My 18xx bin re-visited

Announcement - Help us fight the BOTS! Please report all spam including stuff in your inbox!
Joined
Dec 2, 2017
Messages
79
I am working on the new test process with my FCDS build and have some cycle charts to share.

Since we are at a New Year, I decided to go through the cells I have tested last year on the Opus, the TP4056 and early FCDS and make some charts so I can see the nature of these cells that tested to be in the 18xx mAh range with my Opus setup. I cannot remember all the detailed history for each of these cells, but this grroup of 10 cells reminds me of the 10 pounds I lost last year. I lost those 10 pounds at least 3 times (the same 10 pounds, not 30 pounds).

So I have tested these cells more than once last year, as evidenced by the various marking on the cell jackets. All of these cells will go into a new bucket, based on the newly tested capacity using the FCDS tester. Note that a few of these cells have exceptions, and would be rejected, most would not be suitable for a Powerwall due to lower capacity.


image_jnohid.jpg

image_ncycyj.jpg

image_dmwpfa.jpg

image_nwutqk.jpg

image_wklglb.jpg
 
Joined
Dec 2, 2017
Messages
79
Here is the next batch of 10 cells.... these were charged on TP4056 and storedin the binfully charged for months.

The last cell in this series, SM12 / Slot 6 had the lower resting voltage, itdischarged fastest and had the longest charge time.
The mAh is in the high 17xx... wonder if this cell has higher I.R. ?

The FCDS uses a two tier Internal Resistance measurement based on IEC 61960.

For what it's worth I will check it at the ending relaxed voltage (not sure if that is the best place to check it, but all the cells are relaxed now)and log it for these cells. One of these cells is a Sanyo... care to guess which one ?


image_yhydkw.jpg

image_rxnxqi.jpg

image_puilhu.jpg

image_vtahpb.jpg

image_bfqkrq.jpg

image_sqgfsq.jpg


Here is a photo of the cells and the labeling scheme I am trying instead of barcodes.

Once a cell is tested, I write the Slave Module (SM 10-13), the Slot (0-7) the mAh . Internal Resistance as calculated by the two tier measurement.

I found it a pain to add barcode labels to each cell, and to give meaning to the barcode, I needed to write that number to the database, so it finally dawned on me that the SM,Slot.mAh.IR would be distinct enough to tell me all I need to know about the cell and I can query my CHARGE database using the SM,Slot and finished mAh to pull up a chart in case I want to go back and take a look.


image_heeein.jpg


The .30 Ohm I.R. measurement reported on SM11,Slot1 is due to that slot reporting high.
I moved it to 12,6 and got .15 Ohms .

This is helping to "test the tester" as well.
 

Cherry67

Member
Joined
May 13, 2018
Messages
496
I am sorry to insist on my typical "story", you cannot accurately measure IR in these (2 contact) slots.
For bad cells above 0.1 Ohm that may work, but not accurately at the lower ones.
 
Joined
Dec 2, 2017
Messages
79
No problem. Let's be clear, the test is a two tier DC load test where the I.R. Is calculated.

I was thinking about sending some cells to @Wolf if he would be willing to check the I.R. Using the 4-wire method so I could know the delta between these techniques.

My concern is reliability of measuring at the "bottom" since that is where I can do it easiest on my tester .
 

Cherry67

Member
Joined
May 13, 2018
Messages
496
SecondHandPower said:
No problem. Let's be clear, the test is a two tier DC load test where the I.R. Is calculated.

I was thinking about sending some cells to @Wolf if he would be willing to check the I.R. Using the 4-wire method so I could know the delta between these techniques.

My concern is reliability of measuring at the "bottom" since that is where I can do it easiest on my tester .
Understood.

You know i published a cheap DIY measure circuit for four wire testing ?
Update : https://secondlifestorage.com/t-Kelvin-measurement-4-Wire-connection-Inner-Resistance?page=2

(and an upgrade will follow soon...)
 

daromer

Moderator
Joined
Oct 8, 2016
Messages
5,659
You can potentially half your IR values by just adding 2 more wiring for the Voltage measurement :)
 

Cherry67

Member
Joined
May 13, 2018
Messages
496
daromer said:
You can potentially half your IR values by just adding 2 more wiring for the Voltage measurement :)

That is really NOT the correct answer of a qualified quality engineer, which is focused on datasheet usage .... :D
 

Wolf

Active member
Joined
Sep 25, 2018
Messages
1,613
SecondHandPower said:
No problem. Let's be clear, the test is a two tier DC load test where the I.R. Is calculated.

I was thinking about sending some cells to @Wolf if he would be willing to check the I.R. Using the 4-wire method so I could know the delta between these techniques.

My concern is reliability of measuring at the "bottom" since that is where I can do it easiest on my tester .

SecondHand,

You are more than welcome to send me a couple of cells if you want. I assume you are stateside so ground shipping is not a problem.
I can run them through all my testers alsoto see if our methods agree.

Just PM me and I'll give you my address.

