Limit a fade that uses a cosine function

Projects Programming
1

Hi everyone, quite a novice coder here. I'm working on a project for a physical computing class that has been throwing me for a loop for a week or so now. I'm trying to use cap touch sensors to individually trigger some LEDs and have them fade from low-high-low in a 5 second time span. What makes this code tricky is that I'm trying to fake multi-threading so I can trigger more than one fade at the same time when the sensor is tapped. I've figured out how to keep the fade going through the five second interval using a cosine function and some annoying boolean code.

int value, value2 ;
int ledpin = 2;                           
int ledpin2 = 6;                           
long time=0;

int periode = 5000;
int displace = 500;

int cap1 = A2;
int cap2 = A3;
int cap3 = A4;
int cap4 = A5;
int cap5 = A6;

boolean button1 = false;
unsigned long currenttime1 = 0;

void setup() 
{ 
  Serial.begin(9600);
pinMode(ledpin, OUTPUT);
pinMode(ledpin2, OUTPUT);
  pinMode(cap1, INPUT_PULLUP);  
  pinMode(cap2, INPUT_PULLUP);
  pinMode(cap3, INPUT_PULLUP);
  pinMode(cap4, INPUT_PULLUP);
  pinMode(cap5, INPUT_PULLUP);
} 

void loop() 
{ 
 time = millis();
 value = 128+127*cos(2*PI/periode*time);
 value2 = 128+127*cos(2*PI/periode*(displace-time));

if(millis() > (currenttime1 + 5000)){
  button1 = false;
}
if(button1 == false){
  if(digitalRead(cap4) == HIGH){
    button1 = true;
    currenttime1 = millis();
    }
  }
if(button1 == true){
analogWrite(ledpin, value);           // sets the value (range from 0 to 255) 
}

My problem is, since this cosine equation uses the millis() function, it will start to fade from some random value in between 0-255 because of the millis. I would really like this to start from 0 and ramp to 255 and then back to 0 again. Is there any way I can constrain these values to only begin the fade starting at 0?

Thanks for your help everyone.

2

I would really like this to start from 0 and ramp to 255 and then back to 0 again.

So, what role with cos() play in this scenario? I can't really imagine a relationship between angle (what cos() expects) and time (what millis() reports). If there really is some relationship, you could always base the angle on now minus when the fading started. Since fading starts when now equals fading started, the difference would be 0, which is the value that you want to start at. Except that cos(0) is 1.0, not 0.0...