Actions
Forum Thread Details
  • State Not Answered
  • Replies 5 replies
  • Subscribers 266 subscribers
  • Views 187 views
  • Users 0 members are here
Related
See a helpful answer?

Be sure to click 'more' and select 'suggest as answer'!

If you're the thread creator, be sure to click 'more' then 'Verify as Answer'!

MikroBus GPIO

MATRIX7878

Hello,

     I am using a ZuBoard 1CG and am trying to use the MikroBus as GPIO pins so I can program a parallel LCD screen.  I am getting strange signals coming out of the pins.  Even just alternating GPIO signals creates strange results.  There are large multi-second gaps for some lines, some have random switches and some keep running forever despite a stop condition.  From the schematic, the only IC between the SoC and MikroBus connector is a 1.8 V to 3.3 V converter.  Why would I have strange behavior?

Thank you

  • 0

    It sounds like an open connection somewhere, possibly an 'enable/disable' pin or the like. Or perhaps if your cables are too long and/or are not screened and some interference is getting onto them from some distance away.

    The power supply could be responsible - I don't remember 1.8v to 3.3v converters, but that doesn't mean there aren't any. If your supply voltage is not close enough to 1.8v then the output might not be close enough to 3.3v to drive the circuit. Can you short all the inputs to zero volts and then check to see if all the outputs are stable? Assuming that none of them are square-wave generators by default when power is applied, of course.

    I don't know the ZuBoard, but the more I think about it the more I think there is a dry joint somewhere or even an unsoldered pin. I've seen it before, a pin on a SMD in mid-air, i.e. not soldered to the PCB. Or, if it is a very old ZuBoard, there might be dendrites. I've even seen a PCB where the track layout was wrong close to an IC in such a way that pin 1 was connected to hole 16, pin 2 to pin 15, etc. The quick cure was to desolder the IC and solder it to the other side of the PCB; this was not acceptable to the quality inspectors but it proved the point. The other 99 boards were scrapped without any further work being done on them.

  • 0

    There is no enable/disable pin and the cables connected to MikroBus and the LCD are no longer than three inches.

    This is the converter:

    image

    I cannot short any of the connections for the MikroBus as they go straight from the SoC, to the voltage converter, to the MikroBus.  I have measured the voltage going into the converter and it measures 1.8V.  The VDD_CLICK voltage is 2.5-3.0V.  None of the pins are generators, they are all I/O.

    Is this the joint you are referring to?

    image

    If so, they stopped soldering on that part it seems.  Other than that, there are no SMD midair pins.  The board I have is less than two months old.  I would highly doubt that they would still have that same mistake now a days.

    I do appreciate your efforts, thank you.

  • 0

    Have you checked your design and outputs using the integrated logic analyzer (ILA)? It's easy to add an ILA to your design, and you'll know for sure whether it's a digital design error or a board error.

  • 0 in reply to javagoza

    No, I have not.  I did not know that was a thing to be honest with you.  Xilinx SoC is my 1st SoC and first AMD product.  I have only used GoWin and Quartus before and GoWins ILA is absolute garbage.  I will most certainly do that.  Thank you.  I will post my findings.

  • 0 in reply to javagoza

    As with anything I do, I follow an example and it does not work.  I cannot get an ILA in my design, no matter what I try.  Is this not correct:

    image

    I even instantiated it with a component and this line in my top-level file:

    image

    Can I not connect the ports to an ILA?  Thank you.