Rockbox mail archive
Subject: Re: Re: ajbr20 behaviour with batteries
From: Dirk A. Vahlsing (diava_at_web.de)
> I don't really understand what goodness this patch offers. Can you elaborate
> on your thinking here?
> Daniel Stenberg -- http://rockbox.haxx.se/ -- http://daniel.haxx.se/
> This could be dependant on the batteries used (get some 1850s or 2000s, they
> are great!) the type of adapter you have (Kings or aAz or replacement?) and
> your operating system/USB drivers (do they allow the disk to spindown when
> Of course if one has never done that large a transfer or left the AJBR
> hooked up that long, one might never notice.
> this isn't my case. My AJBR haven't any hardware mod.0
> hhhmmmmm! then ... it's a hardware failure :-?
> I wonder if other archos users consider this an acceptable
> behaviour for their AJBRs.
We were using the unmodified AJBR20s as they were shipped by Archos:
- original 20GB HDD (Hitachi DK23CA-20)
- original batteries (1500mAh NiMH Noname)
- original charger (aAz: 9V_at_600mA)
- firmware Archos 1.27d or RockBox 1.4 (either booted from HDD)
on Win2k/XP USB1.1/2.0 machines (write cahching disabled).
We also ckecked this with HDDs upgraded to 60GB (Toshiba MK6021GAS).
What we noticed is that the archos batteries are discharged during long USB data transfer sessions even with the charger connected and charging.
So depending on the initial battery level the archos shut down/rebooted/turned off USB/ (unpredictable) after a certain time (from minutes to hours) due to discharged batteries. We removed the charger immediately after the "disconnect" happened and checked the battery voltage using RockBox. I can't remember the exact value but it prooved that the batteries were empty.
Since we currently don't want to spend time on boosting the internal charger (simply replacing the current sense resistor in best case or even modifying the whole charger circuit in worst case) I wrote the patch mentioned in my last mailing.
The patch isn't good for anything in that it doesn't solve the problem, but:
- it gives one the chance to watch/control the current battery level to be able to stop the device early enough so that an uncontrolled shutdown or whatever (with data loss or file system corruption) won't happen,
- it removes icons not really interesting from the status bar when connected via USB
- it allways shows the status bar when USB is connected, ignoring the status bar display setting.
Excerpt from the patch description (slightly extended for the sake of completeness)
> Changed blinking behavior of batt and charger:
> - batt icon always shows level, no charging indication
> - batt icon blinks on low batt level
> - charger icon blinks during charging
> - charger icon shown if charger connected but not charging or topoff/trickle charging
> Changed behavior of status line:
> - only batt/charger/clock shown in USB mode
> - status line always shown in USB mode
Daniel: Hope this clarifies what the patch is good for (at least for me and the guys around).
Josť: I think it's not a fault but a feature ;-) and we think it's not a 100% acceptable but we can live with it (having the patch & a future hw-mod to boost the charger in mind).
Chris: What charger are you using? Maybe you or anyone ales can check their recorders for this behaviour?
Werden Sie kreativ! Bei WEB.DE FreeMail heisst es jetzt nicht nur schreiben,
sondern auch gestalten. http://freemail.web.de/features/?mc=021142
Page was last modified "Jan 10 2012" The Rockbox Crew