• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 2 answers
  • Subscribers 63 subscribers
  • Views 203 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

UCD90320: UCD90320 vs UCD90320U

AlexWalton42
Expert 5655 points
Part Number: UCD90320

Hi Team,

Per the datasheet, the “U” suffix part is recommended for “Telecom and Networking” equipment. In addition, they reference SEU events

Features

  • Ultra Low Alpha(ULA) Mold Compound to reduce soft errors caused by alpha particles
  • Single Event Upset (SEU) Detection

This sparked curiosity if customer feedback led to this “U” version being created?  Can you share any information on this?  Ultimately, Should they be concerned about the part performance without the suffix?  And/or should we consider moving to the U suffix part for our applications? 

Preferably, they'd like to have a short call to ask any additional pointed questions. 

Thank you in advance

  • 0
    TI__Guru 60800 points

    Hi

    Both U and Non-U can be used for Telecom and Network.

    We have customer asked to provide sequencer which is less SEU sensitive so we implement the ULA mold compound.

    It is application driven and are they SEU sensitive? 

    Regards

    Yihe 

  • 0 in reply to yihe
    TI__Expert 5655 points

    Hi Yihe,

    My customer has had some unexplained card reboots (all UCD90xxx devices we use are in question, not just the UCD90320), and the existence of a ‘U’ device makes them wonder if we might have had SEU events in the ‘non-U’ devices we use today.

     Has your team detected SEU in the SRAM? 

    The answer… we don’t know.  If the ‘non-U’ device was affected by an SEU event, is there a way we’d be able to know?  A register that records the information that we could read after the fact?  How would an SEU event in the SRAM of a ‘non-U’ device be handled?  Part reset?  Device lock-up?  Something else?  Would the response be random or deterministic?

    Thanks in advance!

  • 0 in reply to AlexWalton42
    TI__Guru 60800 points

    HI

    SEU is most caused by two parts: alpha particles and Neutron beam.

    Having ULA mold compound helps to minimize the impact from alpha particles. but the impact from Neutron is always there.

    We haven't done any Neutron beam test on the part.

    SEU is a soft error and the result from SEU is random. Customers may not not notices this.

    Regards

    Yihe

  • 0 in reply to yihe
    TI__Expert 5655 points

    Hi Yihe,

    Can you help me understand which applications / environments it would be worth considering the "U" device? 

    And can you confirm that there is no way to determine that the "u" variant would've prevented the failures they have seen?

  • 0 in reply to AlexWalton42
    TI__Guru 60800 points

    Hi

    This is up to the customer and they need evaluate the SEU from the system to determine the risk level. 

    At this moment, there is no any signs to indicate that the issue is related to the SEU.

    Even with U variants, it just reduces the FIT rate of the SEU and it can not totally eliminate and can not correct the errors caused by the SEU

    Regards

    Yihe