• State TI Thinks Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 2 answers
  • Subscribers 33 subscribers
  • Views 254 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.

Original question:

LMH1297: EQ Gain Adjustment for LM1297

LMH1297: LMH1297

Mentor Tseng
Prodigy 195 points
Part Number: LMH1297
Other Parts Discussed in Thread: LMH1229

Tool/software:

Hi,

    I see this reply in E2E forums. In my application I use 3 pcs LMH1297 to implement 12G-SDI receiver and transceiver. But I meet long cable (50 meters) issue and it related to this description shown in below.  

"Please note when SDI_OUT is enabled, we expect 10 to 15m of cable reach reduction. SDI-OUT causes cross talk and this reduces cable reach. "  

      

 LMH1297 Setting,

  - Chip 1:SDI_IO is EQ mode. SDI_OUT: loop-through 75-Ω output port. (Rx + loop-through Tx)

  - Chip 2: SDI_IO is EQ mode. SDI_OUT: unused.  (Rx only)

  - Chip 3: SDI_IO is CD mode. SDI_OUT: unused. (Tx only)

  My issue is 12G-SDI Rx can't work normally when Tx is enable @ 50 meters long cable. All test results shown in below is tested in 50 meters long cable.

   - Test 1-1: Chip 2 could receive 12G-SDI video normally when Tx of Chip 1 is "Disable"

   - Test 1-2: Chip 2 can't receive 12G-SDI video (unlocked) when Tx of Chip 1 is "Enable" 

   - Test 2-1: Chip 2 could receive 12G-SDI video normally when Tx of Chip 3 is "Disable"

   - Test 2-2: Chip 2 can't receive 12G-SDI video (unlocked) when Tx of Chip 3 is "Enable" 

   - Test 3-1: Chip 1 could receive 12G-SDI video normally when Tx of Chip 1  is "Disable"

   - Test 3-2: Chip 2 can't receive 12G-SDI video (unlocked) when Tx of Chip 1 is "Enable" 

It seems Tx has some influence to Rx channel, could you help to  confirm this issue? Thanks.

1.  when SDI_OUT is enabled, we expect 10 to 15m of cable reach reduction.  <-- is correct?

2. If above description is correct, which distance is required between Tx and Rx if hope without cable reach reduction?  

3.  In my current layout (SDI connector position), any suggestion to reduce this issue impact? ex: add shielding..

Thanks.

  • 0
    TI__Genius 9209 points

    Hi Mentor,

    Yes, we expect 10 to 15m of cable reach reduction when SDI_OUT is enabled.

    Mentor Tseng said:
    If above description is correct, which distance is required between Tx and Rx if hope without cable reach reduction?  

    There is no complete rule of thumb for this because resonances from crosstalk are difficult to calculate. A general rule is that the further the better, but if the distance is just right the effect might be amplified because of constructive interference. 

    Mentor Tseng said:
    .  In my current layout (SDI connector position), any suggestion to reduce this issue impact? ex: add shielding..

    One way to deal with EMI is to ensure that the correct impedance matching is met in the layout.

    Do you have a GND reference for your 75 ohm input and output traces? To me it looks like a void. You must ensure that the current has a return path by including a GND reference.

    Another suggestion is to use our new device the LMH1229 because it has an improved cable reach performance.

    Best Regards,

    Nick

  • 0 in reply to Nick Peabody
    Prodigy 195 points

    Hi Nick,

      The layout I shown is only layer 1 and 2, 75ohm trace is reference to Layer 3 for impedance required. Totoal Layout is shown in below.

     (1) In same LMH1297 chip, SDI_IO set as receive and it expect 10 to 15m of cable reach reduction when SDI_OUT is enabled. 

     (2) In different LMH1297 (in my case is Chip 2 and chip3), Chip 2 SDI_IO set as Rx and chip 3 SDI_IO set as Tx. Does it has same issue if Tx is enable? 

        --> Expect 10 to 15m of cable reach reduction when  other LMH1297's SDI_IO(Tx, CD mode) is enable. 

    Above 2 items is correct? Do you have any document or experimental data or this constraint? I need to some evidence to close this issue in my project, thanks.

  • 0 in reply to Mentor Tseng
    TI__Genius 9209 points

    Hi Mentor,

    We expect crosstalk to impact cable reach performance, but we have not validated for your case or a similar cases. 

    In our validation of SDI retimers, we have seen cable reach performance drop around 10m when crosstalk from adjacent sources is high. There is no official validation data on cable reach when SDI_OUT is enabled.

    One suggestion is to implement GND stitching spaced calculated distances from one another to provide improved shielding from crosstalk. Here is a good appnote that discusses GND stitching.

    https://www.ti.com/lit/an/snla288/snla288.pdf

    Best Regards,

    Nick