How I Recondition Cells

Added current resting voltage and resistance to right end

 

A nice hard to comprehend chart
This is based on the zero state of charge resistance, which may mean nothing, as resistance at 50% SOC may be completely different
Looking at the discharge voltages at 3 amp hours and grouping the mods into a bunch of resistance group, lowest resistance is Yellow Highest is Black (not really, DASHED LINE RED is higher than black, ran out of colors)
One would assume highest resistance would have lowest volts and vice versa. Assume lol.
discharge was from 90% soc

 

CSV file for the discharge

 

Look how terrible this 2008 cell looks compared to my 2004 2005 cells

 

no not yet, but i am examining the curve on these, and will post more when i know, all 28 cant possibly be "defective". so stay tuned, the test is ongoing now. March 2008 cells. IF they all show this weirdness, I have a source to get a few more to test.

If nothing else, this shows why you need to graph, these would pass a capacity test (the ones I have tested since I posted, yet still show this curve deviation). Mixing these with 2004 2005 with a flatter line would be "interesting" in the cars diagnostics monitoring systems

 

GRAPHING
there is a poor mans way that may provide a higher return on investment than buying a dedicated device

cheap multimeter with serial port output,

more cumbersome and steps, you open a terminal window in your pc and capture the data, then import into excel etc to make graphs, more work but for someone that wants graphs without a CBA unit, there is a possible solution.

(there is a cheap unit 18 dollars if you buy 100 of them, it is 75 on amazon per unit, i think there may be cheaper alternatives)

one could even make one themselves or find a MAKERS unit that has the capability to just read volts and report them up the serial port, i havent looked but it probably exists and is under 10 dollars.

this is something that can be made to work as arduino can connect to your pc by default, hundreds of ways to do this for dollars (dang maybe i need to make an open source CBA using arduino and throw it out there in the public domain)


How to Make a Simple Digital Voltmeter with an Arduino

 

oh i understand, i am speaking to the 12,000 that are reading this lol
i am the person that used a car battery charger for years cause i had 6 of them , i am cheap

i think you can make an interface to log voltages while charging discharging using ONLY the arduino and its usb to your computer.
latest arduino is $8.83

(and the phone call, that interrupts my video, is my sister who uses a spoofed name on her phone, so i didnt answer her, and she was broke down in a prius at a 711, she finally got it home, and i found the brake lights are intermittently not coming on, which means you cant get it in ready mode lol ) now i have to fix that

 

In addition to wirelessly reading all the temps on all 28 modules and their voltage, and recording it for download over an intranet or the internet, I can also add the ability to balance the entire pack using this new design, i need to build it )

Oh, and I can add a SWELL ALERT SENSOR and ALARM lol

If I did what I did to the honda insight, I can also get into the prius computers and start restricting the charge to protect the modules until the issue is dealt with. Pretty endless the things you can add that toyota should have done in the first place

I can get the chips for under 2 bucks each

 

When I state, wait one hour before continuing with a 2 amp charge until end of charge, that is on a module you are just starting with and know nothing about, once it has been cycled at least once, I no longer pause, I change to 2 amps, and continue the charge. If you cycle them and they no longer take more Ah and/or, the capacity stops increasing, then it is complete, that's 3 to 5 cycle normally.

Swelling and gassing are two different things, they are not related. Each can happen without the other. Gassing pressurization and venting are a chemical process and are minimally related to heat, swelling is a manual process and is directly related to heat, this is where the collector BETWEEN the neg and pos anode and cathode physically swells, and that can damage the module, there is a "band" around the assembly that can break, at which point layers of material connect short or disconnect from each other.

Once the module is done and fully operational, if i need to to do further test or low end work at the 6 volt area, and that requires charge discharge, there is no reason to fully charge the module over and over and stress it out, just put in 6.5Ah (assuming a fairly good hi cap reconditioned module) and discharge from there and note the capacity, which should be approx 5.75 Ah, use that "Ah in" and "90% capacity" as the baseline going forward instead of the 8Ah in 6.5 Ah capacity, and you wont be stressing the module over and over as you "try different things"

 

I have become intrigued by what I see here, RED is the freshly worked 2004 and 2005 modules for my honda insight pack. The BLACK are the 28 modules (only odds 1 through 11 so far, charged discharged twice in a row past 48 hours) that I got from one car, all are in serial number series and original to the car. All the black display a similar discharge path, as do the reds, but they are quite different.

The REDS have had the most work done to them, but all have been cycled the required number of times to 100% EOC. The BLACKS have not been worked down at the 6 volt level as the RED were. Even so, the curve the BLACKS display is not one I have seen so prevalent in one pack before.

This graph was based on 90% charge not 100%.

The far RED has 5.8 Ah at 90% charge.

So, I am going to work on a few of these and see where they go with respect to that curve.

 

5 and 7 seem like the most stable, the others all seem to have issues, so I will go work 5 and 7 and see what happens.

 

Actually, I pulled my data and found that module 16 had proven to be the best of that 28 mod pack from one car, and I added it to the mix below, this blue line is mod 16 after several 100% charges, this is in fact a 100% discharge, being compared to 90% discharges, the odd volt curve was still present, so as for any more work I may do, forget 5 and 7, mod 16 is the one to focus on to see if this weird 2008 curve sticks.


The blue line was sourced from below