maemo.org - Talk - View Single Post - Battery Tests, Homemade 3000mAh XL Battery & Q&A (POST #1)

	Estel	2011-07-20 , 08:16
	Posts: 5,028 \| Thanked: 8,613 times \| Joined on Mar 2011	#853

Originally Posted by shadowjk

Estel: interesing find that bme charges at 1050. I never saw it use anything except 950, but I investigated it during PR1.2 times or earlier.

Actually, I checked it now with stock Nokia wall charger, and, indeed, it uses 950 mA. Ho ever, when I use 2A 5V charger (from Virtual USB adapter, modified a little to contain USB charging port with crossed D+/D-), it uses 1050 mA for me. Using even more powerful power source (ATX power supply @ 5V) doesn't result in further increase of current, obviously.

Originally Posted by shadowjk

btw, I expect charge.sh/charge21.sh and bq27200.sh would not work if the power kernel battery module is loaded, as they try access the same hardware and conflict.

Equally, trying to restart bme with power kernel battery module loaded would fail. Both bme and charge*sh access charger chip intermittently and, depending on chance, might actually run both at the same time, but it would be messy to have two things trying to run the charger at the same time.

Strange, cause all of the time I was using it with bq27x00 module loaded (i know, shame on me...), and bq27200.sh works without even smallest flaws both in normal and loop mode. It works even with module + script + bme charging combination.

Also, some time ago - when i didn't know about possible conflict - I was notoriously testing charging via charger.sh without disabling bme, and never noticed any side effects. I mean, without looking into terminal, it was just like bme is charging - You know, battery indicator going all way to up in loop, and communicates about charging when plugged popping up...

Ho ever, charge.sh output was also correct, and when I try it now with charger21.sh I can't find any errors in log.

Also, with kernel module loaded, and even with both kernel module + bq27200.sh working in loop, stopping and starting bme works for me flawlessly, if I'm *not* around ~3200 mV

I'm sure joerg_rw got really good rationale to warn us about possible conflicts with accessing same thing by bme, but i must admin that I have yet to see any bad effect of that, or anyone claimed problems with bq27x00 kernel module. And then, confirm that it is really it - except for inaccurate reading due to wrong calibration. AFAIK, every report about conflict (in kp thread) concluded in discovering different source of problems.

So I'm start to assuming - with all due care and respect to fact, that I may be totally wrong - that there is something, about how it does work, that we don't know about, despite high amount of research done already.

// Edit

Of course my problem with charge*.sh not charging, when started under 3248 mV and bme *disabled* still persist, which make it even more strange.

__________________
N900's aluminum backcover / body replacement
-
N900's HDMI-Out
-
Camera cover MOD
-
Measure battery's real capacity on-device
-
TrueCrypt 7.1 | ereswap | bnf
-
Hardware's mods research is costly. To support my work, please consider donating. Thank You!

Last edited by Estel; 2011-07-20 at 08:18.

Quote & Reply |

The Following User Says Thank You to Estel For This Useful Post:
knuthf