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.

LMH1297: LMH1297 LMH0397 SDI_IO EQ problem when cold

Part Number: LMH1297
Other Parts Discussed in Thread: LMH0397,

We use LMH0397 3G-SDI Bidirectional I/O With Integrated Reclocker and
LMH1297 12G UHD-SDI Bidirectional I/O With Integrated Reclocker

We use the devices in EQ mode by feeding an SDI video into SDI_IO through cable EQ and CDR and output via OUT0 to FPGA.
The chips are running in the power up default. From 0 to 85°C the chips are running fine.

They fail to function at -40°C for 720p60 or 1080p60 so 1.5G or 2.97G speeds, while 270Mbit do work.
Spec says –40°C to +85°C Operating Temperature Range


For debugging tests we used the "Cable EQ Bypass Override" feature
With EQ bypassed the cold issue is gone. But we did not implement these chips to disable the EQ.

    CMD REG VALUE MASK //COMMENTS
    RAW FF 05 07 //Select ConfigIO Register Page
    RAW 27 08 08 //Enable EQ Bypass override with value in Reg 0x27[2]
    RAW 27 04 04 //Bypass Adaptive Cable EQ



As alternative to enable the "–6-dB Launch Amplitude" feature also seem to fix the failing EQ.
But the launch amplitude is ~800mV where this setting is usually not required.
We see that this setting reduces the readback value of the Cable EQ Index (CEI) after adaptation.

    COMMAND REGISTER VALUE MASK //COMMENTS
    RAW FF 05 07 //Select ConfigIO Register Page
    RAW 34 80 80 //Enable –6-dB splitter mode

At cold temperatures the read value of the Cable EQ Index (CEI) after adaptation was higher than at room temperature.
We have the feeling that higher EQ CEI or EQ gain values are in correlation with the cold error.       



Is this a known issue ?
What is your advice?

Can we have manual control over the SDI_IO cable EQ settings like gain or SDI_IO CTLE etc?
Is there a newer programming guide than SNAU204C–March 2017–Revised January 2018

Many Thanks in advance
Marcello

  • Hi Marcello,
    1). There is a feature within the device called auto bypass. With auto bypass enabled, you can set EQ index level where device goes into bypass and also set a limit when device exits bypass mode. There is a default setting that works for evaluation board. Perhaps we may need to adjust this for your board. I can send you this register settings if you like.

    2). The reason -6dB mode works is because in this mode EQ index reading is lower than normal mode. Therefor your EQ index is lower.

    3). We do not recommend using manual control since this has not been characterized and supported.

    Regards,,nasser
  • Hi Nasser,

    to set the two auto bypass registers to higher values seem to avoid this issue for 3G speed as well. We have to check the results with propper temperature tests.

    I have 3 questions in addition:

    1. Is this recommended practice to increase the threshod and disable the EQ for short(2m) cables at cold(-40°C) temperatures?
    2. Is the EQ on or off when it powers up between the two threshold values of e.g. lower=0x15 and upper=0x19?
    3. Is there a similar way to deal with 6G and 12G signals on the LMH1297 as programming guide says the above registers are for ≤3G rates?


    Regards Marcello

  • Hi Marcello,
    1). This depends on the EQ index. If EQ index is low then this should be. Suggestion is to check different cable length. We have other customers who have changed default auto bypass EQ settings.
    2). I believe it is off under the condition you noted.
    3). At 6G or 12G auto bypass is disabled. I suggest to leave this as is.
    Regards,,nasser
  • Thanks for your support Nasser, i expect this will solve our issue. I´ll let you know if we need more assistance.
    Sincerly Marcello