Battery Drain

[tsh]

I think it would be a good idea to have a clear indication that the battery is fully charged. Aren't the LEDs capable of changing colour? The Nokia N900 charges with a slowly blinking yellow light, when finished it changes to stable green (which switches off when you remove the charger cable).

The LEDs are single colour. It would be possible to blink an LED when charged (since there is a driver to do this when below a threshold).


http://git.openpandora.org/cgi-bin/gitweb.cgi?p=pandora-libraries.git;a=blob;f=apps/pndevmapperd.c;hb=HEAD

 

[WizardStan]

If you're curious, I noticed that it seems to depend on how hot the unit has gotten

Interesting. Wrapping it in a towel, I have now caused it to stop charging twice consecutively; it wouldn't start charging again without unplugging and plugging back in. No towel, it charges with no problem.


It seems to me that overheating and not reaching 100% is some kind of stop-charging trigger. It might just happen that when it hits 100% more heat is generated by trickle charging or something.

 

[Prometheus]

I'm glad that, for once, it's not just me.


I wonder if this accounts for any of the cases folks have reported, where they plug the unit in at night to charge, but find in the morning that the battery has somehow run itself down?


EDIT: While I think of it, I notice that my battery loses its first ten percent very quickly (i.e., within a matter of a few hours), but otherwise only loses 1% or less per night (the longest I've left mine is about two days, and it only lost 1% during that time, too). I don't know if it is related to some sort of misreading of the heat and/or charge level (the battery is definitely seated right, of course, and obviously it loses barely anything otherwise), but it might be worth a mention.

 

[Tor]

EDIT: While I think of it, I notice that my battery loses its first ten percent very quickly (i.e., within a matter of a few hours),[..]

If everything else is as it should then a quick initial drain down from 100% would normally be simply because it wasn't fully charged when removed from the charger (as mentioned earlier it'll go to 100% during charging quite quickly, but will still need time to complete the charging. If there's no external indicator to show when it's full it will be difficult to know when the charger can be removed).


Without an indicator the only way to be "safe" is to leave it on the charger for hours, possibly with some calculation. The N800 with a 1500mAh battery and charging with up to 800 mA (at max) will take 2-3 hours, from a reasonably depleted battery.


If you charged overnight, for example, and you still see that 'quick early drain' effect, then there's something else going on.

 

[Prometheus]

I never charge overnight, so I won't be able to observe anything as far as that goes.


As far as getting to 100% quickly, that is very interesting... So, even though the meter ticks up over the course of a number of hours, and eventually reads 100%, it actually isn't?

 

[Tor]

As far as getting to 100% quickly, that is very interesting... So, even though the meter ticks up over the course of a number of hours, and eventually reads 100%, it actually isn't?

It's not at 100% when it reaches 100% initially, no. The reason is that the percentage is measured as a percentage of what's defined to be 100% voltage when not charging. To use a device I own as a reference, that could be, say, 4.20 volts. So, if the battery is at 4.20 volts when not connected to a charger the meter shows 100%. However, to get to that level you have to charge with a higher voltage (so that there's a potential difference to create a current flow). So this device I have will at some point up its charge level to 4.20 volts, and it shows 100%. But current is still flowing into the battery, so if I remove the charger at that point it will soon drop a bit, indicating that it's not full, and when I run something that needs power it'll drop quickly.


So in short, the indicator you see will have to stay at 100% for a while until there's no more current flowing into the battery, and only then will it be fully charged. (The charging circuitry takes care of all this, it constantly adjusts its charging voltage to provide a reasonable charging current, at the same time not going above a maximum voltage setting).

 

[Prometheus]

Aha. Thankyou - that explains it. I had unplugged it as soon as I saw it tick to 100%. :lol:

 

[Korlithiel]

That's somewhat interesting on why it stops charging during the night, though unfortunately not something I can really deal with. I have to put my Pandora top down to charge at night to cover the lights, almost never have my heater on, and it's still rather consistently running down a fair percentage of the battery during the night.


I imagine this problem will go away as it gets colder out for a fair while, but I'm afraid next summer it would be rather problematic not being able to charge for more then a short bit of time without unplugging and plugging it back in.

 

[Prometheus]

^ If it's of any interest, my observation has been that hot weather has no effect - it was just whether or not I had it in its bag whilst charging. I don't know for sure if that's why they sometimes stop charging overnight, I must stress.

 

[Korlithiel]

Prometheus, I know there are no certainties on something so peculiar right now. I just hope it's a heat thing since that means it will stop bothering me in the next several stressful months.

 

[Prometheus]

I have a question for WizardStan and Korlithiel, if I may.


Do you guys happen to have the PowerPax chargers (the one with the figure-of-eight connector on one side), by any chance? Or do you have the Flypower kind (the one with the Europlug and the adapter for your own country) instead?


