• State Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 1 answer
  • Subscribers 33 subscribers
  • Views 175 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.

Original question:

TCAN1462-Q1: CAN FD Topology Design with SIC

TCAN1462-Q1: SIC Value

Zhihong Lin1
Prodigy 60 points
Part Number: TCAN1462-Q1

Tool/software:

Hi TI CAN Expert,

I want to understand deeper about CAN SIC value. 

My current understanding is:

  • When the CAN transceiver works as transmitter without 120ohm termination,  SIC is better than non-SIC.
  • When the CAN transceiver works as transmitter with 120ohm termination, there is no difference for SIC and non-SIC.
  • When the CAN transceiver works as receiver without 120ohm termination, there is no difference for SIC and non-SIC.
  • When the CAN transceiver works as receiver with 120ohm termination,  there is no difference for SIC and non-SIC

 

That means SIC value is only critical when the transceiver of a CAN node  works as transmitter without 120ohm termination, it can improve the signal integrity. Because SIC's main point is to provide low impedance when the transceiver is going to transmit dominant to recessive. During that period, the impedance is high if without SIC, so reflection is critical.

Do I understand correctly?

Regards

Nic

  • +1
    TI__Guru 51106 points

    Hi Nic,

    Yes, your understanding is correct in that SIC reduces ringing on the bus during transition from dominant to recessive, especially in non-ideal termination. I.e., differences between SIC and non-SIC can be minimal if there are no bus impedance mismatches to yield any ringing to begin with.

    However, SIC is not limited to only one specific case (without 120 ohms) as it is more about how the device actively damps the recessive edge regardless of whether the node has the 120 ohms or whether the main terminations are elsewhere on the bus. I.e., it helps in reduce ringing in any network where the bus impedance may deviate (not only at a node lacking 120 ohms)

    • Non-SIC have strong drivers in the dominant state but relies on passive pull-ups through the bus termination for the recessive transition.
      • In a classic well terminated linear bus, this is acceptable.
      • In a bus with multiple stubs, the abrupt release of the bus from dominant to recessive can cause ringing, especially at higher data rates.
    • SIC actively shapes or damps the recessive edge by providing a controlled current when going from dominant to recessive, smoothly transitioning the voltage on CANH and CANL. This reduces ringing that can occur if the bus impedance is not purely 60 ohms or if there are multiple line segments, thanks.

    Best Regards,

    Michael.

  • 0 in reply to Michael Ikwuyum
    Prodigy 60 points

    Hi Michael,

    I checked 1462 datasheet. The resistance in active recessive is 100ohm. However, I checked ISO, the resistance requirement is 37.5-66.5ohm, why it is different?

    Also, 1462 datasheet does not specify min/max resistance spec, only typ. Doesn't min/max matter here? What is the min/max value of 1462?

  • 0 in reply to Zhihong Lin1
    TI__Guru 51106 points

    Hi Nic,

    Please note that they seem to be specifying two different specs I.e., one is differential correlating to the driver impedance in the active recessive state and the other is single ended, optionally implemented to adjust the effective loading on the bus. Hence, comparing the ISO spec to the TCAN RIN spec is more appropriate.

    Further note that the spec is optional and the device do not have to meet the requirement to function, thanks.

    Best Regards,

    Michael.

  • 0 in reply to Michael Ikwuyum
    Prodigy 60 points

    Hi Michael,

    Sorry I missed the differential and single-ended statement. If we looked at the differential spec, the spec is 75-133ohm. Do you mean that it does not need to meet the 75-133ohm spec? Let's say if the SIC differential resistance is 50ohm or 200ohm, it also works?

    Regards

    Nic

  • 0 in reply to Zhihong Lin1
    TI__Guru 51106 points

    Hi Nic,

    The spec says optional for only the single-ended.

    Note that this is an internal spec and 50 or 200 ohms implies the device may not meet the signal integrity expectation.  However, the device specifies 100 ohms, meeting the spec, thanks.

    Best Regards,

    Michael.

  • 0 in reply to Michael Ikwuyum
    Prodigy 60 points

    Hi Michael,

    We tested both TCAN1462 vs TCAN1042 waveform at 5Mbps, using star topology. The topology has 6 stub and each stub is 10 meters. 

    However, don's see the waveform has any big ringing difference. 

    While TI's measurement has quite big difference on ringing.

    Do you know why here? The topology is not worse enough? 

    Regards

    Nic

  • 0 in reply to Zhihong Lin1
    TI__Guru 51106 points

    Hi Nic,

    Yes, your topology is not worse enough to stress the bus sufficiently to highlight the ringing differences and can further modify stub lengths and change termination to help bring out the differences more clearly, thanks.

    Best Regards,

    Michael.