18650 Harvesting Flow Charts

Wolf

Moderator
Joined
Sep 25, 2018
Messages
2,004
As a community we all have our different ways of processing cells and I wanted to take the opportunity by starting this thread to find some common ground on how we all should harvest cells.
This is not meant to be a "be all end all" but more of a guide for new members and maybe some senior members alike.
Constructive critique is appreciated as any input that furthers our voyage will make us all better harvesters.
In this instance I will share my Flow Chart that I use it to process my 18650 cells. Kudos to Daromer who as far as I know was the first one to publish his flow chart.
Other Chemistries and formats are going to be different so hence the title "18650 Harvesting Flow Charts"

My chart assumes that you have a supply of 18650 battery packs and have a good understanding of what to expect the pack has for cells and what the likely mAh is of the cells that you will be harvesting.
For Example a 2.5Ah 14.4V pack will most likely have a 4s1p setup with 2500mAh cells
the same runs true for a 6Ah 11.1V pack will most likely be a 3s3p with 2000mAh cells
You should by now be able to tell if the packs are 3s2p or 4s4p and so on.
This chart also relies heavily on IR (the internal resistance of the cell as measured with a 1kHz AC 4 wire meter) as the initial pivot point.
Meters such as the YR1030, YR1035+, and RC3563 supply us with that measurement, are reliable and with my testing have been accurate.

If I open a pack and I spot cells that are corroded or have physical damage I close it up and wrap duct tape around it as it is much easier to drop off at the recycler than individual cells.
The chart assumes if you open a pack you will separate the plastic casing, electronics, and nickel strip residue for recycling.
BTW please publish your flow charts so we can compare notes.
Here is my Flow chart:

image_bxfygz.jpg


Wolf
 
Last edited:
Last edited:
Redpacket said:
Would be great tocollaborate& get these into the FAQ with Korishan

Yes that is why I started this tread.
FAQ maybe when its refined and if we all can agree, :p

Side note:
I'm still on the hunt for a FET that will open proper at a bit lower V for the ESP32 to get full flow.
Of course I can always rewrite the sketch for a Nano its just the speed of the ESP32 that's the advantage right now.
Wolf
 
Suggestion - maybe2.6, 2.7, or 2.8v instead of 3v to begin charging as2.8v is often the specified low v for 18650 rather than 3.0v. Maybe 'newer' cells are bit less risk than 'older' cells also. For a real worldexample, the RING battery packs (newer cells - e.g. not likely to be stressed much)have a significant % where the cells are at 2.6v and they're perfectly find just charging them directly (using the OPUSs).

OK, just a comment... :)
 
OffGridInTheCity said:
Suggestion - maybe2.6, 2.7, or 2.8v instead of 3v to begin charging as2.8v is often the specified low v for 18650 rather than 3.0v. Maybe 'newer' cells are bit less risk than 'older' cells also. For a real worldexample, the RING battery packs (newer cells - e.g. not likely to be stressed much)have a significant % where the cells are at 2.6v and they're perfectly find just charging them directly (using the OPUSs).

OK, just a comment... :)
Perfect good point.
I feel that if a cell is at ≈ 2.5V and it is a rather young cell with good IR you should have no problem popping it into a charger/tester at 1A.
If a cell is < 2.5V I feel it would be wise to recover charge to 3.0V. The difference in charging time between 2.8V to 3.0V is maybe a half hour or so.
But hey that's why we are discussing it. :)
Wolf
 
Last edited:
Wolf said:
OffGridInTheCity said:
Suggestion - maybe2.6, 2.7, or 2.8v instead of 3v to begin charging as2.8v is often the specified low v for 18650 rather than 3.0v. Maybe 'newer' cells are bit less risk than 'older' cells also. For a real worldexample, the RING battery packs (newer cells - e.g. not likely to be stressed much)have a significant % where the cells are at 2.6v and they're perfectly find just charging them directly (using the OPUSs).

