Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

check state of multiple li ion cells

Status
Not open for further replies.
I have lots of 18650 Li ion cells from laptop battery pack, I wanted to know if there was a way to compare them with each other to determine the good ones, bad ones and ones that can be paralleled together to make a larger capacity pack. I have measured the terminal voltage, discarding the ones reading less than 2.8 volts. most of the packs I have dismantled had two cells in parallel and three or four series so have left good pairs of cells connected together. I was thinking of charging each cell up and discharging for a certain time and grading them on the final terminal voltage. then parallel similar cells and make a pack up with the best ones. because the cells are going to have different wear on them, I'm going to have a PIC monitor each parallel group and charge each group separately rather than in series. are there any other things I should consider?

thanks,

Jules
 
Unfortunatly I can't find the thread so I'll have to repeat it but I suggested the bellow method a ways back to a user. Never implemented it myself but it will allow you to 'fingerprint' a battery pretty accuratly.

After charging (to identical charge states) Discharge them through an adjustable constant current source and sweep the discharge rate from 1/10th C all the way up to 4C over about 2 seconds, you'll need a scope or at least a micro controller with an ADC that can give you a few thousands samples a second. Match up cells that have similar curves. Ideally you'd want to monitor it's peak temperature as well as that will be an extra datapoint for a good match, and you'll want to make sure that any cells that you use in series are VERY close to one another.

Do NOT monitor termination voltage, it is useless to determine state of charge or internal resistance of a Lithium cell, the only way to do that is under actual load which is why I recommend the swept discharge. Seeing as how the discharge sweep I recommend is only 2 seconds you could increase the peak discharge to 10C possibly higher, the higher the better for getting an accurate graph that will 'fingerprint' the battery, make sure the packs have been charged to the same state (I recommend constant current of 1C until 4.1 volts is reached that's about 80% state of charge for most Lithiums) and have been allowed to stabilize to room temperature (72F +/-1 2 degrees)

I only suggested the 4C discharge first because many common cells aren't rated for much more than that, but with it being only a 2 second discharge and a ramp up the risk of fire is extremely low, the primary problem this creates is the higher C you use the higher wattage the constant current source has to be able to handle. Be wary of damaged cells and have a fire extinguisher nearby just for general saftey.

Seeing as how power generated vs resistance is not linear, it's [latex]I^{2}R[/latex] a linear ramp is not ideal, it should be an exponentially increasing sweep if possible, it will be easier to visually spot matching cells as they'll be pretty close to a straight line.

Not easy but I'd be curious if you decide to build something like that as I'm curious as to the specific curves you'll come up with. Refining from the results will allow you very accurate matching of cells, both new and old.

I would be a little wary of final characterization based on this though as measuring the cell voltage at a 1C discharge till at least a 20% charge state will give you a separate graph that will tell the rest of the story about a cells state.
 
Last edited:
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top