• State TI Thinks Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 1 answer
  • Subscribers 101 subscribers
  • Views 606 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.

Original question:

(66AK2H14):About equalizer parameter tuning of SerDes Hi Expert

(66AK2H14):About equalizer parameter tuning of SerDes Hi Expert

Hidekazu
Intellectual 770 points
Other Parts Discussed in Thread: 66AK2H14

Hi Expert


e2e.ti.com/.../677491
It is a continuation of the above thread.

Since the difference in parameters is still unknown, an error has occurred.
There is a difference between the parameters of the IBIS model of 66 AK and the contents of the user guide
Please tell me how to reflect the SIM results.

Best Regards,
Honma

  • 0
    TI__Mastermind 46460 points

    Honma,

    Please provide more details.  What parameter mismatch are you referencing?

    Tom

  • 0 in reply to Tom Johnson 16214
    TI__Guru* 95265 points
    Honma,

    Any updates?

    Regards, Eric
  • 0 in reply to Tom Johnson 16214
    Intellectual 770 points

    Hi Tom

    The presented table is a parameter list of the 66AK2H14 IBIS model (66AK2H14 IBIS-AMI),
    but it does not match the parameters of the user's guide. How do I reflect the simulation result?

    I want to know which parameter value of the user's guide the following parameter value (blue) matches.

    Option Attribute Setting Value
    SWING Output swing 　 1-10
    Cm1 EQ pre cursor tap 0-7
    C1 EQ post cursor tap 0-15
    control Input ATT Controls 0-15
    zero control 　 0-2
    BOOST CTLE Controls 0-15
    Gain 0-3
    Super Boost Enable 　 0-1
    EDGE Tap
    1/2/3/4/5    		 0-127/63/
    63/63/63
    Center Tap
    1/2/3/4/5    		 DFE Controls 0-127/63/
    63/63/63
    Center DFE Enable 0/1


    Best Regards,

    Honma

  • 0 in reply to Hidekazu
    TI__Mastermind 46460 points

    Honma,

    Support for these models is extremely limited due to IP licensing restrictions.  This is why we limit access to these model files and expect all users to be highly proficient with SERDES model simulation.

    You have a glimpse into the complexity and capability of the SERDES blocks through the models.  However, the SERDES blocks implemented in the processor are abstracted by software drivers provided by the SERDES IP owner so that much of the configurability is hidden.  Similarly, the registers documented in the User Guide are only the registers that we were allowed to expose.

    In your other post, Eric pointed you to the diagnostic code available for SERDES link commissioning.  You will need to be familiar with the SERDES configuration options available in the code.  You can then translate that across to the controls available in the models.

    The transmitter block has 3 controls that are programmable.  They are the SWING, CM1 and C1.  The swing settings adjust the output amplitude.  CM1 is the pre-cursor coefficient and C1 is the post cursor coefficient.  These correlate to the settings in Table 14-1 of the UG.

    The receiver equalizer is intended to be used in automatic mode.  Therefore, for your simulation purposes where you need to establish your transmitter settings, the equalizer should be set on automatic mode.  However, as you can see in the Errata document, certain data patterns cause the automatic equalizer to fail.  Therefore, at board commissioning, you must follow the steps discussed in the Link Commissioning Guide to complete the receiver configuration.

    The equalizer block has 2 primary controls available in manual mode.  These are Attenuation and Boost.  All of the remaining CTLE and DFE settings are hidden as stated in the SERDES UG.

    Level and attenuation settings are important.  The equalizer will only function properly as long as the transforms are all linear.  Saturation must not occur at any point in the transmitter or in the receiver as this causes the signals to become non-linear.

    Tom