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.

SN65HVD1040: Comm. Improvement

Part Number: SN65HVD1040

Team, we would like to seek your guidance regarding CAN communication signal quality issues observed in our system.

As illustrated in the attached diagram, the “Before” configuration represents the original product design, while the “After” configuration follows the 120Ohm  termination scheme recommended in the datasheet.

We have evaluated both termination approaches under the same test conditions. However, in both cases, the CAN signals still exhibit noticeable noise and degraded signal integrity.

From our preliminary analysis, we would like to share several hypotheses and request your feedback:

  1. Stub Length Impact
    The current topology may introduce relatively long stubs between the transceiver and the main bus, which could lead to signal reflections and ringing.

  2. Termination Placement
    The 120 Ohm termination may not be placed strictly at the physical ends of the bus, potentially causing impedance mismatch along the transmission line.

  3. Split vs. Standard Termination
    We are currently using standard termination. Would TI recommend split termination (e.g., 60 Ohm + 60 Ohm with center tap to GND via capacitor) to improve common-mode noise performance in this application?

  4. Common-Mode Noise / Ground Potential Difference
    Given this is a UPS / power system environment, there may be ground shifts or common-mode disturbances affecting CAN signal quality.

  5. Common-Mode Choke Usage
    Would adding a common-mode choke at the CAN transceiver interface help suppress noise in this scenario?

  6. Transceiver Drive Strength / Edge Rate
    Could the issue be related to transceiver output drive capability or edge rate control, especially under heavier bus loading or longer cable conditions?

  7. Cable and System-Level Factors
    Factors such as cable length, characteristic impedance, node count, and routing topology (daisy-chain vs. star) may also be contributing to the observed behavior.

Given our system architecture, could you please advise on the optimal termination scheme and any additional design recommendations to improve CAN communication robustness?
8d93a174-7125-4509-aeb6-ed3fb5ef31a6.png

  • Hi Brian,

    1. Yes, please see 10.2.1.1 of the data sheet recommending 0.3 m.

    2. Yes, please see figure 26 recommending the far ends of the bus and would recommend 120 ohm at the first circle missing termination in the after block and then the last 120 ohm as shown. I.e., just those two, while removing the mid termination.

    3. Yes, split termination should further help with noise.

    4. Yes, any noise disturbances can impact signal integrity. See if 1-3,5,6,7 significantly helps improve the impact.

    5. Yes, CMCs should further help as well.

    6. Yes, reducing the speed under heavier loads should help.

    7. Yes, and would suggest 10.2.1.1's recommendations, thanks.

    Best Regards,

    Michael.