Hi Patrick, I'll take the blame for this-- in the implementation of the data bypass on Shimmer2 (and 2R) The external "SO" pin is no longer bidirectional. As a result I2C SDA is broken. The AnEx manual and Shimmer documentation needs to be updated.
There solution is to use the internal SO (SOMI) line. The connector pins are difficult to solder to. You may want to wire to resistors on the shimmer board (I can send you a drawing), or request a raw daughtercard from Shimmer for easy access to those pins. Please email [email protected] and describe which approach suits you and/or preference between hard wired and a connector solution. Sorry for the inconvenience, Ben On Tue, Feb 7, 2012 at 5:50 AM, Patrick Kugler <[email protected]> wrote: > Hi everyone, > > we are currently trying to connect an external sensor to a Shimmer2R > using the AnEx-Board. The sensor has a digital I2C-Interface and worked > fine with a different microcontroller. However when connecting the > sensor to the Shimmer, we ran into some electrical problems with the bus. > > I used a modified version of the Boilerplate firmware, as it already > contains I2C communication with the Mag sensor. However, I could not > communicate with the external sensor. Using an oscilloscope, we found > that the I2C requests are sent correctly to the sensor. When using a > different I2C address (e.g. from the Mag) the external sensor does not > respond (which is correct) and communication with the Mag also still > works fine. When using the correct I2C address for the external sensor, > it does try to send an ACK on the bus by pulling it low. However the bus > does not completely go to GND (as it should), but instead goes to half > VCC (1.5V) - which is totally wrong.... As this is an invalid bus level, > the MSP does not read the ACK and communication is not possible. > > Could this be a problem of the pull-up? At the beginning we assumed we > have to add our own pull-up and added a 4.7 k, but this does not seem to > be necessary. It seems that Shimmer is using a 1.2 k internal Pull-Up, > which is quite low but should be still OK. However I also could not find > any documentation about this. > > Currently I have some theories: > - The 1.2k pull-up is too small for the external sensor to pull the bus > low. Could I remove this pull-up? > - The connection wire to the external Sensor is too long (50cm) > - We have some problems with proper GND or VCC on the external sensor board. > - Using the 4.7k parallel to the internal pull-up we damaged either > Shimmer or the Sensor. > > Any thoughts? > > Regards, > Patrick > > -- > > --------------------------------------------------------- > Dipl.-Inf. Patrick Kugler > Researcher in the Digital Sports Group > > Pattern Recognition Lab > Friedrich-Alexander University Erlangen-Nuremberg > Martensstr. 3, 91058 Erlangen, Germany > > Room: Haberstr. 2, Room 01.006 > Phone: +49 (0)9131 / 85-28980 > E-Mail: [email protected] > Web: www5.informatik.uni-erlangen.de/~kugler > --------------------------------------------------------- > > This E-Mail (and any attachments) may contain confidential > or personal information. Do not use, copy or disclose the > content of this E-Mail without prior approval of the sender. > > _______________________________________________ > 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