I ask as I made an interesting discovery. See, I've ended up with one of each type of charger, and I've found that my Pandora gets considerably hotter when using the PowerPax charger than it does when using the Flypower one.


I mentioned earlier that keeping my Pandora in its bag while charging is a no-go, but this is only the case with the PowerPax charger. Also, the quick dropping of 10% of the charge after the meter reads 100% only seems to occur with the PowerPax charger. With the PowerPax one, I must leave my Pandora outside of its bag, otherwise it will get too hot and stop charging before it reaches 100% - accordingly, trickle-charging will only happen when the Pandora isn't in its bag, for the same reason.


Charging and trickle-charging happens completely normally with the Flypower charger, and the Pandora does not get anywhere near as hot with that one. When using the Flypower unit, my Pandora can be left in its bag to charge. When the Pandora gets to 100% with this charger, it really is at 100%, and can be unplugged shortly afterwards without losing any percentage other than the normal 1% overnight.


So, could either of you confirm what type of charger you both have, please? I'd like to know if I'm onto something here.

 

[WizardStan]

Do you guys happen to have the PowerPax chargers (the one with the figure-of-eight connector on one side), by any chance? Or do you have the Flypower kind (the one with the Europlug and the adapter for your own country) instead?

I've got the Flypower kind but I can't use it because it shipped with the wrong adapter. Instead I use a PSP charger.


Interesting about the PowerPax charger there. If I were to guess based only on what you've described, I would say the PowerPax is supplying more than the rated 5 volts. That would explain why it gets hotter, and why the charge percentage reaches 100% before the battery is full (as Tor explained about the way chargers over charge things)


You wouldn't happen to have a voltmeter you could test them with, would you?


I'm almost completely convinced this is a heat thing. Once it reaches 100%, if the charger is providing too much voltage, that's just more energy that need to be converted to heat during trickle charge. I had to insulate my Pandora pretty well to get it to stop charging, and thinking on it now, it only happened the first time to me because I had accidentally covered it with a spare blanket while it charged overnight.


I think a useful poll might be for those having these charge problems which type of charger they have. I wouldn't be surprised if it was entirely PowerPax.

 

[Prometheus]

Sorry to say that I don't have a voltmeter, no...


Also, I agree, I wouldn't be surprised if all of the folks with that charging issue are those who have the PowerPax charger.

 

[Twiitcher]

I have a PowerPax charger, no problem charging to 100%. just the odd 1% battery drain per hour when system completely off thing.

 

[WizardStan]

I have a PowerPax charger, no problem charging to 100%. just the odd 1% battery drain per hour when system completely off thing.

I doubt all PowerPax will exhibit the overheat issue, but I'm willing to bet all the ones experiencing it have PowerPax chargers. That's based entirely on Prometheus's single example, so the statistics may be slightly skewed.

 

[Twiitcher]

New news from my 1% hourly battery drain issue, I finally got minimenu working - (was an issue I didn't talk about, dev version of fennec pnd was corrupt or something, regardless, I removed the pnd and voila minimenu worked) - and I powered off this AM from minimenu.. 2 hours later I powered back up and low and behold, still at 100% Battery. I will do lengthier tests over the weekend and hopefully get more definative results.


ATM is seems that the shutdown from the Full Desktop OS may not be truly shutting down completely, but minimenu appears to be shutting down properly.


An overnight test resulted in the same 1% battery loss every hour regardless of which OS I shut down from, the above 100% battery after 2 hours off from mini-menu was inaccurate... minimenu took time to update battery status and showed the 1% loss per hour as always..


So issue is not OS related, something is draining 1% of the battery life even when the unit is turned off completely.

 

[Prometheus]

Twiitcher, I'm just curious, here: What format do you use for your SD Cards?


This may seem unrelated, but I observed a little while ago that, on my unit, I couldn't shut down properly from Xfce if mine were ext2 (it would just hang on the blinking cursor, but would never actually power off). I wonder if this is related?

 

[Twiitcher]

Twiitcher, I'm just curious, here: What format do you use for your SD Cards?


This may seem unrelated, but I observed a little while ago that, on my unit, I couldn't shut down properly from Xfce if mine were ext2 (it would just hang on the blinking cursor, but would never actually power off). I wonder if this is related?

I format them using Panasonic SD Formatter to fat32. It appears to shut down completely, from both UI's - i.e. the blinking cursor goes away, the screen goes dark, all leds go out...

 

[Prometheus]

Ah, ok. I guess it's not related, after all. Thanks.

 

[Tor]

I concur with WizardStan, the PowerPax heat issue sounds like a voltage problem (i.e. it provides more than 5V, possibly much more). I would get it checked with a voltmeter, and, if possible, a scope to check if the voltage is clean. (I checked a bunch of chargers before I connected anything to my BeagleBoard, and only a few are actually providing stable, clean, regulated 5V (or 5.1V which is more common). Just for the record: I'm an electronics engineer and I've got access to equipment, fortunately.)