OK, just a comment... :)
Perfect good point.
I feel that if a cell is at ? 2.5V and it is a rather young cell with good IR you should have no problem popping it into a charger/tester at 1A.
If a cell is < 2.5V I feel it would be wise to recover charge to 3.0V. The difference in charging time between 2.8V to 3.0V is maybe a half hour or so.
But hey that's why we are discussing it. :)
Wolf
 
507PowerWall: Not sure what your last post was about. Please if you answer a quoted post that your reply is "outside" of the quoted section. If you want to make sure of the quoted section, always preview before posting.
 
Korishan said:
507PowerWall: Not sure what your last post was about. Please if you answer a quoted post that your reply is "outside" of the quoted section. If you want to make sure of the quoted section, always preview before posting.

I did qoute, but sometimes this happen.... moreover, Im not able to see my own post that you are making reference :(

CG.
 
It seems that only RedPacket is taking heating into account in its processing. How do you do it. With the bare hand approach?

I've found myself how difficult is to physically control cell heating and at the same time cooling the testers without affecting each other.

I'm designing an ESP driven gadget to give those additional features to the existing chargers & testers. Each ESP will have 8 temperature probes that will be mounted into a 3d printed clip that I will dock to each cell. I'll send measurements to a DB and grafana for reading and alerts.
I'm figuring out how to measure voltage so I can log the temp along it. Measuring voltage would be easy, just adding multiplexers and perhaps an ADC for differential reading, but..
Is it possible/reliable to read voltage while being charging or testing with an independent device?
Would be "good" to join the negatives on Opus/LII500/TP4056 cells for simplifying voltage reading circuit?

That counts as a part of my processing flow? ;9
 
SeRiusRod,

Ah the heating of the cell.... or "Heaters" as we know them.

There is one thing I have observed and that is, if the IR is measured first and it falls within my cheat sheet perimeter the chances of having a hot cell is virtually eliminated.
I have tested the so called Sony heaters with above my cheat sheet IR and they have run hot some very hot. The same cells with proper IR did not heat up beyond the normal charging / discharging temp.

So that is why I do not have a temp check in my flow chart.

Wolf
 
Last edited:
I've updated the flow chart to larger size & included IR testing.
Let me know any thoughts....Cell testing flow chart bigger w IR.jpg
 

Attachments

  • Cell testing flow chart bigger w IR.pdf
    28.2 KB · Views: 174
Excellent flow chart. I might do this section in terms of '% of original specs' instead of a fixed 2000mah. Perhaps 80%? As its OK to build a DIY powerwall battery with 1700mah cells that are 80% of 2150mah original specs.
1612278835298.png
 
Note that some power tool high drain low capacity cells can reach below 1V and still be good (near max cap). I recovered some like that (like Samsung 13 and 15Qs).
 
As a noob question regarding the "heaters," I don't think I caught a direct answer as to what is considered a "heater?" Is this a direct temperature measurement with an IR thermometer or some other method?
 
As a noob question regarding the "heaters," I don't think I caught a direct answer as to what is considered a "heater?" Is this a direct temperature measurement with an IR thermometer or some other method?
If a cell is too hot to leave your finger on it - its a heater for sure. 120F/50C is likely too hot for me. https://www.answers.com/Q/What_temperature_is_too_hot_to_touch

When cells are charging up you should touch them (or measure their temps) to verify they don't get hot. A bit warm is OK but not 'hot'. With a little experience, this becomes easy to tell.

Also, once a cell nears full charge... they may take a long time to reach the final 4.20v which can indicate self-discharge. The self-discharge / continued charging can create really hot cells.

Usually you process cells in batches - meaning a group of 100s or 1000s from modem batteries or medical packs OR of a certain type / age. For example, I just processed several thousand RING battery packs and all were OK charged out of the pack, had 100% capacity, and none got hot at all - so I hardly checked after the first several hundred. Just a quick touch as they charged. On the other hand - some medical packs I processed had hot cells every 20 cells or so... so I checked each one very carefully with infrared thermometer and with a 40C cut-off. .
 
As a noob question regarding the "heaters," I don't think I caught a direct answer as to what is considered a "heater?"
@Doc3G Great question. There is a lot of info on what we refer to as "heaters" and many of us have written a lot on it.
Check out this thread.

