delayMicroseconds to millis

notsolowki · April 4, 2018, 7:51pm

how can i re write this in millis so that its non-blocking i did the top part but i dont know what to do with the delaymicroseconds?

unsigned long WaterWait = 4000;
unsigned long Waterdelay =0;

int trig = 12;
int echo = 11;

void setup()
{
  Serial.begin(9600);
  pinMode(trig, OUTPUT);
  pinMode(echo, INPUT); 
}

void loop()
{
 if (millis() - Waterdelay > WaterWait) {
  long t = 0, h = 0, hp = 0;
  digitalWrite(trig, LOW);
  delayMicroseconds(2);
  digitalWrite(trig, HIGH);
  delayMicroseconds(10);
  digitalWrite(trig, LOW);
  t = pulseIn(echo, HIGH);
  h = t / 45;
  h = h - 5;  // offset correction
  h = 35 - h;  // water height, 0 - 50 cm
  hp = 2.7 * h;  // distance in %, 0-100 %
  Serial.print(hp);
  Serial.println(F("%"));
  Waterdelay = millis();
 }
}

system · April 4, 2018, 7:57pm

Is it critical to get rid of the very short blocking delayMicroseconds?
In fact, those delays are almost certainly shorter than pulseIn

notsolowki · April 4, 2018, 8:01pm

TolpuddleSartre:
Is it critical to get rid of the very short blocking delayMicroseconds?
In fact, those delays are almost certainly shorter than pulseIn

i dont know i have alot of other timers in the background this is just a small part of code that im going to integrate into my main sketch. so far i have come up with this,

unsigned long WaterWait = 4000;
unsigned long Waterdelay =0;
unsigned long WaterWait2 = 10;
unsigned long Waterdelay2 =0;
unsigned long Waterdelay3 =0;

int trig = 12;
int echo = 11;

void setup()
{
  Serial.begin(9600);
  pinMode(trig, OUTPUT);
  pinMode(echo, INPUT); 
}

void loop()
{
 if (millis() - Waterdelay > WaterWait) {
  long t = 0, h = 0, hp = 0;
  digitalWrite(trig, LOW);
 if (millis() - Waterdelay2 > WaterWait2) {
  digitalWrite(trig, HIGH);
 if (millis() - Waterdelay3 > WaterWait2) {
  digitalWrite(trig, LOW);
  t = pulseIn(echo, HIGH);
  h = t / 45;
  h = h - 5;  // offset correction
  h = 35 - h;  // water height, 0 - 50 cm
  hp = 2.7 * h;  // distance in %, 0-100 %
  Waterdelay2 = millis();
  Waterdelay2 = millis();
  Waterdelay = millis();
  Serial.print(hp);
  Serial.println(F("%"));
 }
 }
 }
}

system · April 4, 2018, 8:05pm

const unsigned long waterWait = 4000;
unsigned long waterdelay;

const byte trig = 12;
const byte  echo = 11;

void setup()
{
  Serial.begin(9600);
  pinMode(trig, OUTPUT);
  digitalWrite(trig, LOW);
  pinMode(echo, INPUT);
}

void loop()
{
 if (millis() - waterdelay > waterWait) {
  digitalWrite(trig, HIGH);
  delayMicroseconds(10);
  digitalWrite(trig, LOW);
  unsigned long t = pulseIn(echo, HIGH);
  unsigned long h = t / 45;
  h = h - 5;  // offset correction
  h = 35 - h;  // water height, 0 - 50 cm
  unsigned long hp = 2.7 * h;  // distance in %, 0-100 %
  Serial.print(hp);
  Serial.println(F("%"));
  waterdelay += waterWait;
 }
}

(I hate too much code) This code has not been compiled and it has not been tested.

system · April 4, 2018, 8:08pm

The code you posted won't work, and anyway you need to learn the difference between "millis()" and "micros()"

notsolowki · April 4, 2018, 8:11pm

TolpuddleSartre:
The code you posted won't work, and anyway you need to learn the difference between "millis()" and "micros()"

'
so there is really nothing i can do about the delaymicroseconds?

