Re: [Paparazzi-devel] Issue using I2C on LPC

[Christoph Niemann]

Hi,

next I tried to do the same with the booz-board. 
My Hardware was: Booz 1.1 Board and Booz 1.1 IMU. As I wrote up to now, I had no luck, getting the MS2100 to work, so I'm going t use a HMC5883L.
First I tried using  <subsystem name="imu"           type="b2_v1.2"/> since this seemed to be the most similar to me.
Result: In this way the LPC doesn't do anything. No telemetry, no LED, nothing.
Second I tried <subsystem name="imu"           type="b2_v1.1"/> since I really try to use a booz 1.1 IMU. But I commented out the MAG-Stuff in the Makefile like in the attachment (makefile_1) and tried to load the hmc-module with <load name="mag_hmc5843.xml"/> like I did before with the TWOG.
Result: the LPC-Starts and everything without the MAG works. The RAW-Message appears, but its containing only zeros. The LA says SDA and SCL are high. Might this be an interrupt vector problem?
Third I tried the <subsystem name="imu"           type="b2_v1.1"/> subsystem again but changed the makefile like in makefile_2 . I copied the HMC-Stuff from the 1.2 makefile and didn't load the hmc-module seperately.
Result: Like with the 1.2 Makefile the LPC doesn't start anymore.

I'm a little bit desperate now.

Could it be a problem with the Interrupt-assignment?
Might it be a Hardware-Problem? Broken I2C on LPC?
What to do? Could someone try to reproduce this?
Booz-Imu-Users: What MAG do you use and how?

Thanks for your help

Christoph

2012/7/18 Christoph Niemann <address@hidden>

... blocking capacitor ... :-) not clocking capacitor

Cheers

Christoph

2012/7/17 Gautier Hattenberger <address@hidden>

With the latest v3.9 (v3.9.2_testing-26-g527837e), it is working fine. The i2c is sending the request to the ETS baro (which is not connected for me, so I have no reply after that).

G.

Le 17/07/2012 18:15, Christoph Niemann a écrit :

Hi Gautier,

this one isn't working in the new version but did in the old. The version was in my first mail (sorry, can't copy it this mail is from my mobile). As I wrote it works will with the tiny and ppzuav-imu.

Cheers

Christoph

Am 17.07.2012 13:39 schrieb "Gautier Hattenberger" <address@hidden>:

The file you have send is the one that works, right ? Can you also send the latest one that is not working. And the result of ./paparazzi_version (or ./paparazzi_git_version)

Gautier

On 17/07/2012 09:57, Christoph Niemann wrote:

Hi Gautier,

currently the booz board isn't mounted into the airframe. I wanted to get the avionics working completely before mounting it, so currently I can't start the Motors. I wanted to start the investigation by taking something, that was working before. So I took old repository for my Funjet and tried it with the old one. I attached the airframe file. I just changed some things needed to get it compiled (attitude->ahrs and ublox_lea5h->ublox).  Everything else was the same. In the old version I get (in this case) the BARO_ETS Messagge and the I2C works, in the new one not.
This morning I also tried something else: I took a plane containing a Tiny13 and ppzuav IMU (also I2C) and it was working -> so I'm suspecting it to be my fault in the configuration somewhere, but I don't know where.

Cheers

Christoph

2012/7/17 Gautier Hattenberger <address@hidden>

Hi Christoph,

Can you start the motors ? Can you send your airframe file (or use pastebin or alike) so we can see the what is your exact configuration. Thanks

Gautier

On 16/07/2012 23:23, Christoph Niemann wrote:

Dear Community,

some weeks ago, I wrote a mail about my problem, using the old booz IMU with the HMC5843 or HMC5883L with the lpc based booz ap.
Investigating this further, I found that for me, using the pretty current 3.9 branch (v3.9.2_testing-3-g38ebd01-dirty), the I2C for LPC doesn't seem to work at all.
I tried both the hmc5843 and hmc5883l on the booz board. It didn't work. Next I tried those boards using a TWOG, it did't work. I suspected there is an issue with the hmc driver, so I tried an ETS Baro, it didn't work either. So I tried the code from about a year ago (95a65a7-dirty). The baro was working.
What could be the Problem? I used a logic anaylzer, to investigate it further. in the working version, there was data going through the bus. In the new version both lines where high.
Every help would be appreciated.

Best Regards

Christoph

_______________________________________________
Paparazzi-devel mailing list
address@hidden
https://lists.nongnu.org/mailman/listinfo/paparazzi-devel

_______________________________________________
Paparazzi-devel mailing list
address@hidden
https://lists.nongnu.org/mailman/listinfo/paparazzi-devel

_______________________________________________
Paparazzi-devel mailing list
address@hidden
https://lists.nongnu.org/mailman/listinfo/paparazzi-devel

_______________________________________________
Paparazzi-devel mailing list
address@hidden
https://lists.nongnu.org/mailman/listinfo/paparazzi-devel

_______________________________________________
Paparazzi-devel mailing list
address@hidden
https://lists.nongnu.org/mailman/listinfo/paparazzi-devel

_______________________________________________
Paparazzi-devel mailing list
address@hidden
https://lists.nongnu.org/mailman/listinfo/paparazzi-devel

imu_b2_v1.1.makefile_1
Description: Binary data

imu_b2_v1.1.makefile_2
Description: Binary data