Send buglog mailing list submissions to
[email protected]
To subscribe or unsubscribe via the World Wide Web, visit
http://lists.openmoko.org/mailman/listinfo/buglog
or, via email, send a message with subject or body 'help' to
[email protected]
You can reach the person managing the list at
[email protected]
When replying, please edit your Subject line so it is more specific
than "Re: Contents of buglog digest..."
Today's Topics:
1. Re: Openmoko Bug #1158: Charging stops even while connected
(Openmoko Public Trac)
2. Re: Openmoko Bug #1158: Charging stops even while connected
(Openmoko Public Trac)
3. Re: Openmoko Bug #1158: Charging stops even while connected
(Openmoko Public Trac)
4. Re: Openmoko Bug #1158: Charging stops even while connected
(Openmoko Public Trac)
5. Re: Openmoko Bug #1158: Charging stops even while connected
(Openmoko Public Trac)
6. Re: Openmoko Bug #1158: Charging stops even while connected
(Openmoko Public Trac)
7. Re: Openmoko Bug #1158: Charging stops even while connected
(Openmoko Public Trac)
--- Begin Message ---
#1158: Charging stops even while connected
---------------------------------+------------------------------------------
Reporter: p...@… | Owner: openmoko-devel
Type: defect | Status: reopened
Priority: high | Milestone: Om2008.10
Component: System Software | Version: unspecified
Severity: blocker | Resolution:
Keywords: power management | Haspatch: 0
Blockedby: | Estimated:
Patchreview: | Blocking:
Reproducible: sometimes |
---------------------------------+------------------------------------------
Comment(by balajirrao):
Replying to [comment:47 lindi]:
> Replying to [comment:46 andy]:
> > > 2008-12-16T21:28:29+0200 1229455709 Battery Not_charging 4163000
10312 1003348 293 Li-ion 1 367860 3932100 100 1 0.21
> >
> > Not really sure that's evil. The GSM unit is connected directly to
battery... the charger rightly decided the battery was full and stopped
charging, then the GSM unit is eating 10mA direct from battery. I guess
after some interval the charger will see if it should charge again.
>
> It should charge yes but it does not seem to since the capacity starts
to go down:
>
> {{{
> 2008-12-16T21:30:30+0200 1229455830 Battery Not_charging 4160000 12375
1003348 293 Li-ion 1 311220 3932100 100 1 0.15
> 2008-12-16T21:32:31+0200 1229455951 Battery Not_charging 4158000 16687
1003348 293 Li-ion 1 227160 3932100 100 1 0.13
> 2008-12-16T21:34:32+0200 1229456072 Battery Not_charging 4155000 58875
1003348 293 Li-ion 1 64320 3932100 100 1 0.19
> 2008-12-16T21:36:34+0200 1229456194 Battery Not_charging 4158000 11625
1003348 293 Li-ion 1 388380 3932100 100 1 0.18
> 2008-12-16T21:38:35+0200 1229456315 Battery Not_charging 4158000 10312
1003348 293 Li-ion 1 367020 3932100 100 1 0.11
> 2008-12-16T21:40:36+0200 1229456436 Battery Not_charging 4158000 20250
1003348 293 Li-ion 1 186840 3932100 100 1 0.08
> 2008-12-16T21:42:37+0200 1229456557 Battery Not_charging 4158000 21937
1003348 293 Li-ion 1 172380 3932100 100 1 0.05
> 2008-12-16T21:44:39+0200 1229456679 Battery Not_charging 4158000 10312
1003348 293 Li-ion 1 366540 3932100 100 1 0.03
> 2008-12-16T21:46:40+0200 1229456800 Battery Not_charging 4155000 20250
1003348 293 Li-ion 1 186600 3932100 100 1 0.09
> 2008-12-16T21:48:41+0200 1229456921 Battery Not_charging 4155000 20625
1003348 293 Li-ion 1 183120 3932100 100 1 0.06
> 2008-12-16T21:50:42+0200 1229457042 Battery Not_charging 4155000 10312
1003348 293 Li-ion 1 366180 3932100 100 1 0.03
> 2008-12-16T21:52:44+0200 1229457164 Battery Not_charging 4155000 9937
1003348 293 Li-ion 1 291720 3932100 99 1 0.05
> 2008-12-16T21:54:45+0200 1229457285 Battery Not_charging 4155000 15375
1003348 296 Li-ion 1 245400 3932100 99 1 0.47
> 2008-12-16T21:56:47+0200 1229457407 Battery Not_charging 4155000 10312
1003348 296 Li-ion 1 365700 3932100 99 1 0.71
> 2008-12-16T21:58:48+0200 1229457528 Battery Not_charging 4152000 13875
1003348 298 Li-ion 1 358980 3932100 99 1 0.82
> 2008-12-16T22:00:50+0200 1229457650 Battery Not_charging 4152000 10312
1003348 298 Li-ion 1 365340 3932100 99 1 0.93
> 2008-12-16T22:02:52+0200 1229457772 Battery Not_charging 4152000 19500
1003348 301 Li-ion 1 193140 3932100 99 1 1.03
> 2008-12-16T22:04:53+0200 1229457893 Battery Not_charging 4144000 43500
1003348 303 Li-ion 1 189180 3932100 99 1 0.74
> 2008-12-16T22:06:55+0200 1229458015 Battery Not_charging 4121000 188437
1003348 313 Li-ion 1 19860 3932100 99 1 0.53
> 2008-12-16T22:08:56+0200 1229458136 Battery Not_charging 4118000 189937
1003348 316 Li-ion 1 19620 3932100 98 1 0.45
> 2008-12-16T22:10:57+0200 1229458257 Battery Not_charging 4110000 206062
1003348 321 Li-ion 1 18060 3932100 98 1 0.33
> }}}
>
>
> > > 00 07 7b e7 28 19 ff ff 00 02 00 33 2a f0 19 00
> >
> > Charger is enabled here.
