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.

TVP5158 acquisition problem

Other Parts Discussed in Thread: TVP5158

Hello,

 

I'm using TVP5158 for capturing 4 channel (all NTSC or all PAL) on DM6467 in 4-ch interleaved mode. In some cases it happens that a horizontal line appears on the acquired video and shifts upwards in the following frames and then starts back from the bottom. This effects happens on all the four channels. Analysing the frames I noticed that the horizontal line is taken from a channel different from the correct channel. The lines are swapped in the frames. The problem seems to be in the de-interleaving procedure from the superframe to the four frame buffers. It seems that a wrong channel number is associated to some lines in the metadata of the superframe, so the lines are swapped from a frame to another one and viceversa in de-interleaving process. Resetting the TVP5158 the problem disappears.

 

Am I missing any configuration?

 

Thanks,

 

Luca

  • Hello Luca,

    I don't recall seeing this issue before.

    Are you using the sample code provided at the TVP5158 driver download web site?

        http://software-dl.ti.com/dsps/dsps_public_sw/dsps_swops_houston/ANALOG_VIDEO/DM_TVP5158_Driver.htm

    Regards,

    Jim

  • Hi,

    Installing the latest TVP patch (I was using a very old one), the problem does not occur anymore.

    But I noticed a new problem in a particular scenario: if I connect the same video signal (the video signal from the same camera duplicated via an amplifier), sometimes the acquisition starts getting corrupted and in the superframe that the video port receives some lines seem to be switched among different channels.

    For example, if I connect signal A, B and C to the TVP and signal A and B are from the same camera, most of the times the acquisition works correctly, but sometimes the lines of the frame A or B are switched with the lines of the frame C. When the corruption starts, lines keep on being switched until the reset of the TVP. This happens only if the same signal is connected to different inputs of the same TVP. Do you think it's a problem in detecting the synchronism? Is there any register of the TVP I should set to a particular value to have a more robust acquisition?

     

    Thanks,

     

     

    Luca