• State Verified Answer
  • +1 person also asked this people also asked this
  • Replies 10 replies
  • Subscribers 29 subscribers
  • Views 458 views
  • Users 0 members are here
Options
Related

AD9162 does not achieve CGS anymore

jlecomte23

Hello.  I have designed a system that uses multiple AD9162 devices and one of the devices no longer passes the CGS stage.  This device was working fine, but then started failing in the CGS stage.  I have checked all of the power supplies and clocks and SYSREFs going to the devices, everything looks ok.  I was able to compare the bad device boot sequence to a know good device boot sequence and the results are listed below.

All register readings matched with the exception of the following:

REG ADDR | GOOD DEVICE| BAD DEVICE

0x092 | 0x01 | 0x03 (This value did clear to a 0x01 and remained that way from that point on, so I think it's ok)

0x034 | 0x24 | 0x14

0x302 | 0x07 | 0x00

0x470 | 0xFF | 0x00 (This is where CGS fails, I did not go any further)

I am using a total of 64 AD9162 devices (8 devices on 8 different PCBs) and a total of approximately 5 or 6 of them have this same issue.  Any help would be much appreciated.Thank you,

John

Parents
  • 0
    •  Analog Employees 
    on Aug 13, 2021 1:53 PM

    CGS failures are typically (90%) caused by clocking issues on one side of the link or the other.  Ensue you are providing the exact reference clock needed for the lane rates you are expecting.  Also check the signal level and signal integrity of the clocks.  Also, is SYNCOUT always low or is it toggling?

  • +1 on Aug 13, 2021 3:02 PM in reply to deljones

    Just a couple of comments.  The SYNCOUT is always low during the CGS phase on the channel that is failing.  Regarding the clock, I am running at 2GHz.  I use a common clock to drive 8 DACs on a single board.  The clock is distributed as LVPECL using a ADCLK954BCPZ chip.  I have looked at the clock on a high speed scope and I do pass the DAC Clock DLL status check so I believe the clock integrity is good.  On the board I am testing now, 7 out of 8 DACs work fine.  The one DAC that is not working used to work, but then failed and hasn't worked since.  There was no known event that caused the failure.

    I will try the PRBS test to see what results that produces.  I will let you know ASAP.

    Thank you,

    John

  • 0 on Aug 15, 2021 1:53 PM in reply to jlecomte23

    I have figure out my problem. My board and configuration are right. There are a bug in the datasheet of AD9163. One of the RSV bit must be set to 1, but say nothing in the datasheet. 

Reply Children
No Data