>
> I get exactly the same line from dump_regs even now that capacity is
> already at 97%. Why is battery capacity dropping even if charger is
> enabled?
>
autores = 1 from the dump => supposed to restart charging at 96% of
vbatmax. vbatmax (reg 0x44) is defined as 4.0 V. So, it expects vbat to
come to 3.84 V - which might never happen ? Seeing the specs - table 93 -
we can increase the value of vbatmax upto 4.3 V ? Can it be useful here ?
--
Ticket URL: <https://docs.openmoko.org/trac/ticket/1158#comment:48>
docs.openmoko.org <http://docs.openmoko.org/trac/>
openmoko trac
--- End Message ---
--- Begin Message ---
#1158: Charging stops even while connected
---------------------------------+------------------------------------------
Reporter: p...@… | Owner: openmoko-devel
Type: defect | Status: reopened
Priority: high | Milestone: Om2008.10
Component: System Software | Version: unspecified
Severity: blocker | Resolution:
Keywords: power management | Haspatch: 0
Blockedby: | Estimated:
Patchreview: | Blocking:
Reproducible: sometimes |
---------------------------------+------------------------------------------
Comment(by andy):
lindi charger is quite complex state machine, in which it can do various
steps like wait or fast-charge and so on. If the charger is disabled, it
would turn all of it off and be certain it would never charge. When the
charger is enabled, it just gives it opportunity to charge depending on
its "program".
Balaji, yes it seems Vth(res) is too low. It's not that it would "never
happen" to reach 3.84V but it will require great deal of discharge
pointlessly.
We can either disable autostop so BATFULL just continues to be in constant
current charge mode (not good for battery for permanently tethered device
I guess), or set MBCC2 vmax (is it really this one that sets Vth(res)? I
didn't see it actually say it) in mach_gta02.c regulator init stuff.
--
Ticket URL: <https://docs.openmoko.org/trac/ticket/1158#comment:49>
docs.openmoko.org <http://docs.openmoko.org/trac/>
openmoko trac
--- End Message ---
--- Begin Message ---
#1158: Charging stops even while connected
---------------------------------+------------------------------------------
Reporter: p...@… | Owner: openmoko-devel
Type: defect | Status: reopened
Priority: high | Milestone: Om2008.10
Component: System Software | Version: unspecified
Severity: blocker | Resolution:
Keywords: power management | Haspatch: 0
Blockedby: | Estimated:
Patchreview: | Blocking:
Reproducible: sometimes |
---------------------------------+------------------------------------------
Comment(by balajirrao):
Replying to [comment:49 andy]:
> lindi charger is quite complex state machine, in which it can do various
steps like wait or fast-charge and so on. If the charger is disabled, it
would turn all of it off and be certain it would never charge. When the
charger is enabled, it just gives it opportunity to charge depending on
its "program".
>
> Balaji, yes it seems Vth(res) is too low. It's not that it would "never
happen" to reach 3.84V but it will require great deal of discharge
pointlessly.
>
> We can either disable autostop so BATFULL just continues to be in
constant current charge mode (not good for battery for permanently
tethered device I guess), or set MBCC2 vmax (is it really this one that
sets Vth(res)? I didn't see it actually say it) in mach_gta02.c regulator
init stuff.
Yes, its MBCC2. Will set it to around (4.2 ~ 4.3) volts. Will be the first
thing I'll try tomorrow morning. Hopefully it'll work..