That being said, and knowing what we know you really shouldn't even experience a "heater unless you are doing some sort of experiment.
The internal resistance (IR) of a cell is the guiding light that will tell you if you have a healthy cell or not. That simple.
If the cell passes the IR test, now mind you it is with a proper IR meter such as a YR1035+ or RC3563 4 wire 1kHz tester, then you can be relatively sure you will not have a heater.
Is this a 100% guarantee, you will never have a heater at proper IR readings? No.
But one thing I can say is that after testing over 12000 cells and processing at least 8000 of them (≈4000 had a defect one way or another including high IR) I have never had a "heater" as long as the IR checked out.
PS any cell during charging at 1A that gets to be >45° C would be in my book a "heater". Matter of fact I set my temp limit on my chargers to 40°C
and a proper cell should really not exceed 35°C during a C/D/C cycle.
1618919406379.png


Wolf
 
@Doc3G Great question. There is a lot of info on what we refer to as "heaters" and many of us have written a lot on it.
Check out this thread.

That being said, and knowing what we know you really shouldn't even experience a "heater unless you are doing some sort of experiment.
The internal resistance (IR) of a cell is the guiding light that will tell you if you have a healthy cell or not. That simple.
If the cell passes the IR test, now mind you it is with a proper IR meter such as a YR1035+ or RC3563 4 wire 1kHz tester, then you can be relatively sure you will not have a heater.
Is this a 100% guarantee, you will never have a heater at proper IR readings? No.
But one thing I can say is that after testing over 12000 cells and processing at least 8000 of them (≈4000 had a defect one way or another including high IR) I have never had a "heater" as long as the IR checked out.
PS any cell during charging at 1A that gets to be >45° C would be in my book a "heater". Matter of fact I set my temp limit on my chargers to 40°C
and a proper cell should really not exceed 35°C during a C/D/C cycle.
View attachment 24730


Wolf

Thanks for the info, I think I could pull off that case mod. I don't have a printer but my sister-in-law does so I can likely convince her to print one for me. I looked at the two meters that you listed and they were reasonably priced. My question is, what is that fancy setup you posted and how many people do I need to murder to get something like that?

Is there not an off-the-shelf solution that provides reasonably accurate readings for IR, capacity, voltage, etc and that can charge and test?
 
My question is, what is that fancy setup you posted and how many people do I need to murder to get something like that?
🤣 Ohhh you have no idea......
Is there not an off-the-shelf solution that provides reasonably accurate readings for IR, capacity, voltage, etc and that can charge and test?
Sort of. The equipment I use is the MCC (Megacellcharger) "hardware" and MCM (Megacellmonitor) "software".
20210104_201704.jpg
The initial release of the hardware and software was plagued with issues that have since been somewhat resolved.
The MCM software developer has been really great and relatively responsive. https://www.facebook.com/groups/646858045823609
The hardware guys not so much. See my post here https://secondlifestorage.com/index...ellcharger-has-shipped-yippee.9855/post-69397 and here https://www.facebook.com/groups/646858045823609/permalink/1000652300444180

Now once I got both of them dialed in I am relatively satisfied with the results.

To get proper IR and V readings you need to go with a RC3563 (my favorite) or an YR1035+ see this thread https://secondlifestorage.com/index.php?threads/ir-meter-recommendations.10361

Question how many cells do you plan on processing? If you are in the market for many thousands as in 10000 plus yea the MCC is a wise investment.
I did my first 6000+ with 14 chargers. Very time consuming recording everything into a spreadsheet.
20190211_062350 (1).jpg
With the MCM I can copy and paste the results into a Raw datasheet which in turn I can link into my record of the cells tested.
1619012518886.png
And last but not least my spreadsheet were once all cells have been tested and entered for the battery build I can then sort by criteria and pull the best of the best. This represents the testing cycle of my flow chart. Matter of fact I have just started my 4th battery build.
1619012817524.png
Wolf
 
Back
Top