• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 33 subscribers
  • Views 238 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.

DS90UB953-Q1: Link_det is 0 or unstable on some PCBs after thermal shock

Amotz kats
Prodigy 60 points
Part Number: DS90UB953-Q1


Hi,

We're using the devices mounted on our PCBs for over 2 years now, and having reasonable performance so far without any functional problems.

[Our design is based on synchronous mode.] 

Recently we  initiated automotive environmental testing, and got the following problem after unpowered thermal shock test on some of the PCBs: 

1. No link with the receiver board (954).

2. when debug from the serializer side LINK_DET is either 0 or unstable.

3. The full GENERAL_STATUS register is found in one of the following values 0x00, 0x01, 0x20, 0x21 (not clear from datasheet what is bit [5]).

4. the power, PDB, boot straps are ok.

The PCBs were fine before the test, and the failures are clearly related to the physical stress, however we do not observe any visible/xray problem on the PCB or device.

Can you advice what could lead to such a problem?

should we look for a specific solder joint? trace?

known weakness of the device?

Is there any clue from bit [5] value?

thanks,

amotz

  

thanks,

amotz

 

  • 0
    TI__Mastermind 18755 points

    Hi Amotz,

    Are you able to provide some details about the thermal shock test just for my curiosity? 

    Also if you are in sync mode and you are not able to LOCK to the back channel then you should see unlock on the 954 as well, can you confirm?

    Bit 0 in GENERLA STATUS indicates that the back channel from the 954 cannot be recovered and bit 5 also indicated back channel errors (Im not actually sure why its reserved) so it seems like an issue with the back channel.  Are you using PoC? can you try to power the serializer locally?  Double verify all the power pins.

    Is there any stressing on the 954 side?

    Regards,

    Nick

  • 0 in reply to Nicholas Darash
    Prodigy 60 points

    Hi Nick,

    Thanks a lot for your prompt support!

    The unpowered thermal shock test consists on 1000 temperature cycles of "soaking" the PCBA for 1 hour at each of the max and min storage temperatures (-40, 105) with 2 minutes transition time between the min and max (and vice versa) temperatures.

    Yes, i can confirm the LOCK is 0 also on 954 side (DEVICE_STS = 0xD3). The full registers table is copied below.

    It is PoC based camera, and the power seems stable. We can communicate with the serializer from the camera MCU, without any problem (excluding the missing lock...). However, we will follow your advice and try to see of powering the serializer directly, has any effect on the link.

    thanks,

    amotz

    954 Registers readout:

        0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f    0123456789abcdef
    00: 64 00 1e 20 d3 01 00 fe 1c 10 7a 7a 81 00 00 01    d.? ??.???zz?..?
    10: 00 21 41 61 00 00 00 00 80 61 a8 e3 dd 00 04 04    .!Aa....?a???.??
    20: 00 01 00 01 00 00 00 00 00 00 00 00 00 00 00 00    .?.?............
    30: 00 00 00 71 40 00 00 00 00 00 00 01 14 6f 00 40    ...q@......??o.@
    40: 00 a9 71 01 00 00 00 00 00 00 00 12 01 00 02 00    .?q?.......??.?.
    50: 00 00 00 03 01 00 00 00 5e 03 00 30 00 30 e0 6c    ...??...^?.0.0?l
    60: 58 e0 e0 00 30 60 8a 6c 58 8a 88 00 60 7c 1a 32    X??.0`?lX??.`|?2
    70: 2b 2c e4 01 e0 04 60 c5 00 01 00 00 00 00 00 00    +,????`?.?......
    80: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00    ................
    90: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00    ................
    a0: 02 0f 00 00 08 19 00 00 00 00 00 00 00 00 00 00    ??..??..........
    b0: 1c 13 1f 08 25 00 18 00 9c 33 83 74 80 00 00 00    ????%.?.?3?t?...
    c0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00    ................
    d0: 00 43 94 03 60 f2 00 02 00 01 00 00 00 00 00 00    .C??`?.?.?......
    e0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00    ................
    f0: 5f 55 42 39 35 34 00 00 00 00 00 00 00 00 00 00    _UB954..........

            

  • 0 in reply to Amotz kats
    TI__Mastermind 18755 points

    Hi Amotz,

    If the serdes pair cannot lock after temperature cycles then it leads me to believe that the issue is either power related or channel quality related.  Try doing an ABA swap with a part that hasn't gone through this process. If it still fails then it just be a hardware issue.

    Regards,

    Nick