• State
  • Locked Locked
  • Replies 5 replies
  • Subscribers 34 subscribers
  • Views 127 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.

SN65DPHY440SS: questions about simulation and EQ gain

Ruixin Feng
Intellectual 2565 points

Part Number: SN65DPHY440SS

Hi team,

Customer has several questions about SN65DPHY440SS. hope you could help to answer. thanks!
1. Simulation Model Clarification (S-Parameter vs. IBIS-AMI)
TI.com provides S-parameter models (.s4p) instead of traditional IBIS or IBIS-AMI models.
  • As an active device (Retimer), we typically expect an IBIS-AMI model for SerDes-type simulations to account for non-linear behavior and clock recovery.
  • S-parameters are generally considered passive models. How can we perform a reliable system-level Signal Integrity (SI) simulation (e.g., in ADS, Sigrity, or HyperLynx) using only S-parameters for an active retimer?
  • Does this model accurately reflect the internal gain, equalization, and redriving characteristics, or is there a hidden/separate model available for active simulation?
2. Full EQ Gain Curve Request
The datasheet only lists EQ gain values for specific frequency points: 500MHz and 750MHz.
  • Does TI have a complete EQ frequency response curve showing the gain across the entire spectrum?
  • Specifically, what is the expected EQ gain at lower frequencies, such as 250MHz (500Mbps)? We need this data to evaluate signal over-equalization risks.
3. Link Insertion Loss (IL) Budget Calculation
Based on the datasheet, the maximum EQ gain at 750MHz (1.5Gbps) is 5dB. According to the MIPI specification, the allowable insertion loss at this frequency is 2dB.
  • Is it correct to assume the total supported link insertion loss is approximately 5dB (EQ) + 2dB (Spec) = 7dB?
  • Can we use this 7dB value as a hard constraint for our total PCB trace and connector loss budget? If not, what is the recommended methodology for calculating the maximum channel loss this chip can compensate for?
Best Regards,
Rayna
  • 0
    TI__Guru**** 187071 points

    Rayna

    1. Simulation Model Clarification (S-Parameter vs. IBIS-AMI)

    The S-parameter is a return loss measurement so it will not suitable for simulation. Unfortunately we do not have a IBIS-AMI model for this device. The only model we have is an encrypted HSPICE model, but we can not provide any support for it.

    2. Full EQ Gain Curve Request
    The datasheet only lists EQ gain values for specific frequency points: 500MHz and 750MHz.
    • Does TI have a complete EQ frequency response curve showing the gain across the entire spectrum?
    • Specifically, what is the expected EQ gain at lower frequencies, such as 250MHz (500Mbps)? We need this data to evaluate signal over-equalization risks.

    I am having trouble uploading the EQ curve data, so I will send it to you over the WebEx.

    3. Link Insertion Loss (IL) Budget Calculation

    I had an extensive discussion on this topic, please see this E2E thread,  SN75DPHY440SS: Does the D-PHY 2nsec limit apply to the output side independent of the input side?  and let me know if there are additional questions.

    Thanks

    David   

  • 0
    TI__Intellectual 2565 points

    Hi David,

    thanks for your reply!

    for question 3, customer sensor output is followed by a long cable, so they need to implement this chip at the cable's far end to compensate for long-line insertion loss. Currently, they need to define the cable loss specifications based on the chip's compensation capability to proceed with the cable design and fabrication. Since the sensor's MIPI data rate is 500Mbps (with a corresponding Nyquist frequency of 250MHz), they aim to determine the allowable cable loss at this 250MHz frequency point according to the chip's performance. should customer use 5dB(device EQ) + 2dB (MIPI requirement) = 7dB to design the cable? 

    Best Regards,

    Rayna 

  • 0 in reply to Ruixin Feng
    TI__Guru**** 187071 points

    Rayna

    Does the sensor TX provide any insertion loss compensation? If not, then the insertion loss compensation comes from the DPHY440 RX. The 4dB compensation is at 500MHz and 5dB compensation is at 750MHz  For 500Mbps which is Nyquist frequency 250MHz, if you look at the max RX compensation at 250MHz with the s4p file I provided, it is around 2dB, so they should use this as their guideline for the cable insertion loss.

    Thanks

    David

  • 0 in reply to David (ASIC) Liu
    TI__Intellectual 2565 points

    Hi David,

    Just to confirm, our device compensation is 2dB, and MIPI standard can accept 2dB loss, so the cable insertion loss can be 4dB. is that correct understanding? thanks!

    Rayna 

  • 0 in reply to Ruixin Feng
    TI__Guru**** 187071 points

    Rayna

    Yes, your understanding is correct.

    Thanks

    David