James Van Artsdalen
Enlightened
Chargers use what's called a "negative delta v" algorithm to detect when the battery is charged. I'm not sure when this is measured but you're looking for a tiny twitch in voltage, about 0.005v. They charger switches to float and sets the green LED when this happens.
The problem is that there's some noise in the voltage and the "signal" may be seen amongst the noise, long before the battery is actually charged.
One solution is to ignore all "twitches" and charge away. This way you definitely don't stop early and definitely fill the battery, for the couple of cycles before the battery dies from over-charging (and as the capacity reduces from overcharge). Needless to say the prudent course of action is to stop early if in doubt.
The better solution is to look for a temperature rise at about the same time as the “twitch”. I’ve heard rumor of a Sanyo charger that does this but haven’t seen specs since Sanyo’s web site appears devoid of useful information.
Another thing to keep in mind is that the 1HU’s timeout probably prevents a D cell from completely charging from a fully discharged state: it takes more than 11.5 hours to fill an 8 AH battery at ~ 800 mA. Once again one solution might be to pump more current but that’s risky if the user has a 2200 mAH D cell (these do exist).
What I do is leave a set of batteries in the charger all the time and trade the run-down batteries for the charged set. Since the charged set has been trickle charging for a few days they’re fully topped off, even the last 5% you don’t get in a normal charge cycle.
If I notice one finish early I pop it out and back in on the assumption that it has a false negative delta v. New D cells I usually charge twice regardless on the assumption they didn’t really finish the first time.
The problem is that there's some noise in the voltage and the "signal" may be seen amongst the noise, long before the battery is actually charged.
One solution is to ignore all "twitches" and charge away. This way you definitely don't stop early and definitely fill the battery, for the couple of cycles before the battery dies from over-charging (and as the capacity reduces from overcharge). Needless to say the prudent course of action is to stop early if in doubt.
The better solution is to look for a temperature rise at about the same time as the “twitch”. I’ve heard rumor of a Sanyo charger that does this but haven’t seen specs since Sanyo’s web site appears devoid of useful information.
Another thing to keep in mind is that the 1HU’s timeout probably prevents a D cell from completely charging from a fully discharged state: it takes more than 11.5 hours to fill an 8 AH battery at ~ 800 mA. Once again one solution might be to pump more current but that’s risky if the user has a 2200 mAH D cell (these do exist).
What I do is leave a set of batteries in the charger all the time and trade the run-down batteries for the charged set. Since the charged set has been trickle charging for a few days they’re fully topped off, even the last 5% you don’t get in a normal charge cycle.
If I notice one finish early I pop it out and back in on the assumption that it has a false negative delta v. New D cells I usually charge twice regardless on the assumption they didn’t really finish the first time.
