• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 19 subscribers
  • Views 226 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.

TXS0102: Jittern issue on the Falling edge of the input pulses

SIDDHARTH ROY
Intellectual 510 points
Part Number: TXS0102

Tool/software:

We are facing a weird issue with one of the Flow meter sensor output signal. Below is the description and I need someone to suggest a solution.

 

  1. We have interfaced the flow meter output with our on-board Microcontroller. The flow sensor  output is pulsed waveform whose frequency is relative to the flow rate.
  2. Our microcontroller works on 3V3 logic Voltage and due to this we had to use a Level shifter between the Sensor output and MCU input pin.The sensor output is an open drain, hence we had pulled up the output to 5V using 10K resistor.TP26 is 5V logic input to the Level shifter IC and TP27 is 3V3 logic output of the IC. Please see attached Flow sensor schematic snapshot.


  3. We are using TI IC TXS0102DUCR Level shifter IC which is generally used for open drain applications.
  4. Below is the screenshot of the of one of pulse’s Falling edge from the sensor output which is input to the Level shifter IC U4 ( TP 26). Please note that this is when the sensor output is connected to the level translator. When we check the output in isolation from the translator, the below behaviour is not seen.


  5. When we zoomed more on the red indicated area on oscilloscope , we are finding multiple unwanted pulses which lasts for few uS.See below image for reference:


  6. Because of the above behaviour, our microcontroller is receiving multiple interrupts causing flow rate calculation to be faulty.
  7. We also tested our Level shifter circuit with a function generator input and found that the circuit works perfectly well.
  8. We have also checked by removing the external pullup on the 5V rail but this behaviour is still evident.
  9. When we isolate the output of sensor from the level translator, this behaviour is not there. 

We would like to understand what can cause this behaviour?  We are able to see the proper pulse train, the only issue while using the level translator is that,  at the 5v logic input side itself, these additional jitter is appearing with every falling edge of the pulses.

Regards,

Siddharth

  • 0
    Guru 298810 points

    The TXS is a passive switch; its edge accelerators cannot keep the signal high for more than a few nanoseconds. This looks as if the sensor does not actually pull the line low at all times.

    Please show an oscilloscope trace with both A1 and B1.

  • 0 in reply to Clemens Ladisch
    Intellectual 510 points

    Hi Clemens,

    I currently do not have the sanpshot of the A1. But when I checked the waveform, it was replica of the B1 with voltage level changed to 3V3. So , A1 is refelecting whatever is on the B1 side. The issue is that unwanted pulses on the falling edge of the B1 side . Do you think it has anything to do with the edge accelators? is there any false triggering? Currently the sensr cable is approx 10cm. Do you think I should try reducing it.
    The same sensor with the 10cm cable works with Arduino board which is durectly a 5V logic.

    Regards,

    Siddharth

  • 0 in reply to SIDDHARTH ROY
    Guru 298810 points

    The edge accelerator could pull the line up, but not down. It would be possible for ringing to cause false triggering, but you have relative long time intervals where nothing pulls the line down.

    There is some ringing on the falling edges, but your problem has a much lower frequency.

    Please note that the ringing happens at each of the falling edges. This shows that the sensor stops and starts driving the line low.