--
Ticket URL: <https://docs.openmoko.org/trac/ticket/1158#comment:50>
docs.openmoko.org <http://docs.openmoko.org/trac/>
openmoko trac
--- End Message ---
--- Begin Message ---
#1158: Charging stops even while connected
---------------------------------+------------------------------------------
Reporter: p...@… | Owner: openmoko-devel
Type: defect | Status: reopened
Priority: high | Milestone: Om2008.10
Component: System Software | Version: unspecified
Severity: blocker | Resolution:
Keywords: power management | Haspatch: 0
Blockedby: | Estimated:
Patchreview: | Blocking:
Reproducible: sometimes |
---------------------------------+------------------------------------------
Comment(by lindi):
Do I understand it correctly: Since GSM chip is somehow directly connected
to the battery there is no way to have battery sit idle if the phone is
registered to GSM network and connected to wall charger? Do you have any
idea if constant charging/discharging of the battery reduces battery
lifetime significantly?
--
Ticket URL: <https://docs.openmoko.org/trac/ticket/1158#comment:51>
docs.openmoko.org <http://docs.openmoko.org/trac/>
openmoko trac
--- End Message ---
--- Begin Message ---
#1158: Charging stops even while connected
---------------------------------+------------------------------------------
Reporter: p...@… | Owner: openmoko-devel
Type: defect | Status: reopened
Priority: high | Milestone: Om2008.10
Component: System Software | Version: unspecified
Severity: blocker | Resolution:
Keywords: power management | Haspatch: 0
Blockedby: | Estimated:
Patchreview: | Blocking:
Reproducible: sometimes |
---------------------------------+------------------------------------------
Comment(by werner):
But is it really at 4.0V ? I looked at the sources of u-boot, Qi, and the
kernel, and they all agree that they set it to 4.2V and/or never change it
from there. I decoded the dump above and it says
44 MBCC2 = 0x28 -- MBC charger control 2
vbatcond
00 (0x0): 2_7V
vmax -- Vbat(float)prog
1010 (0xa): 4000+20*10 = 4200 mV
vresdebtime -- debounce time for Vth(RES)
0 (0x0): 32s
The tool for it is something I wrote a long time ago. It's in
http://svn.openmoko.org/developers/werner/greg/
To decode a raw dump of whitespace-separated hex bytes starting at
register 0 (like the one included above), run
./ds.pl
and paste the raw dump.
--
Ticket URL: <https://docs.openmoko.org/trac/ticket/1158#comment:52>
docs.openmoko.org <http://docs.openmoko.org/trac/>
openmoko trac
--- End Message ---
--- Begin Message ---
#1158: Charging stops even while connected
---------------------------------+------------------------------------------
Reporter: p...@… | Owner: openmoko-devel
Type: defect | Status: reopened
Priority: high | Milestone: Om2008.10
Component: System Software | Version: unspecified
Severity: blocker | Resolution:
Keywords: power management | Haspatch: 0
Blockedby: | Estimated:
Patchreview: | Blocking:
Reproducible: sometimes |
---------------------------------+------------------------------------------
Comment(by sdhillon):
I'm not sure if this actually helps, but I am running this in crontab:
* * * * * root echo 500 > /sys/devices/platform/s3c2440-i2c/i2c-
adapter/i2c-0/0-0073/force_usb_limit_dangerous
* * * * * root echo -n "fast_cccv" >
/sys/devices/platform/s3c2440-i2c/i2c-adapter/i2c-0/0-0073/chgmode
and it seems to cause the charger to kick back in when it stops.
--
Ticket URL: <https://docs.openmoko.org/trac/ticket/1158#comment:53>
docs.openmoko.org <http://docs.openmoko.org/trac/>
openmoko trac
--- End Message ---
--- Begin Message ---
#1158: Charging stops even while connected
---------------------------------+------------------------------------------
Reporter: p...@… | Owner: openmoko-devel
Type: defect | Status: reopened
Priority: high | Milestone: Om2008.10
Component: System Software | Version: unspecified
Severity: blocker | Resolution:
Keywords: power management | Haspatch: 0
Blockedby: | Estimated:
Patchreview: | Blocking:
Reproducible: sometimes |
---------------------------------+------------------------------------------
Comment(by andy):
Yeah Balaji Werner's right about the voltage, that value is shifted up a
couple of bits.
It is still worth trying increasing this thing though because the voltage
over discharge of the battery stays up for a long time and only decays
slowly until you get near the end.
--
Ticket URL: <https://docs.openmoko.org/trac/ticket/1158#comment:54>
docs.openmoko.org <http://docs.openmoko.org/trac/>
openmoko trac
--- End Message ---
_______________________________________________
buglog mailing list
[email protected]
http://lists.openmoko.org/mailman/listinfo/buglog
