hi chris,

my guess is that the pressure sensor is getting the i2c events from the magnetometer; did you check that?

i have a simple driver-level solution that i'll provide shortly. it wasn't in the bmp085 driver because shimmer doesn't have any daughter cards that have two i2c devices.

stay tuned,

steve

On 07/18/2012 05:14 PM, Chris B wrote:
Hi

I'm coming up against an issue on Shimmer2r when wiring in both the
magnetometer (kinematics board) and pressure sensor (GPS/Press board)
into a 'BoilerPlate'-style app (i.e. a general do-it-all app). Upon
adding the pressure sensor hooks, and without calling it's init
anywhere, the app will no longer get to the Magnetometer.readDone event
when sampling the magnetometer. To demonstrate, i have gone and added
the pressure sensor hooks (NB no call to PSInit.init is present) to the
BoilerPlate app in contrib. i.e.

adding this to BoilerPlateAppC.nc

    components PressureSensorC;
    BoilerPlateC.PSInit         -> PressureSensorC.Init;
    BoilerPlateC.PSStdControl   -> PressureSensorC.StdControl;
    BoilerPlateC.PressureSensor -> PressureSensorC.PressureSensor;

and this to BoilerPlate.nc

       interface Init as PSInit;
       interface PressureSensor;
       interface StdControl as PSStdControl;

    event void PressureSensor.tempAvailable(int16_t * data){
    }

    event void PressureSensor.pressAvailable(int32_t * data){
    }

The result is no streamed sample output to the ShimmerConnect host
software, since the app never gets to see the Magnetometer.readDone
event (checked by sticking an led toggle there).

I know i could just add a compile flag and some ifdef statements to get
around it, but then the app loses it's generality. Any ideas on this?
Can the wiring of the two sensors coexist without interfering with the
other?

thanks

Chris


_______________________________________________
Shimmer-users mailing list
[email protected]
https://lists.eecs.harvard.edu/mailman/listinfo/shimmer-users

_______________________________________________
Shimmer-users mailing list
[email protected]
https://lists.eecs.harvard.edu/mailman/listinfo/shimmer-users

Reply via email to