• State
  • Locked Locked
  • Replies 5 replies
  • Subscribers 99 subscribers
  • Views 715 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Interface a CMOS sensor with the DM355

SAKHO
Intellectual 390 points
Other Parts Discussed in Thread: TMS320DM355, TVP5146

Hello everybody,

At present I work on the demo board TMS320DM355 and exactly on the VPSS part. I make of the image capture  with the sensor CMOS MT9J003. 
I shall in fact want to interface this sensor with the DM355 via Imager Daugther Card.

I manage to select the Imager interface instead of TVP5146. Then, the problem which I have is the following one:

During the boot of demo board, I have message saying to me:

 

vpfe-capture vpfe-capture: could not setup input for mt9t031 

vpfe-capture: vpfe capture clocks disabled

vpfe-capture: probe of vpfe-capture failed with error -14

vpfe_register_ccdc_device: DM355 CCDC

dm355_ccdc: probe of dm355_ccdc failed with error -22

 

Then, that would be what the problem?

I work with an uImage stemming from the Arago project which I arrange all the source code. The sensor CMOS is piloted by bus I2C and I shall want to make a raw capture without treatment behind. I already know that there is a controller MSP430 (which controls the sensor), but is it he who shall raise problem? How controller manages to communicate with my sensor and how to send the good commands and the good signals to make the capture of image?

It seems that the driver of the sensor MT9T031 can be used for the sensor MT9J003. The difference is not enormous. You confirm it to me? Otherwise, it is necessary to develop a driver specially for my sensor? 

If there are modifications to make on the code source, which files it will be necessary to modify and where exactly?

I really need help. Then, thank you for dedicating me a little of your time.

SAKHO

 

  • 0
    Intellectual 700 points

    Be aware that the MT9J003 is a 1v8 interface.  Also, the I2C registers for set up and control are considerably different.

     

  • 0 in reply to I Jeffray
    Intellectual 390 points

    Thank you for your answer.

    Then, that has to correspond to what this 1v8 interface? What it implies in the interfacing?

    Effectively, I noticed that I2C registers is really different and I have a lot of problem to communicate with this bus. The MT9J003 provides a 16-bit register address space and it is a problem to access it with the DM355 EVM I2C bus.  If you have ideas for the I2C bus so that the CMOS  sensor MT9J003 can communicate, I shall be very grateful to you.

    Once again furthermore, thank you for your help.

    SAKHO

     

  • 0 in reply to SAKHO
    Intellectual 700 points

    It should be no problem at all to access it from the DM355 I2C bus... it's just two bytes rather than one for each access.  

    Regarding the 1v8 ... just a heads-up that it's different from the 3v3 levels of the MT9P031 sensor so you'd need to add level shifters or configure the DaVinci IO to be 1v8 (I don't know if that's possible)

     

  • 0 in reply to I Jeffray
    Intellectual 390 points

    But, it is the I2C bus which manages automatically the access in two steps ( 2 Bytes) according to the size of the register address (16 bit for the MT9J003)?

    I modified, in fact, the driver of the sensor MT9T031 by changing the names of registers and the addresses (16 bits on the MT9J003). Just with these changes, I manage always not to communicate on the I2C bus. Then, I began coding the operations of writing and reading in two stages. But, always nothing...

    I do not see how the I2C interface of the DM355 manages to manage when the register address is 16 bits...

    Still thank you.

    SAKHO

     

  • 0 in reply to I Jeffray
    Prodigy 185 points

    I don`t know if it is really 1.8V only interface. In the absolute maximums there is a 3.1V for VDD_IO and in the other tables (like the table 27 and table 30) it is mentioned that the VDD_IO can be raised to 2.8V. I think that the VDD_IO can be operated at 1.8V or 2.8V and the table 25 refers only to the VDD_IO=1.8V option (described above the table).  The datasheet is really a little confuse.