›

ADC Keypad Due+RADDS

GRB352GRB352
in Repetier-Firmware
I had to make a few changes to the ADC keypad code from CONTROLLER_ZONESTAR and I wanted to see if you thought is would cause any problems (delays, etc) .  I'm trying this because I can't get the regular 4x4 matrix to work reliably on my Due+RADDS using any combination of available pins. (I get Phantom/ghost button presses, when no button has been pressed). The 4x4 matrix worked fine on the RAMPS, but I think maybe having 3.2v ref instead of 5v made matrix key press detection problematic.

Basically I just commented out the way it got the value for ADC and replaced it with an analogRead() because I wasn't getting a value from the old code....then I changed the rest of the code to work for the scale of the value I was getting (max of 1023, not 4000).

to see the value for each pressed button I temporarily uncomment out the part that serial prints the ADC value. 

inline void uiCheckSlowKeys(uint16_t &action) {
   struct {
    uint16_t min;
    uint16_t max;
    uint16_t action;
    } keys[] = {
     {   0,   15, UI_ACTION_IDENT01 },   
     {   160,  180, UI_ACTION_IDENT02 },   
     {  302,  322, UI_ACTION_IDENT03  },   
     { 417,  437, UI_ACTION_IDENT04 },
     {   510,   530, UI_ACTION_IDENT05  },   
     {   590,   610, UI_ACTION_IDENT06  },   
     {  655,  675, UI_ACTION_IDENT07 },   
     { 715,  735, UI_ACTION_IDENT08 },  
     { 765,   785, UI_ACTION_IDENT09  },   
     { 811,   831, UI_ACTION_IDENT10  },   
     { 850,  870, UI_ACTION_IDENT11 },   
     { 885,  902, UI_ACTION_IDENT12 },  
     { 915,   935, UI_ACTION_IDENT13  },   
     { 945,   965, UI_ACTION_IDENT14  },   
     {  971,  991, UI_ACTION_IDENT15 },         
     {  999,  1015, UI_ACTION_IDENT16  }   

  };
  const uint8_t numOfKeys = sizeof(keys) / sizeof(keys[0]);

  extern volatile uint16_t osAnalogInputValues[ANALOG_INPUTS];
//  uint16_t adc = osAnalogInputValues[KEYPAD_ANALOG_INDEX] >> (ANALOG_REDUCE_BITS);
 uint16_t adc = analogRead(ADC_KEYPAD_PIN)  ;


//  if (adc < 4000) {
   if (adc < 1016) {   
    for (int8_t i = 0; i < numOfKeys; ++i) {
      if ((adc >= keys[i].min) && (adc < keys[i].max)) {
        action = keys[i].action;
 //  Serial.print("ADC V:") ;
  //   Serial.println(adc);
        return;
    } else {   
  //  Serial.print("ADC X:") ;
   // Serial.println(adc);
      }
    }
  }  
  }

Comments

  • GRB352GRB352
    Oh, I also had to use a diff pin for ADC_KEYPAD_PIN, pin5 wasn't working reliably so I'm using TEMP_4_PIN. 
  • RepetierRepetier
    Using analogRead is strictly forbidden. We read adc inside a timer loop so this breaks all other adc readings hemce the access over the analog read array we read.

    temp4 pin is only ok if it is a thermocouple pin, normal temperature pins have a 4.7k resistor connected to pin so changing the result.
  • GRB352GRB352
    ok, so what do I need to do to fix it?
     uint16_t adc = osAnalogInputValues[KEYPAD_ANALOG_INDEX] >> (ANALOG_REDUCE_BITS);
    doesn't give me a value if I:
    Serial.println(adc);


    also, if I use the ADC_KEYPAD_PIN with analogread() I get strange values starting off with a solid 781 until you hit the buttons on the keypad for a while. So when I hook my ADC keypad up to the temp pin instead I get a solid 1023 right off the bat with analogread() and all my key presses come up with consistent values (also with analogread() )

    if analogread() will break other readings then thats the sort of thing I was worried about. what do I need to do to get this working?:
     uint16_t adc = osAnalogInputValues[KEYPAD_ANALOG_INDEX] >> (ANALOG_REDUCE_BITS);


    I suppose I could be getting a reading that Serial.print might not be able to display, but if that is the case then I wont know what values to map to my key presses.
Sign In or Register to comment.