• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Subscribers 33 subscribers
  • Views 404 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.

TCAN4550-Q1: About the waveform disturbance due to the common mode choke coil.

Kengo.Y
Guru 11255 points
Part Number: TCAN4550-Q1
Other Parts Discussed in Thread: TCAN4550

Hi team,

Could you give me your advice about the waveform disturbance due to the Common Mode Choke Coil（CMCC)?
*The waveform becomes clear when the CMCC is removed.

My customer use ''DLW32SH510XK2''.(https://www.murata.com/en-us/products/productdetail.aspx?partno=DLW32SH510XK2%23)


*Schematic


*Waveform
[1]
IC3_CANH/L : Input
CN5_CANH/L : Output


[2]
IC3_CANH/L : Output
CN5_CANH/L : Input



I think this phenomenon is causing the CMCC.
What kind of SPEC do you think is influential?
If you have recommended coil, please let me know.

And then, about waveform No2, I think the undershoot is large at rises up.
Why does it happen?

Sincerely.
Kengo.

  • 0
    TI__Guru 53575 points

    Hi Kengo,

    This looks like common mode ringing on the bus due to some resonance in the system. This is common when significant parasitics (capacitance from bus caps and cabling, inductance from CMC) exist on a long or complex bus. From your description, it sounds like the inductance from the CMC is significantly contributing to this. The effective ringing can be suppressed by reducing the parasitic sources in the network, simplifying bus layout, and ensuring optimal termination selection and placement. 

    Could you please elaborate on the difference between the IC3 and CN5 CAN channels? Based on the schematic, it looks like TCAN4550 is driving the clean CN5 channel. Is the IC3 channel also driven by a TCAN4550? How far apart are the nodes that are being teste? Could We see the IC3 channel both sending and receiving during the same transition (currently one D>R, one, R>D)?

    I believe TCAN4550 was evaluated the ACT45B-101-2P-TL003 CMC as is shown in the EVM BOM. This appears to be a different value (100uH) than the one in your testing (51uH). A higher inductance may be able to help filter out this ringing. 

    I noticed that 60-oohm termination (R6+R7) is used on this node. Is this the only place in the system where termination is placed? If so, this may subject the system to excess ringing as the signal reflects off unterminated endpoints. Ideally, 120-ohms of termination should be placed at each endpoint of the network. 

    The dip seen in the second figure is most commonly cause by asymmetry somewhere in the CAN system - the CAN transceiver, CMC, routing, and bus are all potential locations this asymmetry can occur. Is this also seen without the CMC? Are the CAN traces on the board routed as differential pairs (same length)? 

    Regards,
    Eric Schott

  • 0 in reply to Eric Schott1
    Guru 11255 points

    Hi Eric,

    Thank you so much for your reply.

    I understand that this issue is due to the common-mode choke coil.
    IC3 and CN5 are driven by TCAN4550.

    By the way, I'd like to know about input signal regulation.
    According to the datasheet, I think the input signal is a regulation of differential voltage.
    In other words, is there any problem if the differential voltage is within the specified range no matter how much the Dominant(0) signal changes?


    Is my understanding is correct?
    0(Dominant) : There is difference voltage of 0.9V(Min)
    1(Recessive) : There is difference voltage of 0.5V(Max)

    Does VIT(ENdom) mean that the input regulation for Wakeup signal from sleep mode?


    Sincerely.
    Kengo.

  • +1 in reply to Kengo.Y
    TI__Guru 53575 points

    Hi Kengo,

    Your understanding on these thresholds is correct. As long as the differential signal can meet these thresholds, the receiver will be able to interpret the data on the bus. The common-mode voltage need only stay within the specified range (+-12V). Because of this behavior, the above pictured noise on the common-mode does not compromise the data in any way, as the differential signal remains in tact. However, This ringing would contribute to emissions as there is no differential equivalent to cancel any radiated energies. Because it's typically the goal of a CMC to reduce radiated emissions and increase signal symmetry, it's not desirable that this choke appears to make both of these matters worse. 

    Have you been able to test with a different CMC to see if we see similar results? Do you know how complex this CAN network is? Number of nodes, spacing between, and length of the bus?

    Regards,
    Eric Schott