Li Ion Charging of Very Old Cells

Having satisfied myself that leaving Li-on cells permanently on charge at a nominal 3.9-4.0V appears to do no harm to the cell I turned my attention to the pile of recovered cells I have. These were mostly from laptop battery packs but some came from old handheld radios. I have been using cells recovered from old Apple laptop batteries for many years now and never given it much thought. They just worked and allowed me to power many projects away from the bench. The newer cells I have from mobile phones and cameras are not the subject of this experiment.

I easily found a research paper that dealt with extreme discharging. However, I am not examining cells that have been forcibly discharged to the point of polarity reversal. Instead, I am looking at cells that self-discharged down to almost 0 volts.

It would be a mistake to believe a cell left to self discharge to this point is ruined. The self-proclaimed experts never present any research to back their assertions and most seem to be plagiarizing from each other. 

 One example of the utter nonsense on this subject: Do not boost lithium-based batteries back to life that have dwelled below 1.5V/cell for a week or longer. Copper shunts may have formed inside the cells that can lead to a partial or total electrical short. 

Some people accept this as gospel but without ever questioning the basis for such a claim. What is the definition of partial electrical short? Where did "1 week" come from? Is 6 days 23 hours safe? Why wasn't the threshold 8 days 12 hours 52 minutes? I wonder how many believe the earth is flat?

So I am running a trial to at least see if a very old cell recovered from a battery pack can be useful. Some of the cells in this trial are close to 20 years old taken from old two-way battery packs. Those removed from old laptop battery packs were at least a decade old. Before charging the terminal voltage was less than 2.7V. Many were below 1V and some as low as 0.2V. 

I doubt any of these cells had seen a charger for at least 5 years, and some would be close to 10 years. I applied a constant 2mA until the cell reached around 2.7V. Then I used my CC/CV charging circuit to charge to 4V. I am leaving these to sit at 4V between discharge trials for several days to a week. That's to reflect how I intend to use these cells in real life. 

It's too early to post data on what has happened but the initial results suggest the cells are useful. Once parallel combination of two 18650 cells recorded just over 3000mAh on the first trial. Not bad for cells that had been left discharged for over 5 years then charged to 4V or about 80% of capacity.

I'll provide a full update in due course but for now it appears that a Li-Ion cell can be left sitting discharged for several years. To restore it to service charge at a few mA till the cell voltage rises to 2.7V. Then it can be charged with my CC/CV circuit mentioned earlier.

The only challenge I have faced to date is separating the cells. The pouch style cell has connecting tabs easily damaged and once broken you need a spot welder to attach a copper flake to the aluminum tab for soldering. Not something I can do so I reluctantly discard such cells.

Joule Smasher Led Flasher - Far better than the Discrete Version

While I waited for the battery life tests to run their course I wondered just how flat a AA battery could be before the Joule Smasher Led Flasher stopped working.

I had some AAA batteries that had stopped powering the discrete version simply because the voltage step across the LED was too small. So I transferred those AAA batteries to a Joule Smasher and started another course of measurements.

One of the batteries had a cell voltage of 763mV so I am interested to watch this one carefully. I missed the original AG3 battery failing because I went on a holiday.

Clearly though the Joule Smasher far exceeds the performance of the discrete version. If you must build a discrete version contact me because I have about 100 of these destined for the tip.

Emergency Workshop Lighting - V1.0

After testing some Li ion cells I realised how useless those Chinese sourced 18650 cells can be. As the pile of cells grew I wondered what I could do with them. 

It dawned on my one evening as I closed up the workshop that there are times when a sudden power outage could leave me vulnerable when exiting the dark workshop. I had previously made my own flashlight with high intensity white LEDS for those frequent nigh-time excursions to the toilet. 

So I combined my CC/CV charging standard with a white LED driver gave me a box with plenty of light to safely exit.  I used some diode bridges so I can use any 8-30V mains derived AC or DC power source and a pass through design so lights can be daisy chained. I ended up installing three such lights across my storage bays and as soon as the power is lost the lights turn on for 10 minutes. Plenty of time in my workshop to turn everything off and exit safely. 

If the power comes back on within the 10 minutes they turn off and start recharging. Otherwise they turn off and wait with a mosfet isolating the battery from the current drain of the charging circuit when no power is present.

So in hot weather I simply turn off the power. A broad estimate is a current draw of 100mA from the battery when running. That's means even the rubbish cells I fitted will give over an hour of light knowing that the batteries will slightly discharge for 10 minutes before going into near total shutdown. The micro, being asleep most of the time, consumers very little current so I expect that the lights remain in a good to go state. 

In practice during summer I expect I will turn them on for perhaps two hours a week in the cool of the morning when I can "escape" to the shed while it is relatively comfortable to wield the soldering iron. 

The next version will have a switch fitted to allow for manually turning the light on and off during power outages when I may need a lantern. A shift to better batteries is also planned which will allow for both a longer run time and still brighter lighting. No point half-doing something.