Identifying dangerous cells

Battery

Battery

New member
Joined
Jan 4, 2017
Messages
11
Experiment:
Resetting Current Interrupt Devices (CID's), What are the features of a heatdamaged cell?

I have just measured the internal resistance from a few cellswith initiated CID's.These cells were froman oldeight cellIBMbattery of(cream paper wrapped cells)and a 6 cell Dell battery(4.3V 28A SDISamsung cells 2013?).Out of interestI'mnot sure who made the super old looking IBM cells. They aremarked on the side withEHPPT2B and some other codesthat varies with each cell eg[012055]they haveaSstampedon the negative terminal. They use a similarfont and layout asSanyo cells.

In the IBM pack all eightcells had triggered CID's and 5/6in the Dell. Once the CID's were popped back in the IBM Cells were sittingat 1.5-3V and theSamsungat4V. Considering their salvaged voltagesI assumeboth packswould havefailed at a reasonably high sate of charge, for theSamsung 4.3V cells ~4.05V is approximately 80% SOC. Also the IBM cells seemed to vent some gas/electrolyte when I reset them.

Since almost all the CID's were triggered in the packsand thattheyindependent devices we can assume that the packs got hot also triggering thePositiveTemperature Coefficient element(PTC). The PTC can be triggered in two ways reversibly and in the case of extreme temperaturesirreversiblywhich leads to permanent damage of the cell*.

Results:
Internal resistance inmiliOhms:
IBM:2992,456,300,808,300,301,1517,250 (LOW 0% SOC)
SAM:176,200,250,220,310,172 ("working" cell)(High 80%SOC)

Internal resistance measurements are often wrong with cheap equipmentso this is but a qualitative test. For typical good cells I measure~60-150 miliOhmsTypically ~100miliOhms at 100% SOC**.What does this all mean? With fewstatistics it is hard to draw just yetthe linebut itlooks like at a high state of chargean IR of200 miliOhmsor higher could be asign of significant internal damage and reason enough to discard the cell. For a low state of charge perhaps 50-100%** more would be acceptable ~300 miliOhm for borderline OK cells.

Here is the general rule I have found posted on forums for3.6V (~0-33% SOC ***):

Milli-Ohm
Battery Voltage
Ranking
75-150mOhm 3.6VExcellent
150-250mOhm3.6VGood
250-350mOhm3.6VMarginal
350-500mOhm3.6VPoor
Above 500mOhm3.6VFail

So it seems the general wisdom is probably correct. One side note I would make issince some battery chemistriesretain a significant level of charge at 3.6V (Panasonic NCR18650B *** 33%).All things are not equal for a low SOC IR measurement until around3.4V, For now I willjust measure all IR at 100%SOC ... Unless this is a bad idea? I really should measure the IR for some cells before and after charging.

Next to do is test the capacity of these cells somewhere safe....

References:
*Safety mechanisms in lithium-ion batteries(PDF)
Journal of Power Sources 155 (2006) 401414
**IR vs SOC(PDF)
(I don't think the lithium cells here are 18650)
The table shows that when the SOC decreases from 100 to 15% the total internal resistance Ri (=Rp+Ro) increases with 50-100%,especially due to Rp. Rp has a more dynamic character in comparison with Ro which stays nearly constant. [Ri = Internal resistance (general, total)(=Rp+Ro)] [Ro =Battery internal "ohmic" resistance] [Rp =Battery internal "polarization" resistance]
***SOC vs voltage
 
It depends a little on how you are measuring it, when i tested my first pack using essentially kelvin connections and a constant current load i was getting between 22-55 for panasonic 2200mah cells once fully charged, and just for error checking, i then hooked them up in series and retested to rule out the possibility for a measurement error and was within 5 milliohm,

Also the ESR does actually change somewhat depending on how long you test for, and how small the test current, at 5mA my esr reading where quite a bit lower, which i believe was surface charge rather than true chemical resistance, while at a short pulse of 15A, the esr was almost double, likely due to limited reaction rates in a cell that has been sitting doing nothing for a week.
 
In my youtube perusing, I came across this really easy to follow video:



He sure made things a lot simpler and easier to follow :)

He does mention that when testing cells, to compare apples->apples, oranges->oranges, you should measure internal resistance at the same voltage as resistance changes as voltage changes. So, I'm guessing you'd want to charge low voltage to cells to a common minimum to test low resistance, then charge to full capacity (or as he mentions, significantlyincrease battery life by slightly lower voltage; 4.1v instead of 4.2v due to SEI buildup), or just under and make sure all cells tested are at the same voltage and test resistance at that point.


Would it be a good idea to measure resistance at bottom end as well as top end? Would it make a big difference? I'm thinking it'd be a good indicator as to how the cell handles over time and maybe even be a predictor of how long the cell might last b4 it becomes beyond rechargability.

