• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 30 subscribers
  • Views 120 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.

DIX9211: BFRAME Signal latency & Frame synchronization

Samuel TUGLER
Prodigy 20 points
Part Number: DIX9211
Other Parts Discussed in Thread: PCM9211, DIX4192

Tool/software:

Hi,

I'm exploring whether the DIX9211 would be a suitable transceiver for Ultranet

To summarize, one half of an Ultranet cable is a twisted pair carrying 8 channels of 48 kHz audio muxed together onto a 192 kHz AES3 signal. 
The data looks something like this:

Sync/demuxing is done with the help of LRCK and the BFRAME.

Currently, I'm testing the DIR part of the chip; the output will go into an FPGA to do channel demuxing, but right now I'm just looking at my logic analyzer.
The chip seems to be working correctly; decoding works just fine.

However, I just noticed the horrendous delay between a B-frame preamble and the BFRAME signal actually being raised (which, after re-looking at the datasheet seems normal).

Here is a logic analyzer capture.

Is there an alternative way of signaling the B-frame that would be more presice?

Or if not possible with the chip is there another that wouldn't have such delay?

Thanks,

  • 0
    TI__Mastermind 30761 points

    Hello Samuel,

     I am not aware of any other way of signaling B-frame.  Beside, PCM9211(DIX9211) , we have  DIX4192,  but I don't know  if that would have less delay (PCM9211 is our most popular and advanced device ) 

    Regards,

    Arash

  • 0 in reply to Arash Loloee
    Prodigy 20 points

    Hi,

    Well then, if there is no way of having the block start signal exactly at the correct frame, it appears that I won't be able to use this chip for my application. 

    I looked over the datasheet of the DIX4192 and it does have the required output signal for the DIR but the delay between the block start frame and said signal isn't specified anywhere.

    I might order one to test if this delay is present.

    I must say while for typical AES3, this delay shouldn't be a problem, I find it weird that a sync signal is delayed by so much what is the typical use case of this signal ?

    Thanks for your reply.,
    Samuel

  • +1 in reply to Samuel TUGLER
    TI__Mastermind 30761 points

    Hi Samuel,

    These are mature and popular parts and yet I didn't find any prior enquiry about B-frame to give you more info as a reference. All I can say is that  it can be used to indicate the start of the frame to the decoding DSP  ( you can read more in   Page 33 of the  datasheet)  or may be it can be used  to sync audio and  video in some applications.

     If you want to order DIX9211, i would suggest to get PCM9211 which is identical device with an extra ADC on board, may be later on the ADC can come handy.

    Kind regards,

    Arash

  • 0 in reply to Arash Loloee
    Prodigy 20 points

    Hi,

    Yes I agree that these are mature and well known, and my application is probably out of spec.

    Ordering the PCM9211 is most likely unnecessary but I'll keep it in mind.

    Thanks,
    Samuel

  • 0 in reply to Samuel TUGLER
    Prodigy 20 points

    Hi,

    So, after re-reading the datasheet and doing a new capture with the logic analyser, I can confirm that I'm an idiot.

    It was late, and I got confused while reading and thought that tlate was the delay between a sample on DATA and BFRAME.

    I now realize that tlate is the global latency of the system, including data and sync signals.
    And that the delay between DATA and BFRAME is ~17BCK.

    I apologize for any confusion,
    Thanks for the help,
    Samuel