• State TI Thinks Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 2 answers
  • Subscribers 47 subscribers
  • Views 845 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.

TSC2007: Non-responsive TSC

Bond - James Bond
Expert 5230 points

Part Number: TSC2007

Hello TSC2007 Experts,

I found several units with an unresponsive touch screen.

When inspecting the units I noticed I can correctly access the TSC2007 via I2C but I got an almost stable value for X,Y and Z no matter where I touch the screen.

After some testing it has started working and it still does.

 

I was wondering if our setup is missing some initialization commands.

Is there any setting for the TSC2007 which might have an effect BETWEEN power cycles? Any register which is kept in a NVM?

 

The touch controller is connected to a I.MX cpu running Linux. During my test I did not use the touchscreen linux driver but I accessed the controller registers via i2c bus.

  • 0
    TI__Guru 64755 points

    Hello,

    Can you clarify what updates were performed in your testing that removed the issues/challenges with the TSC2007?  You mentioned "After some testing it has started working and it still does."

    There is a defined POR sequence that must be followed between power-on/off cycling which may be causing some of the challenges you're seeing if it's not being followed.  There is not any NVM used to store register settings.  

  • 0 in reply to Collin Wells
    Prodigy 10 points
    Hello,

    we didn't any power cycle during our test. We only "played" with i2c commands for a while, trying different values in Command Byte.
    I cannot say precisely all the steps we did, but suddenly the touchscreen started to provide valid X&Y values.
    After this moment, the TSC2007 has kept providing correct information even after power cycles or with old source code.
    I2C runs at 85kbps.

    The VDD on ramp in our board is not satisfied (it's 1kV/s), but let me say 12kV/s is insane!
    VDD is by definition slow, as it's plenty of decoupling capacitors around the board.
    12kV/s requires an external logic to drive separately the TSC2007 power supply: basically we should create an externally POR circuit to drive TSC2007 internal POR..
    Is it possibile to force a software reset by I2C?

    Regards
  • 0 in reply to Riccardo Cusatelli
    TI__Guru 64755 points

    Hello,

    Hmm, that sounds pretty peculiar.  Usually something has to change for products to being working differently either in a more positive or negative manner.  Have you repeated the experiment and found similar results that initial readings are incorrect and then future ones suddenly become correct?  Perhaps you could capture an I2C sequence during the times when the product is behaving strange for us to look at.

    12kV/s in a more common unit for slew is only 0.012V/us which should be possible with most switching/linear supplies but I agree if you're charging a lot of capacitance through some cabling or additional series resistance then you may struggle to meet those limits.  In this case an additional POR circuit may be required as you've suggested to meet the requirements.

    There is not a way to SW reset the TSC2007 through I2C or any other commands/sequences.