Wolf
 
Joined
Dec 2, 2017
Messages
79
Hi @Wolf,

I pitcked out more than a couple... the 6 greenies are Samsung ICR18650-22 [F|FU|H|HU] ranging from 19xx to 14xx mAHlh.

On the Sanyo UR18650F's, there is a leaker/heater and a really low mAhand 18xx, 17xx.

The silver writing shows my most recent test results
Here is the format:
module.slot.mAh.two_tierOhms

All cells are at their resting voltage after discharge to 3.0v. The resting voltage is where I ran the two-tier I.R. test routing.


image_xfiyhp.jpg

I have the deplete/charge/dischargecycle charts for most of these cellsif they would be of use, they look similar to the onesposted above,

Just curious how close thistwo-tier measurement may beto the 4 wire and if the numbers track with your study of these cell types.. Unfortunately, there are several slots involved on my setup, so that will add some extra scatter to my numbers. These were tested once on the two-tier... I did not repeat the measurement until a mode appeared.

I will understand if you don't want to bother with this after all, just let me know if you want to try these
 

Wolf

Active member
Joined
Sep 25, 2018
Messages
1,613
SecondHandPower said:
Hi @Wolf,

I pitcked out more than a couple... the 6 greenies are Samsung ICR18650-22 [F|FU|H|HU] ranging from 19xx to 14xx mAHlh.

On the Sanyo UR18650F's, there is a leaker/heater and a really low mAhand 18xx, 17xx.

The silver writing shows my most recent test results
Here is the format:
module.slot.mAh.two_tierOhms

All cells are at their resting voltage after discharge to 3.0v. The resting voltage is where I ran the two-tier I.R. test routing.


image_xfiyhp.jpg

I have the deplete/charge/dischargecycle charts for most of these cellsif they would be of use, they look similar to the onesposted above,

Just curious how close thistwo-tier measurement may beto the 4 wire and if the numbers track with your study of these cell types.. Unfortunately, there are several slots involved on my setup, so that will add some extra scatter to my numbers. These were tested once on the two-tier... I did not repeat the measurement until a mode appeared.

I will understand if you don't want to bother with this after all, just let me know if you want to try these
Sounds good I will PM you my address. I am looking forward to doing this, Just so happens I am doing anew test on the ICR18650 -22[F|FU|H|HU] series of batteries #1441 to #1485 in the sheet. I have80 or so of these cells and binned any over 60m? IR to be tested later. The ones below 60m? are being tested as I am writing this.The first setto #1452 have been recorded and show interesting results. :) Once you look at the sheet you will see.
The procedure I dois an initial test of IR and V to see if the cell is > 1V and IR is60m? or below. Then full charge and IR and V again and recorded in the sheet. Then results of mAh recorded from the Opus. After a couple of days usually almost a week if not more a second IR and V test recorded then after 14 days again (usually 3 weeks or more) as my backlog of the 2ndand 3rd test is catching up with me. :dodgy:
But I will prioritise yours! :D
I have 14 chargers running 24x7 LOL
Only so many hours in the day.

image_ikzhsd.jpg

image_ibjufr.jpg


Wolf
 
Joined
Dec 2, 2017
Messages
79
Ok, so Thanks @Wolf for offering to take a look at those (10) cells to see what they say to you and the 4 wire testing you do.

I have been continuing with my clean-up efforts... I am trying to get all the previously tested cells into their proper bin once I have tested everything with my new FCDS test process.

I have one more batch of photos. These are the last of the 18xx and into some 19xx that apparently were tested with the FCDS before I started capturing graphs.
I am including a picture of the cell with the graph because I am trying out a couple of image editing and screen shot programs.

If you wonder about all the scribbling on my cells, here's how that happened:

The silver color is for the most recent testing with the FCDS tester where I captured the data from thetest cycle to make graphs.
Blue writing couldbe an Opus charge/dischargeor early FCDS when I did not have data capture.
Black writing could be Opus or the TP4056 charge with Opus orHW-586 discharge board...

...So you can see why i wanted to revisit my 18xx bin and test everything againwith my newFCDStest.


image_shqyio.jpg


I have marked up the first couple of charts to point out something I have seen only on this type of cell. [CGR18650CG, Japanese] I am sure the spec sheet may have the explanation, but these cells have the lowest mAh I have seen from a resting voltage discharge while having the healthiest mAh rating after a full charge/discharge.

Take a look at the initial discharge from the resting voltage (the left side of the chart where the mAh rises for the fist part of the test cycle.)
Compare to some of the other charts and cell types.


image_fmalnq.jpg

See ? This guy is different. There is another one later on, so you can see this is not a tester or slot issue, we are able to seesome differences in the cell specs of different manufacturers. Some cells are "stronger" at the bottom, some take longer to charge, some heat up... etc. etc.


image_ulhcio.jpg

image_bgelbq.jpg


During this exercise, I may havefound a technique to help improve screening lower mAh cells. However,I noticed at least one cell type that does notfit my rule for this experimentalmodel.

I will start a new thread on this idea of how to improvetester slot utilization by binning lower mAh candidates in a shorter time on the tester.
 
Top