Pandora charging issues


I recently upgraded to a 1GHz Pandora (fantastic machine btw) and although it's been charging without too many issues, today it really doesn't want to play ball. I've switched batteries and tried via AC and USB but nothing seems to work.

When I plug in AC, it says:


twl4030_bci twl4030_bci: spurious CHG_PRES irqs detected (7), disabling charger


And a grep for this type of thing gives:


dmesg | grep CHG_PRES | cut -f2- -d ']' | sort -u
twl4030_bci twl4030_bci: spurious CHG_PRES irqs detected (5), disabling charger
twl4030_bci twl4030_bci: spurious CHG_PRES irqs detected (6), disabling charger
twl4030_bci twl4030_bci: spurious CHG_PRES irqs detected (7), disabling charger


Does anyone have any idea what this might be?
 
Wow. I have never seen that particular error. I'll have to check the driver, see if there's any indication in there as to what it means.

In the meantime there's some simple stuff we can try. You've obviously tried turning it off and on again, given the battery switching. Have you tried removing the battery and just letting it sit for about 30 minutes to give the capacitors time to drain? It's possible the charge chip has gotten into some weird state and just needs a good reset which will happen as soon as any stray power is used up. Also try running it on just AC power for a while: turn it off, remove the battery, then plug in the AC charger. Doesn't always work but doesn't hurt, just make sure you turn it off and unplug the charger before putting the battery back in.
 
Thanks, got this as well btw:


[ 6538.400634] twl4030_bci twl4030_bci: battery temperature out of range

I've seen this for both batteries so doesn't look related to the battery itself.

Just tried to run on AC only but it doesn't do anything so I suppose it needs the battery to be in.

As per your suggestion, I'll leave it off for a little bit.
 
Ok, leaving it off didn't help (same message) however, switching to another power supply (an old one from Craig's shop which I still had lying around) did the trick.

I even got it to charge off USB when switching to another USB charger as well.... I guess the 1GHz model (at least mine) is a bit pickier than the classic Pandora. I checked the old power supply against the new one and they were both 5V / 2A so not sure why one worked and the other didn't.

Lucky that I ordered two back in 2009 :D
 
The connector on the charger that didn't work... was it centre positive or centre negative?  The pandora needs a centre positive connection.  (though I would normally expect a reversed polarity to result in magic smoke escaping)

- Neelix
 
The first error, "spurious CHG_PRES", occurs when there are a lot of charge request interrupts in a very short period of time. Basically if it looks like the charger has been plugged in and unplugged over and over a bunch of times in less than a second then it simply disables the charger. So it's getting hot, and rapidly connecting/disconnecting: definitely sounds like a faulty charger, which it looks like you already figured out. :)

The errors in question, however, simply disable the AC charger: the USB charge should have still worked, but maybe that was a different problem.

Glad you had a spare charger. I should have thought of suggesting that, sorry.
 
The adapter was an official OpenPandora one which I received with the 1GHz model so the connector should be correct. (In fact, it worked a few times before it packed in.)

Thanks for looking into the error, WizardStan!
 
 
@Ben: Has this been resolved already?

I have similair problems, that I cannot charge using the AC port. I've got two official AC-chargers, no difference there. I've checked the DC voltage of each of the adapters using a multimeter, they output 5V as expected.

I've seen similair error messages reported in dmesg. Both:

- battery temperature out of range

- spurious CHG_PRES irqs detected

My only way to charge the battery is using the USB port. Which is a pain:

- I cannot use the official AC-chargers bought for this purpose.

- cannot charge the batteries when these are completely drained, I need to boot into linux first, such that the software can enable the USB charger

:(

help!
 
If it thinks that battery temperature is out of range and it's not very hot or cold in your environment, maybe the thermistor pin (middle pin of 3 pins on the battery) is not properly connected. Try reseating the battery.
 
Last edited by a moderator:
Shouldn't both USB an AC react the same if the temperature-readout or battery was the problem?
 
I've had a few Chargers with poorly designed barrel connectors it seemed with the Pandora and even other devices it would only connect properly if I rotated the connector in specific angles.. Which you wouldn't think would happen but it does.. I have a Radio Shack brand charger I bought for the Pandora I ended up moving it to my powered USB 2.0 Hub which is stationary on my TV setup.. I've seen this issue on the charger I got from the first batch Pandora... That charger had some other issues too..
 
Last edited by a moderator:
Since yesterday, I am a Pandora owner, again.

Unfortunatly, the device does not charge without problems. I am sure (not a 100%), it charged yesterday after I turned the device on. But today, I can't charge it. The led does not light up.

I currently can only charge the device via USB, but thats slow as hell.

I don't own any device to check the charger. Are there common issues with the Pandora? Any ideas, what I can check?
 
Well like I said in a prior post, some barrel jacks are fiddly with the Pandora. If you rotate the jack does it sometimes bring up the charging light? Also I would suggest finding someone to check the charger, I know the one provided on the first batch pandora's by Craig's shop were sketchy and poorly put together. The one EvilDragon sourced after he took over were a lot better quality. That said a charger made for the PSP would work with the Pandora if you just want to find something else to try.

Are you on the latest and greatest Firmware image(SZ 1.76)? I know there has been some changes with charging software a few major patches ago.
 
Firstly, check the battery is making good contact. In my case my battery compartment is stuffed with little wads of folded paper to ensure the battery retains in contact in all situations, and I expect most people need that little foam pads or no. Your symptoms perhaps seems to agree with the sense pin (the middle one) not making good contact, so the system will still run but the battery controller will be utterly perplexed.

One way to test that is to cat various files under (from memory) /sys/class/power_supply/bq_0??/ (hex number at the end there) and in that folder there will be files called things like charge_full, current_now and things like that. Or, perhaps simpler, you could just download this: http://repo.openpandora.org/?page=detail&app=sysinfo.wb, install and launch it and it'll show you all that in graphical form. It should be able to tell you things like the battery full charge when switched to the battery panel (use shoulder buttons) and the time to empty, and you can observe if anything happens if you plug in a charger. It's a very handy tool.
 
I mostly had issues about charging because of high temperatures reached by the battery... you can check if it's the case using "dmesg"
 
I mostly had issues about charging because of high temperatures reached by the battery... you can check if it's the case using "dmesg"

That could also be simulated using a bad sense pin connection I think; lithum battery thermistor protection is usually positive, so if the system detects a high resistance it may assume the thermistor's indicating an over-temperature.
 
I sometimes have problems charging when my battery is completely drained. The battery needs to have some power left to power up the charging citcuit or something and it seems to trickle in very slowly, thus not charging straight away when plugged in.
Usually I can just leave it plugged in for a while and the charging light will come on after an hour or so.

I try to not completely drain my battery to prevent this from happening.
 
Yes, if you let it charge down below a specific voltage, it won't have enough power to run the CPU, but the charge chip is simpler and can work on a lower differential. The system is set up so that the CPU need so run and boot the kernel and probably init linux before the charge LED can come on, because it's controlled via a script I think. Even if the battery is good, if you wake a unit by plugging it in, the power light will come on before the charge light. Note, you can't discharge the battery this low by using it - you need to have not used the system for long enough for battery self-discharge to take it that low.
 
Back
Top