• State TI Thinks Resolved
  • Locked Locked
  • Replies 2 replies
  • Answers 1 answer
  • Subscribers 33 subscribers
  • Views 903 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:

SN65LV1224B Data Clock Frequency Range for Locking

SN65LV1224B: DS92LV1212AMSAX/NOPB Deserializer Replaced With SN65LV1224BDB - No LOCK (Random Data)...

Eric Brown
Prodigy 50 points
Part Number: SN65LV1224B
Other Parts Discussed in Thread: SN65LV1023A, SCAN921226H, SCAN921224

Good day folks,  

 

We have an application where LVDS data is being supplied by a device that we cannot command to send Sync data.  Up until it went EOL we had been using DS92LV1212AMSAX/NOPB Deserializer with complete success.  When that went away we tried the recommended SN65LV1224BDB.  We were not able to get LOCK to stick (our Deserializer won't stay locked if it does momentarily acquire lock) .  The recovered rate is 264 Mbps (so we are supplying a 22 MHz Ref Clock). 

 

We tried various other Deserializers and none seem to work.  We have tried adjusting the Ref Clock (using more precise sources, etc.) but to no avail.  As I mentioned above, we do not have the option to force the data source device to issue a Sync sequence so we must rely on the Deserializer locking with the random data method.  I read on this forum that one of the nice things about the "old" 1212 was that it could accept a clock with a 5% tolerance whereas the new parts need something like +/-100 ppm. 

 

Is there any part option available today that has that wider tolerance on the Ref Clock it accepts and can operate with that 22 MHz clock?  If one of the Enhanced Performance devices would work, we would be willing to use that. 

 

Many thanks, and best! 

Eric

  • 0
    TI__Genius 10250 points
    Hello Eric,

    The SN651224B should be able to lock to random data. However, as you mentioned, it does require a REFCLK which is the same frequency as the transmit clock on the other side with +/-100ppm accuracy. The following app note discusses the differences between DS92LV1212A and SN65LV1224B:
    Migrating from DS92LV1021A/1212A 10-bit SerDes to SN65LV1023A/1224B: (www.ti.com/.../slla435 ).

    In terms of similar parts, there is SCAN921226H and SCAN921224 which have similar functionality to DS92LV1212A (incl. the 5% tolerance on REFCLK). However, they are not pin-to-pin compatible (bigger package), since they have extra pins for JTAG.

    Regards,
    Yaser
  • 0 in reply to Yaser Ibrahim
    Prodigy 50 points

    Good day Yaser, 

     

    Many thanks for the fast reply with great info.  I was not aware of that App Note and it's got lots of good info. 

     

    I plan to try the SCAN921226HSM/NOPB with a spin of a test board.  That will take some weeks of course; I'll reply once I get results from that attempt. 

     

    Thanks again for the support, and best, 
    Eric