• State TI Thinks Resolved
  • Replies 7 replies
  • Answers 1 answer
  • Subscribers 34 subscribers
  • Views 104 views
  • Users 0 members are here
Support feedback
Options
Options
Related

TCAN1145-Q1: Delay time of TX/RX

Ethan Liu
Intellectual 1415 points
Part Number: TCAN1145-Q1

Tool/software:

Hi team,

1. When using the TCAN1145DRQ1, I noticed a 200mV difference between the dominant common-mode voltage and the recessive common-mode voltage. Is this normal? What are the main reasons for this discrepancy between the dominant and recessive common-mode voltages? Is a difference of less than 500mV considered normal?

2. Is the TCAN1145's TX and RX delay time related to RL (termination resistor)? Does this delay time correlate with the length of the canL and canH traces? I've measured different delay times with different RL resistances, with a 30-40ns difference between 120Ω and 60Ω. Because the CANH and CANL traces are quite long (about 26cm), I suspect the delay time is related to the traces. Is there a specific linear relationship between the canH and canL trace lengths and the delay time?

RL = 60Ω, ch3 is tx, ch4 is RX, ch1 is canl, ch2 is canh

RL = 60Ω, ch3 is tx, ch4 is RX, ch1 is canl, ch2 is canh

BR,

Ethan

  • 0
    TI__Intellectual 1415 points

    Hi team,

    Any updates about this?

    BR,

    Ethan

  • 0 in reply to Ethan Liu
    TI__Guru 74252 points

    Ethan,

    Thanks for posting to E2E. 

    Ethan Liu said:
    1. When using the TCAN1145DRQ1, I noticed a 200mV difference between the dominant common-mode voltage and the recessive common-mode voltage. Is this normal? What are the main reasons for this discrepancy between the dominant and recessive common-mode voltages? Is a difference of less than 500mV considered normal?

    Are you only testing one device, or on a bus with other transceivers? Is the dominant common-mode voltage greater or lesser than the recessive common-mode voltage?

    Ethan Liu said:
    2. Is the TCAN1145's TX and RX delay time related to RL (termination resistor)? Does this delay time correlate with the length of the canL and canH traces? I've measured different delay times with different RL resistances, with a 30-40ns difference between 120Ω and 60Ω. Because the CANH and CANL traces are quite long (about 26cm), I suspect the delay time is related to the traces. Is there a specific linear relationship between the canH and canL trace lengths and the delay time?

    The delay time could be a result of both. With 120 ohms there is less current being driven and thus slower rise/fall times, which may result in delay. If the traces are the same between both measurements, it may explain the difference compared to datasheet specification, but should be the same difference between 120 ohms and 60 ohms. I think the main difference is the termination resistance.

    Is it possible to share a schematic and more waveforms of the common-mode voltage difference?

    Regards,

    Eric Hackett 

  • 0 in reply to Eric Hackett
    TI__Intellectual 1415 points

    Hi Eric,

    The schematic is shown as below, we are only testing one device. As you can seen in below wavefroms, the dominant common-mode voltage is greater than the recessive common-mode voltage after the shake. Could you help us analyze the possible reasons, and if it is accepetable for this 200mV difference?

    Also, is it normal for Vcm to jitter near the edge of CANH/L? Could you analyze the possible reasons and are there any solutions?

    And could you please tell me the ranges of the two parameters: common mode voltage step Vcm (step) and peak-to-peak common mode voltage Vcm(p-p)? We don't find that in the datasheet.

    (Yellow: CANH, Blue: CANL, Red: Vcm, Purple: VCC)

    For below waveforms, is it normal for CANH and CANL to jitter near the falling edge of RX? Could you analyze the possible reasons and are there any solutions?

    (Yellow: CANH, Blue: CANL, Purple: RX, Green: TX)

    BR,

    Ethan

  • 0 in reply to Ethan Liu
    TI__Intellectual 1415 points

    Hi Eric,

    Any updates about this, thanks for the support~

    BR,

    Ethan

  • 0 in reply to Ethan Liu
    TI__Guru 74252 points

    Ethan,

    Sorry for the delay, I am on business travel currently and Eric Schott is out of office. Please allow until Friday, 23/08/2025 for a response.

    Regards,

    Eric Hackett 

  • 0 in reply to Eric Hackett
    TI__Intellectual 1415 points

    Hi Eric,

    Thanks for your reply, because customer is test their borad and meet issues in TCAN1145-Q1 urgently, we sincerely hope we can receive your reply as soon as possible, appreciate!

    BR,

    Ethan

  • 0 in reply to Ethan Liu
    TI__Guru 53555 points

    Hi Ethan,

    Seeing a small amount of ringing on the common mode during state transitions is typical, especially when no CMC is present. This is in part caused by small variations in the CAN driver but also significantly by measurement setups. Scope probes, bus cabling, PCB routing and coupling can all cause this ringing to be exaggerated. This is similarly true for the digital signals which is why we can see the same effect on the RXD signal. Measuring the common mode shift in this way is not the most reliable way to predict emissions energies. This is why we instead measure emissions directly in our certification testing. This is also the reason why we do not specify behaviors such as common mode shift - because there are so many system variables that can impact this reading and ultimately where it is important (emissions) is evaluated through more direct means. 

    I don't see the waveforms here as any cause for concern. Is the customer currently experiencing an issue with the TCAN1145 in their system? Are they having trouble meeting any emissions requirements?

    Regards,
    Eric Schott