• State TI Thinks Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 1 answer
  • Subscribers 46 subscribers
  • Views 1121 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.

Original question:

[FAQ] TSC: Touch Coordinates are Unstable

TSC2046: Vref not stabilizing on TSC2046

Marten Beels
Prodigy 150 points
Part Number: TSC2046


Hello,

We have noticed that some of our products using the TSC2046 have an unresponsive touchscreen.  This seems to be because the detected coordinates are changing and not stabilized.  I have noticed that on the PCBs that exhibit this symptom, the 2.5 V Vref does not stabilize until 100- 200 uS after the touch was detected.  Some PCBs do not have this issue, and Vref is stable.  The problem follows the PCB (not the touchscreen).  We are not using any external capacitors or filtering on the touchscreen lines. 

Why would we see this variability and what can we do to improve the Vref stabilization time?

The 2.5 V Vref on a board with a responsive touchscreen:

Vref on a board showing the slow stabilization:

  • 0
    TI__Mastermind 22950 points

    Hello Marten,

    Can you please share a schematic as well as the commands sent to the device? An I2C transaction would be fine. 

    Regards,

    Aaron

  • 0 in reply to Aaron Estrada
    Prodigy 150 points

    Hello Aaron,

    Thank you for the reply, here is a portion of our schematic, and I will try to get some information on the commands from our engineers.

  • 0 in reply to Marten Beels
    Prodigy 150 points

    The commands we send are:

    The command we send are:
    - Read ADC Y: 10010011
    - Read ADC X: 11010011
    - Read ADC Z1: 10110011
    - Read ADC Z2: 11000011
    - ADC OFF: 10000000

  • 0 in reply to Marten Beels
    Prodigy 150 points

    In more detail:

    1. Read ADC Y up to 8 times, looking for 2 consecutive that are the same values
    2. Read ADC X up to 8 times, looking for 2 consecutive that are the same values
    3. Read ADC Z1 up to 8 times, looking for 2 consecutive that are the same values
    4. Read ADC Z2 up to 8 times, looking for 2 consecutive that are the same values
    5. ADC off
  • 0 in reply to Marten Beels
    TI__Mastermind 22950 points

    Hello Marten,

    The schematic doesn't show anything of concern and looks fine. Looking back at the scope shots provided in the initial post, why is Vref powered for less time in the second image? Could it be Vref is not on when trying to use the touch screen?

    Regards,

    Aaron

  • 0 in reply to Aaron Estrada
    Prodigy 150 points

    Thanks for the reply, Aaron.

    The Vref seems to be "on" for as long as the screen is pressed.  So with a longer touch, the scope shows Vref powered on for longer. 

    I have done some more testing in the meanwhile, and perhaps it is a small difference in noise levels.  Instead of requiring two consecutive reads to be equal, we are now allowing for two consecutive readings to be within 1 LSB of each other to accept a touch.  With that change, a previously unresponsive touch screen is now very responsive.  Perhaps the criteria for accepting a touch was too stringent, and some of our PCBs are slightly more susceptible to noise than others.  We'll continue testing this change, but I think it may work. 

    I still don't know why Vref seems to 'sag', but it seems to still work with a slightly looser touch acceptance criteria.  Maybe our solution is just a work around, but it seems to be ok so far. 

  • 0 in reply to Marten Beels
    TI__Mastermind 22950 points

    Hello Marten,

    Glad to hear the touch screen is more responsive now that the requirements are less stringent. I will go ahead and close this thread but feel free to comment back if more questions arise. 


    Regards,

    Aaron