After a failed attempt to revive an SLA with a shorted cell:
http://fieldlines.com/board/index.php/topic,144491.0.html... I decided to try it on a slightly healthier subject, and so far the results are promising.
This one is a 12V 5AH SLA that had lost some capacity (presumably from being on float for extended periods of time) and just wasn't performing up to snuff when put to the test.
I more or less was aware of it's entire service 'career', and this is what made it so appealing to try again with the watering idea. It had never been abused in the sense of multiple C discharge rates or left for dead or anything, it just floated for too long and gave up too much water to hit full capacity anymore.
The fun part is dealing with striation that happens as a result of adding water to the cells; since the electrolyte isn't freely floating around inside, I got creative with my mind's eye and ran where it took me.
These are the steps I took to revive it:
In as much of a charged state as I could get it, I watered it (very slowly, drop by drop) until I could just barely see a layer of 'flowing' water above the plates/separators. This took several passes, just going around to the next cell as the last one soaked in. Keeping air from being trapped was a challenge also since the hole for the valve boot is so small. I used an old irrigation syringe I got from the oral surgeon when my son had his wisdom teeth pulled. Works like a champ. Just gotta be patient.
Once topped up with water, let it sit overnight on a very light floating charge, the idea being to just stir it a little. I didn't want to over-do it, concerned about oxidation of the plates.
Then a nice slow discharge at about the C/20 rate, in this case into a buck converter driving an LED. I didn't get all scientific about plotting the current draw or anything, because at this point I was not super hopeful, just curious if this was plausible or not. Took it down to 12.00V and disconnected it.
After the slow discharge, ran a slow charge, somewhere in the C/30 to C/25 range, and let it come all the way up to 15V. This took a while obviously, and once it hit 15V, I held it there for another 24 hours. Wanted to make sure I got an even better stir.
Next, same as before, slow discharge at C/20, into the buck/LED. Ran almost twice as long this time around before hitting the 12V mark.
Feeling froggy, I took it all the way down to 10.5V, disconnected it and let it sit for an hour or so, then began the ultra slow charge. C/50'ish.
She's been on charge now for a couple days, and is grabbing ahold very well, and every once and a while I do a rudimentary check for impedance, looks promising. Current terminal voltage ATM is 12.55, with charge current something like 100mA or so. I had to bump it up a notch as it got over 12V, since the voltage difference between the charger and the battery was diminishing. This just involved changing a ballast resistance to keep it around 100mA. Might have to do it again in a couple days. Will see.
Next time I do one, I'll go quite a bit more scientific (plots, graphs, etc) but I was on more of a 'ok I wanna know if this is really feasible or not', and not so concerned with the details of how much gain I got from it.
More to come as it progresses...
Steve
The idea here was simple, really. Coax the dried out portions of the plates into giving up any sulfate into the electrolyte, since there's no real way to 'stir' an SLA.