• State Resolved
  • Replies 3 replies
  • Subscribers 32 subscribers
  • Views 90 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

TCAN4550: Strange waveform on MOSI Line

Sukmin Hong
Prodigy 110 points
Part Number: TCAN4550

Tool/software:

Hello, TI Team.

I am working on TCAN4550, thanks to e2eTI Team.  I could make good system with it.

TCAN4550 is working well with with MCU, and there seems to be no problem, SPI, CAN is also working well.

.

But When I measured SPI by oscilloscope, there seems to be some strange waveform in MOSI>

.
Below is MOSI of my system captured by oscilloscope.

And I wonder what a triangle shaped waveform is.

.

There are no noisy signals around SPI, So I think it wouldn't be other noise.

SPI is working with 18Mhz, and below triangle 's frequency is about 180kHz.. Amplitude is about 1V.
.

Could you give me any idea about that? Or Have you seen any of these waveforms before?

.

I don't think that's what's affecting it since spi is working fine, but I'm worried that there might be some potential issue.

2.  And one more question.

Is damping resistor is essential? 

SPI is very short between MCU and TCAN4550 about 6~7mil.

I can't see the ringing, but I'll try put it if needed.

Thanks.

  • +1
    TI__Mastermind 25810 points

    Hello Sukmin,

    I have not seen this type of waveform before.  The MOSI is an Input signal to the TCAN4550, so the source of this noise may not be related to the TCAN4550.

    Do you see any type of noise on any other signals?  Can you capture the other SPI signals at the same time so we can see if there is any coupling or correlation between the other waveforms? 

    What does the power look like for VIO and VSUP?  Since this distortion is only on the LOW level, there could be noise on the GND or Power resulting in essentially a GND shift.

    All of the valid data bits appear to be in between the distortion pulses.  Do the data bits ever overlap the distortion pulses, and if so what does that waveform look like?  Does the low level of the data bit waveform follow the distortion pulse, or does it drive to full low?

    Given the noise pulse frequency is 180kHz and the bit rate is 18MHz, they are multiples of each other there may be a correlation.  If the noise pulses only occur between bursts of data bits, then it may be that the driver is transitioning the pin to Hi-Z between bursts of data bits causing it to slowly rise, an then the pin becomes driven Low at the beginning of the next data burst which is why it has a fast fall time.

    The serial dampening resistors are usually only needed for long SPI traces and may not be needed if the signal traces are very short.  If you saw a lot of overshoots on the rise and fall time transitions, then I would recommend them.

    Regards,

    Jonathan

  • 0 in reply to Jonathan Nerger
    Prodigy 110 points

    Hello, Jonathan.


    I will check everything you said.

    Check VIO, VSUP, and other waveform also.

    Thanks for your debugging guide. 

    But if I can't find anything with it, if then I may have to ask you one more time with the some more detailed data.

    Thanks!! :)

  • 0 in reply to Sukmin Hong
    TI__Mastermind 25810 points

    Hello Sukmin,

    Good luck in your debug efforts.  If you have any additional questions, please feel free to ask.

    Regards,

    Jonathan