that runs on a battery you could probably stop the clock with a relay or transistor that turns the clock battery off at the required time, just for one example.

The constant current sink should work all the way down to around 0.8v so you know for sure it always works at 0.9v or so.
LEDs vs Bulbs, LEDs are winning.

User avatar
evahle
Posts: 186
Joined: Mon Aug 13, 2007 3:48 am
Contact:

Re: To evaluate performance of used 18650 cells...

Post by evahle » Sat Jun 23, 2018 8:25 pm

That sounds good to me. My tester works pretty well though, and it only costs $4.79. I hate designing anything that I can get cheaper and better somewhere else.

Post Reply

Who is online

Users browsing this forum: No registered users and 8 guests