• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 2 answers
  • Subscribers 33 subscribers
  • Views 670 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.

DS25BR110: Jitter limits for verification LVDS verification

Marion Loessl
Expert 3675 points
Part Number: DS25BR110
Other Parts Discussed in Thread: DS25BR100

Hi!

my customer wants to verify a through a DS25BR110 processes LVDS signal without an error identification/correction connected afterwards.

the only parameter for the input signal is the differential input high threshold. High big can the jitter on the LVDS input of DS25BR100 get, so that the signal is still correctly identified?

Is there material or an app note regarding that topic?

Measured signal in and output:

Kind regards,

Marion

  • 0
    TI__Mastermind 43795 points

    Hi Marion,

    The attenuation of the input signal is not a problem in this case.  It looks like there is still an open eye at the input side.  The 200Mbps datarate is relatively slow for the DS25BR110, but it does look like the input equalization is having some positive impact on the output signal.

    The glitch in the output signal could just be something related to the measurment location and the receiver loading.

    If the output signal is disconnected from the Rx load and terminated with a 100 Ohm resistor instead does the waveform look better?

    Regards,

    Lee

  • 0 in reply to Lee Sledjeski
    TI__Expert 3675 points

    Hi Lee,

    unfortunately this topic is still open.

    The data rate is actually 960MBit/s. The crashes of the signal were caused by false regulation of the impedance.

    The system works without errors, but with only this measurements, the customer can not know how much spare tolerance he has.

    So again the question:

    What is the max value for the jitter on the input of the LVDS signal, so the Signal is detected correctly?

    Is there an app note similar to snla053a, which descries the Verification?

    Kind regards,

    Marion

  • 0 in reply to Marion Loessl
    TI__Mastermind 43795 points

    Hi Marion,

    There is no app note like this for the DS25BR110.  The BLVDS devices are serializer/deserializer devices and the signal is sampled and must meet certain timing limitations.  The DS25BR110 is a just an analog buffer/equalizer.  The output is a direct response to the input signal.

    The threshold of the DS25BR110 is guaranteed to be +100 or -100mV.  The scope waveforms show > 30mV of margin to this specification.  Typically the real threshold is < 25mV, so the actual margin is over 100mV.

    Regards,

    Lee

  • 0 in reply to Lee Sledjeski
    TI__Expert 3675 points

    Hello Lee,

    sry for asking another question, but I need to make sure, I understand:

    1. So from a "timing" perspective, as of this (from Wikipedia):

    Our device would pass on this kind of "error"

    2. From a Threshold perspective, you say, that the signal looks good right?

    So the +-100mV threshold is meant for the eye itself right? the >30 and <25mV is the deltaV, that is defining the open eye?

    Kind regards,

    Marion

  • 0 in reply to Marion Loessl
    TI__Mastermind 43795 points

    Hi Marioin,

    1. Yes, our device would pass on any periodic or other deterministic jitter not associated with attenuation of the signal.

    2. Yes, the signal looks good.  The open eye is approximately +/- 130mV in amplitude. This leaves a 30mV margin to our datasheet specification.

    The <25mV reference is based on typical device threshold data which is much smaller than the 100mV specification.

    Regards,

    Lee