Post Go back to editing

ADAU1442 GPIO output is always high

Hi,

I simply want to toggle MP0 pin high (+3.3V) and low (0V) from Sigma Studio. I have connected a voltage meter between Pin 2 (gnd) and pin 3 (MP0) on the ADAU1442 evaluation board 50-pin header but no matter what I do, the MP0 pin stays high (+3.3V).

Am I doing something wrong? 

Earlier I successfully used one of the MP pins as input and it worked fine. But I cannot get any of the MP pins to work as an output, they stay fixed at +3.3V.

Best regards,

Kim



Changed wording.
[edited by: HifiDSP at 6:39 PM (GMT 0) on 21 Nov 2019]
  • Hello Kim,

    ( I like it when people put their names at the bottom of the post. It is more friendly to 

    me But I am kind of old fashioned) 

    I actually pulled out my eval board and also the GPIO board and I had no problems at all getting MP0 to function as an output. I used the same settings as you did. So I have no idea why you cannot get it to work. 

    I looked at Pin 3 and I also looked at the MP0 test point that is nearby. You do not need to use the internal pull-up since there is an external one but it does not cause it to not work. I did try that and I attached the GPIO board and set the MP0 functions on that board and I could not replicate your problem. So there has to be a simple reason. I suggest you try a new project from scratch that is very simple and see if it works. 

    What version of SigmaStudio are you using?

    Are you using I2C? Place a DC cell in your project and run it to a readback cell to be certain you are actually talking to the PC? Do the meters in your project work?

    Dave T

  • Hi Dave,

    I also like it when people put their real names at the bottom.

    I am really glad to get your support on this issue and hope you are right that it is something simple.

    Sigma Studio 4.5 and I2C. I have placed a readback cell in the project and verified that the connection to the PC is working by reading back the value every time I toggle the switch. 

    The project file: gpio_out_test.zip

    I have connected the voltage meter like this (black to gnd/pin2 and red to MP0/pin3):

    Sometimes after uploading the project the voltage is 0 V and other times 3.3 V. When toggling the switch, I can successfully read back the correct value (1 or 0) but the voltage displayed at the voltage meter does not change. Am I doing something wrong when measuring the voltage this way?

    I assume it is unlikely that there is something wrong with the DSP since it works fine otherwise. 

    Best regards,

    Kim T

  • Hello Kim,

    That is a very good test with the readback. So I am assured that you are actually communicating with the part. 

    I have no idea how you got that project file onto the post but I cannot download it. You have to Zip it first then attach it to the post. It is a hassle that I deal with daily! 

    I cannot see what you are doing wrong. Try touching the test point that is right next to the connector. The MP0 test point. There is only a resistor between the two points but you never know what could have failed. 

    Try MP1. See if you get the same results. I do want to try your program...

    Dave T

  • Hello Dave,

    I have uploaded the project file as zip now (edited my previous post).

    I have found the reason why the voltage did not change and it is very simple. The problem was the pin numbering. I should have checked it more carefully before posting - I am very sorry about that. When outputting on GPIO_0 in Sigma Studio, the physical output is on MP2. I still find that a bit strange but it works. From the GPIO expansion board schematics document and old pictures of the board:

    Have you seen this mismatch before? Maybe it has something do do with the revision (ADAU144XEB REV B) of the old board.

    Kim T.

  • Hello Kim,

    Ah! I have a Rev C board so you have an older revision. So I went into our files to look at the schematic to see if it is different and I found this list of known errors. So I think you should have this information:

    Known errors in EVAL-ADAU144XEBZ Rev B:

     

    J20-3 Should be MP0 – is MP8 and MP2

    J20-5 Should be MP1 – is MP1 w/o 100 ohm isolation resistor, MP9 and MP3

    J20-7 Should be MP2 – is MP0

    J20-9 Should be MP3 – is MP3 w/o 100 ohm isolation resistor and MP1 and MP9

    J20-11 Should be MP4 – is MP4 w/o 100 ohm isolation resistor

    J20-13 Should be MP5 – is MP5 w/o 100 ohm isolation resistor

    J20-15 Should be MP6 – is MP6 w/o 100 ohm isolation resistor

    J20-17 Should be MP7 – is MP7 w/o 100 ohm isolation resistor

    J20-19 Should be MP8 – is MP8 w/o 100 ohm isolation resistor and MP2

    J20-21 Should be MP9 – is MP9 w/o 100 ohm isolation resistor and MP1 and MP3

    J20-23 Should be MP10 – is MP10 w/o 100 ohm isolation resistor

    J20-25 Should be MP11 -  is MP11 w/o 100 ohm isolation resistor

     

    So this is the list. I do not understand when it says "is MP8 and MP2"? Does this mean that both of the MP ports are shorted together? 

    I think I need to get you the schematic for the exact board you are using. I looked up in our files and I am not 100% certain which file I need to send. So can you give me the data code on the PCB? This will be a number listed directly below the part number of the board. It is YYMMDD_x format. 

    Dave T

  • Hi Dave,

    Thanks again. Do you have a document with all the known errors for this board (in case there are more errors than the ones related to the MP pins)?

    I have uploaded some old pictures of the board (taken before I removed the dust with compressed air :-) 

    ADAU144x.zip

    On my board MP0 and MP8 are connected through a 100 Ohm resistor. I.e. when I measure the resistance (when board does not have power) between pin 3 and 19, I get 100 Ohm.  

    Kim T.

  • Hello Kim,

    Thanks for the pictures. This board is from before we started putting on a date code on the silk screen that I can reference to the layout file. When I get back in the office tomorrow I will look through the files to see one that has the silkscreen like this board. 

    Regarding a complete document, that is what I pasted into my last post. If I find other details I will let you know. 

    Dave T