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.

TCAN1043-Q1: rise time and fall time spec

Part Number: TCAN1043-Q1
Other Parts Discussed in Thread: TCAN1043

Hi,

   About rise time and fall time of TX,RX,CANH,CANL, one customer ask to follow the spec related to data rate. That is <10% of the period(1/data rate). When data rate come to 5Mbps, the CANH and CANL can not meet the requirement which is 20ns no matter how we adjust the circuit.

   Is this request reasonable?

   And is the device possible to drive CAN bus that fast?

  • Hi Zehui,

    The rise/fall times are specified as tR and tF, in datasheet which has a typical value of 45 ns. This was measured on a simulated differential bus load of 60-ohms and 100pF. No cabling, secondary transceivers, or other external loads were applied to the circuit. We do not specify individual signal rise/fall time (CANH alone or CANL alone), only that of the differential signal. 

    As we do not specify maximum and minimum values for these characteristics on the datasheet, it is difficult to point to these measurements as a discrepancy for the device. But usually we'll see a minimum tR of 28 ns and a minimum tF of 29 ns, thus I don't think 20 ns rise/fall time is possible for this device. 

    Regards,

    Sean

  • Hi Sean,

        On waht condition can you measure minimum tR of 28 ns and a minimum tF of 29 ns ?

        Do you know any type of CAN transceiver can drive the edge that fast?

  • The condition is just what I described above, same as the condition in datasheet. These are not required parameters for CAN, so we don't specify min/max, only typical for reference. The differential signal rise/fall time is not critical to CAN since the thing that really matters for arbitration is loop delay.

    May I know why customer has that requirement on rise/fall time? 20ns is in the range of SIC transceivers so if this requirement is to stand they would need to move to TCAN1463, which has a 25 ns typical value.

    Regards,

    Sean

  • Hi Sean,

        In fact, I don't really think there is a strong reason for that requirement on rise/fall time. They require that as a matter of experience. Normal signal have the 10% of period rise/fall time. I wonder if this is the suitable when it comes to CAN. Or do you ever encounter such requirement?  

  • Hi Zehui,

    Actually that's a ridiculous requirement to us because these are not required by the CAN specification. Rise and fall time are the time it takes for the differential voltage to transition from 10% to 90% of its final dominant value, but the CAN standard never specified them to meet 10% of the period. Is there a specific concern that the customer has regarding this behavior and how it impacts their system? Or are these parameters required for a different part of their specification? 

    Would you please send me an email to s-guo2@ti.com regarding this issue.

    Regards,

    Sean

  • Hi Sean,

        I don't think it's a very reasonable requirement either. Because it's confidential, I can't show the original document. But a few customers do require a short Tr/Tf time like 20ns, 25ns, 30ns for no good reason. Since you and I don't approval it, maybe I could argue with them.

  • Hi Zehui,

    Thanks for your understanding! As I said if the customer definitely requires 20ns tr/tf value, I would recommend to move to TCAN1463 which is a CAN SIC transceiver. It is for sure that TCAN1043 won't meet this requirement no matter how you adjust the circuits.

    Also if they still don't agree with us please send me an email directly so we can provide further support and let my team know this issue.

    Regards,

    Sean