accelerometer driver

Harald Welte laforge at openmoko.org
Fri Aug 29 08:12:00 CEST 2008


Hi Wolfgang,

On Fri, Aug 29, 2008 at 01:20:44PM +0800, Wolfgang Spraul wrote:

> I have heard a lot of mysteries around the accelerometers.
> Are they very sensitive or not? How sensitive? Are they calibrated  
> somewhere?

They are factory-calibratet for an operating voltage of 2.5V.  According the
data sheet this calibration can be used from 2.16 to 3.6V operating voltage.
Openmoko uses them at 3.3V, so they need not to be re-calibrated.

The factory calibration data is automatically read from device-internal
non-volatile memory every time the device is powered up.

> In the factory we saw big variations of values they produced, when lying 
> flat on a table. Nobody could really understand why.

According to the data sheet, the zero-g level offset accuracy should be +/-
60mg.  This should mean a perfectly unaccelerated device (x=0g, y=0g, z=1g
[earth gravity]) should have values that change up to +/- two digits.

Any bigger values are out of spec and either mean a software or hardware bug,
or a defective acceleration sensor.

if hardware+software are verified to be OK, the acceleration sensors do have a
self-test mode by which an electrostatic force actuates the MEMS.  I would
assume that this self-test is part of the production testing process of the
GTA02 - or if it isn't yet, it probably should become part of it.

> Do we need 2 accelerometers, if they are already 3D? We are planning to 
> take one out in future products... What do you think?

Sean always required 2 accelerometers, 3D each.  I never questioned this
requirement.  I believe the main reason was to be able to reliably detect
rotation of the device around its own axis (or any axis going throug the
device).

> Until today there are bugs in the driver talking to both at the same  
> time I think, in other words even if there is some advantage to having  
> two, they are not working right yet...

well, driver bugs just mean that there are apparently no R&D resources for
fixing the bugs...

-- 
- Harald Welte <laforge at openmoko.org>          	        http://openmoko.org/
============================================================================
Software for the world's first truly open Free Software mobile phone



More information about the openmoko-kernel mailing list