• State
  • Locked Locked
  • Replies 5 replies
  • Subscribers 32 subscribers
  • Views 302 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

TCAN1146-Q1: sampling point difference with NXP device

Songzhen Guo
Intellectual 2770 points
Part Number: TCAN1146-Q1 
Hi Expert,

This is a urgent issue for customer. there are 1000pcs of TCAN1146-Q1 was used in the new project. Could you help on that? 

The sampling point of the CAN control module of the main CPU configured in the software is 81.25% (SP=81.25% tQ=16).
When the customer tested the CAN sampling point, it was found that the sampling of the TJA1145(NXP) board was 81.25%.
When TCAN1146 is tested at the same test environment. The sampling point of the TCAN1146 board was tested to be 87.5%. 
The only difference between these two boards is the CAN transceiver. Will your CAN transceiver affect the sampling point?
What we can do to optimize this? 
Here is the schematic for TCAN1146-Q1:
 

Best Regards
Songzhen Guo
  • 0
    TI__Guru 60002 points

    Songzen,

    The sampling point of a CAN system is based on loop delay and any possible ringing in the system; you must wait for the guaranteed time the bit will settle into stead-state. Can you elaborate on the test performed to determine the sampling point of each bit? Do you how their test determines what is the best sample point for the bit? I'm curious if it is based on ringing or loop delay, or something else entirely.

    Regards,

    Eric Hackett 

  • 0 in reply to Eric Hackett
    TI__Intellectual 2770 points

    Hi Eric,

    Thanks for your feedback.

    Here is the test result: Could you comment on that and give us suggestions? Thanks.

    Test Environment:

    VH6501 of VECTOR standard test scrip environment, it's common test environment of CAN communication

    After the power is stabilized, the equipment starts to test process. When the sampling point is 81.25% and non-standard wiring harness state. TCAN1146-Q1 will fail. And TJA1145 can pass.

    Best Regards

    Songzhen Guo

  • 0 in reply to Songzhen Guo
    TI__Guru 60002 points

    Songzen,

    Based on that information, it may be a ringing issue in the system. Is it possible to share waveforms of the test when the TCAN1146-Q1 is being used, and a schematic? If the schematic can't be shared on this public forum, it can be shared to my email, e-hackett@ti.com. One other point I noticed, when the sample point is set to 75% both devices work, but 81.75% does not work. Are these both the same exact tests with the sample points shifted?

    Regards,

    Eric Hackett  

  • 0 in reply to Eric Hackett
    TI__Intellectual 2770 points

    Hi Eric,

    Thanks for your feedback.

    If the waveform is not good enough, which part can we optimize? Could you give us suggestions? Thanks.

    I got some information from customer:

    1. For the incorrect result, the waveform is abnormal. Because the wire harness can cause those issue. But the wire harness is defined by end customer, they can's change this. Our customer's board is connected to a equipment, we can't get the waveform when it testing.

    2.Customer test two situation to isolate the power supply, the tests are all failed.

    one is TCAN1146DRQ1 is powered by 12V(Pin 10 is connected to 12V Power),

    another is TCAN1146DRQ1 is powered by 5V LDO (Pin 10 is connected to 5V).

    Here is the schematic and PCB Layout:

    a. TCAN1146DRQ1 is powered by 12V(Pin 10 is connected to 12V Power):

    b. TCAN1146DRQ1 is powered by 5V LDO (Pin 10 is connected to 5V)

    Best Regards

    Songzhen Guo

  • 0 in reply to Songzhen Guo
    TI__Guru 60002 points

    Songzhen,

    Thanks for this information, the power supply voltage shouldn't have a significant effect on the CAN bus waveform. And I understand that the cable harness is causing the abnormal waveform, what I don't understand is why the sample point of 75% works for both devices, but the 81% sample point doesn't work for the TCAN1146-Q1.

    Typically placing the sample point later in the waveform is more reliable because any ringing caused by cable parasitics will settle later in the waveform. This is why I was asking for waveforms of CANH, CANL and RXD: I am curious to see what is causing the earlier sample point to work but the later to not work for the TCAN1146-Q1. Are the tests the same between the 75% and 81% sample points? Or is something different?

    Regards,

    Eric Hackett