Kori
 
I have had packs where all CIDs are triggered on all cells that is. Im pretty sure it was due to charging and heat. They all came from a place where they had the laptops in place where it was hot and charged them all the time. Of course i also had randoms cells too.

I have not had any good success with popped cids since all had low capacity and also high ir. The CID pop for a reason.


So basically my though here is:
* They did pop for a reason -> And since they did gas something have happened inside of them. Ie mos likely degraded.
* How can we safely ease the load inside the cell and make sure that the function of the cid do remain? How can we check this function after? Most of us just press the cid gently but it feels like we may mess with that function :)
 
Korishan


Would it be a good idea to measure resistance at bottom end as well as top end? Would it make a big difference? I'm thinking it'd be a good indicator as to how the cell handles over time and maybe even be a predictor of how long the cell might last b4 it becomes beyond rechargability.

Kori

I'm not sure if taking two measurements will give you much more information.I like the top end because after discharging my cells are automatically charged and i take the reading then 4.2V is ~100% +-5% S.O.C for most cells. Different cell chemistries converge to 0% around 3.4V and this would give the highest possible IR so it may be useful... idk
 
I had four that had the CID popped. I reset the CID and they had around 3.8v without even charging. They showed no signs of abuse or anything like leaking. I capacity tested them all around 2200mah. I let them sit a good 2 weeks and they didn't loose any voltage so I threw them in my packs. I wish I would have remembered which ones they were so I could check up on them. I haven't tested the theory with any other cells yet. Too many good ones to process than to worry about refreshing bad ones.
 
Sorry to revive an old thread but I've been trying to measure the IR of my recycled and nearly fully charged 18650s. I used the instructions from the Youtube video.
Using a voltmeter, alligator clips and a 20W 1Ohm resistor https://www.amazon.co.uk/gp/product/B07D7Y61ZR

The problem I am having is that the voltage reading is exactly the same when measuring the voltage of the battery directly and through the resistor. Could my resistor be damaged? I tried two resistors with same outcome.

Measuring the resistance across the ends of the resistor shows 1.4.


What would account for this?
Thanks

image_kccfwj.jpg
 
dwarflundgren said:
Sorry to revive an old thread but I've been trying to measure the IR of my recycled and nearly fully charged 18650s. I used the instructions from the Youtube video.
Using a voltmeter, alligator clips and a 20W 1Ohm resistor https://www.amazon.co.uk/gp/product/B07D7Y61ZR

The problem I am having is that the voltage reading is exactly the same when measuring the voltage of the battery directly and through the resistor. Could my resistor be damaged? I tried two resistors with same outcome.

Measuring the resistance across the ends of the resistor shows 1.4.


What would account for this?
Thanks
Click to expand...

@dwarflundgren

There are several ways to measure IR (Internal Resistance) of a cell.
The right way and the wrong way.

The right way is with a 4 wire kelvin 1kHz m? meter just like the manufacturer spec sheets shows.
Example:

image_iozttf.jpg

Meters like this are readily available for ~$50.00 US.

image_usygaz.jpg


And then there is the wrong way.
Enough said.

Wolf
 
dwarflundgren said:
Is this the wrong way?
Click to expand...

That Video shows you the DC Internal resistance. It is another way but not the prefered way especially for powerwalls.

The manufacturer does all IR with AC Internal resistance measurements.
Here is a video to explain the right way and the difference between AC and DC IR and why.


Wolf
 
For an Ebike battery built from recycled laptop batteries which is the more significant internal resistance to measure?
AC Internal Impedence or DC Internal Resistance? Presumably DC Internal Resistance right?

I believe I was following the correct procedure to measure DC Internal Resistance but getting the same voltage readings. I'll get the resistors used in your video and re-test to see if it makes a difference.
 

dwarflundgren said:
For an Ebike battery built from recycled laptop batteries which is the more significant internal resistance to measure?
AC Internal Impedence or DC Internal Resistance? Presumably DC Internal Resistance right?

I believe I was following the correct procedure to measure DC Internal Resistance but getting the same voltage readings. I'll get the resistors used in your video and re-test to see if it makes a difference.
Click to expand...

Good question.

I don't know as I only go by what the manufacturer recommends on how to test IR which is with a 1kHz AC impedance meter.
Additionally here is a PDF you may find interesting.

https://hiokiusa.com/wp-content/uploads/pdf/18740-UG_WP_BT4560E1-56E.pdf


Not my video just some guy on youtube that explains it.

Wolf
 
i assume there was a valid reason for the cid to trip.
the big unknown is whether it will reliably trip again if it needs to.
cells that appear open go into the scrap bucket.
i suspect there is some sort of undesired reaction going on in old cells that generates gas.
i have seen packs that were very old but unused with all cells tripped.
 
You must log in or register to reply here.
Back
Top