• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 1 answer
  • Subscribers 21 subscribers
  • Views 423 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.

LMH1983: Timing margin between HIN and VIN as well as HIN and FIN for proper operation.

user4779049
Expert 2345 points
Part Number: LMH1983
Other Parts Discussed in Thread: LMH1981

 Please advise me on my question bellows:

 

Question:

So, can you let me know the timing margin between HIN and VIN

as well as HIN and FIN for proper operarion?

Backgrand:

I am using combination of the LMH1981 and LMH1983.

Generally the HSOUT, VSOUT and OEOUT from the LMH1981 are

Directly input to the corresponding input of the LMH1983.

However, I put some glue logic between LMH1981 and LMH1983

Witch case some amount of unpredictable delays.

 

Referring the timing charts on the datasheet of the LMH1981,

the leading edge of HSOUT and VSOUT/OEOUT are exactly same

on the Even field.

I concerns that the unpredictable delays may case unexpected results.

 

Mita

  • 0
    TI__Guru** 107030 points

    Hi Mita-san,

    V and F are used for auto format detection and TOF alignment, if the delay in glue logic is in the nano-second range, I think this should not be matter.

    All clock outputs are based on PLL1 and H is the reference to PLL1. Again the delay in glue logic is not matter.

  • 0 in reply to Noel Fung
    Expert 2345 points

    Noel-san,

     

    Tank you for your prompt answer.

    Please allow me to discuss this topic more deeply.

     

    My concern is not the absolute delay of the F, V and H, but skew of H and V and/or H and F.

     

    I concern that such skew may case the trouble on the format detection function.

     

    I suppose that format detection logic n the LMH1983 measures the HSYNC period using

    the 27 MHz. It also count the number of lines in the VSYNC interval.

     

    If the H and V change their state at the same instance and are sampled with the 27 MHz clock,

    the phase of the sampled V is +1, 0 or -1 clock period against the phase of the sampled H.

    This fools the format detection logic.

     

    If V or F status is referred one clock later the H transition, this problem can be prohibited.

    Is this correct?

     

    Mita

  • 0 in reply to user4779049
    Expert 2345 points

    Noel-san,

     

    Let me recall previous my question.

    Instead, please allow me to confirm the fact bellow.

     

    Does the LMH1983 tolerate several nano-seconds of skew between HSIN and VSIN

    as well as the skew between HSIN and FIN, for the stable format detection?

     

    I appreciate it if you can give me feedback shortly.

     

    Mita

  • +1 in reply to user4779049
    TI__Guru** 107030 points

    Hi Mita-san,

    The LMH1983 uses a two dimensional look up table to determine format - it looks at V and F to determine if it is looking at an interlaced or progressive format, and it then measures Hin frequency - using the combination of H frequency and interlace/progressive, it determines format.

    In other words, the phase between H-V or H-F is not matter.