[Gta04-owner] problems with battery loading

H. Nikolaus Schaller hns at goldelico.com
Fri May 27 12:13:54 CEST 2016


> Am 26.05.2016 um 23:33 schrieb <t.munker at gmx.net> <t.munker at gmx.net>:
> Hi,
> thanks for your answer so far!
> I still have some questions:
> Am Thu, 26 May 2016 23:07:29 +0200
> schrieb "H. Nikolaus Schaller" <hns at goldelico.com>:
>> Hi,
>>> Am 26.05.2016 um 20:14 schrieb <t.munker at gmx.net>
>>> <t.munker at gmx.net>:
>>> Hi,
>>> i've got some problem with loading my battery.
>>> Symptoms are:
>>> When plugging usb line in at the phone, phone wakes up and starts
>>> loading, if there is power.  
>> Even without the cable connected on the other end?
> The phone even wakes up without the other cable end connected. But does
> not start loading, since there is no power...

That is really strange. Wake-up means that there is some power detected on
the VBUS line. But how connecting just a cable with open other end can do this
is strange.

>>> Plugging in at the other end of the usb line is not detected (when
>>> first plugged in at phone, phone wakes up, but can't load since
>>> there is no power and goes sleeping again)
>>> After some seconds (mostly) (sometimes even about half an hour) it
>>> stops loading, even when battery capacity is still critical.  
>> Hm. Looks like broken cable.
> Since i used different cables, i would suspect the connector.
>>> Old openmoko power adapter doesn't seem to work at all.
>>> To get the phone usable i need to watch it closely, and unplug and
>>> replug it, if it stops loading.
>>> Environment:
>>> I use QTMoko. Since this has some lag until plugging is detected, i
>>> use neocontrol to view power status.
>>> I resoldered my usb-port, maybe something went wrong there. But
>>> right after resoldering it, it was working good and now got slowly
>>> worse... Also the sd-card got already corrupted at some point, i
>>> will reinstall next days to exclude a software error
>>> Where i need help:
>>> Is there some way to determine why the phone stops loading?  
>> I don't know about tools but latest QtMoko still uses the old 3.7
>> kernel which does not have a good charger driver. The main problem is
>> that it simply stops charging as soon as the USB voltage inside the
>> GTA04 drops below 4.5V. And does not restart unless you unplug +
>> replug the cable.
>> This may have a reason in a weak cable or weak power supply,
>> especially with almost empty battery when the charging current going
>> through the cable (in parallel to operating the GTA04) is highest.
>> This may be ~800 mA...
>>> When i want to check my usb-port, is this possible with a simple
>>> ohm-/volt-/ampere-meter? What are the readings that i should expect
>>> between the usb-pins when connected correctly to the board?  
>> A common source of trouble is if the GND or the shield is not really
>> connected. This may happen during repair. So you should check if you
>> have 0V on GND
>> 3.3V on data lines (may be there only for a very short test cycle so
>> you can't notice) ca. 1V on the ID line
>> 5V on the VBUS line
>> For charging, the data lines are not used. Therefore having a good 5V
>> input is most important.
> i don't have a clue about the usb-standard, so i describe here what i
> hopefully understood:
> When i look at the big connector side of my usb
> cable (other side connected to the phone), i guess the left of the outer
> pins should be GND and have 0V (and 0Ohm) to the shielding.

Yes. Or no. Some cables have 0 Ohm to shielding others don't. But the GTA04
should have 0 Ohm between both.

> The two inner pins are the data lines, of which one should have 3.3V
> for some micro seconds and the other is the ID Line and should have 1V
> (both against the GND, and shield).


> The right outer pin then is the VBUS line and should measure 5V to GND.


> Is there something to say about the resistances between the lines on
> the phone? I want to check if i did some bad connections...

If you take an ohmmeter (with low voltage!) you can usually see some values
above kOhms.

> I guess i have to measure voltages, while the gta04 is running?


> If this
> should be measured while gta04 is off and connected to some power
> source, how can i measure them (where should i point my measure tip)? 

Hm. It could also be a broken ESD chip. It has some diodes inside which protect
against overvoltage but are high resistance if voltages are ok.

If these diodes allow reverse-current, a voltage on some data line may rise
VBUS to a level that is detected as a charger wakeup event.

On the other hand it still does not explain why it happens with a cable not plugged
to a power supply...


More information about the Gta04-owner mailing list