notsolowki · April 4, 2018, 8:12pm

TolpuddleSartre:
The code you posted won't work, and anyway you need to learn the difference between "millis()" and "micros()"

right now this is what im using

unsigned long WaterWait = 4000;
unsigned long Waterdelay =0;
unsigned long WaterWait2 = 10;
unsigned long WaterWait3 = 20;
unsigned long Waterdelay2 =0;
unsigned long Waterdelay3 =0;

int trig = 12;
int echo = 11;

void setup()
{
  Serial.begin(9600);
  pinMode(trig, OUTPUT);
  pinMode(echo, INPUT); 
}

void loop()
{
 if (millis() - Waterdelay > WaterWait) {
  long t = 0, h = 0, hp = 0;
  digitalWrite(trig, LOW);
 if (micros() - Waterdelay2 > WaterWait2) {
  digitalWrite(trig, HIGH);
 if (micros() - Waterdelay3 > WaterWait3) {
  digitalWrite(trig, LOW);
  t = pulseIn(echo, HIGH);
  h = t / 45;
  h = h - 5;  // offset correction
  h = 35 - h;  // water height, 0 - 50 cm
  hp = 2.7 * h;  // distance in %, 0-100 %
  Waterdelay2 = millis();
  Waterdelay3 = millis();
  Waterdelay = millis();
  Serial.print(F("The Water Level Is At "));
  Serial.print(hp);
  Serial.println(F("%"));
 }
 }
 }
}

system · April 4, 2018, 8:22pm

You're worried about a 12 microsecond blocking (my code blocks for 10 microseconds) every four MILLION microseconds.

That's not something you should be worrying about at the moment.

Given that the speed of sound is something like 8.5mm in 25 microseconds, pulseIn blocks for much longer than your delays.

Your revised code still does not work, and you have demonstrated that you still need to learn the difference between milliseconds and microseconds.

DKWatson · April 4, 2018, 8:23pm

Two things: you cannot delay for 2 microseconds or 10 microseconds.

The off-the-shelf micros() function has a resolution of 1/256th of about a millisecond, so pick 4, 8 or 12.

Usually for non-blocking you would mark the time and carry on, returning from time to time, via the loop, to see if a prescribed period has passed. You would do it with micros() the same way you would do it with millis(). In reality though how far are you going to get through your code before your time runs out? Is it worth it? Or just stick with your delays?

system · April 4, 2018, 8:25pm

The two microsecond delay is redundant anyway, as my code earlier demonstrated.

GrOnThOs · April 4, 2018, 8:55pm

 Serial.print(hp);
  Serial.println(F("%"));

This serial prints, blocking much more than 10 microseconds.

system · April 4, 2018, 8:56pm

GrOnThOs:

 Serial.print(hp);

Serial.println(F("%"));




This serial prints, blocking much more than 10 microseconds.

Not every four seconds it doesn't.

GrOnThOs · April 4, 2018, 9:32pm

This unnecessary six time calls of millis function blocking much more than 10 microsecond.

system · April 4, 2018, 9:36pm

GrOnThOs:
This unnecessary six time calls of millis function blocking much more than 10 microsecond.

Four calls to millis, that I can see in the most-recently posted code.

GrOnThOs · April 4, 2018, 9:54pm

Four calls to millis, that I can see in the most-recently posted code.

Six times to the first post but you lost my point.

system · April 4, 2018, 9:57pm

GrOnThOs:
Six times to the first post but you lost my point.

The first post has only two calls to millis - you need to pay attention to detail.
It's very important if you want to get anywhere in software.

oqibidipo · April 4, 2018, 10:18pm

DKWatson:
Two things: you cannot delay for 2 microseconds or 10 microseconds.

The off-the-shelf micros() function has a resolution of 1/256th of about a millisecond, so pick 4, 8 or 12.

Nonsense. delayMicroseconds() has nothing to do with micros().

github.com

arduino/Arduino/blob/1.8.5/hardware/arduino/avr/cores/arduino/wiring.c#L120


      
          	while (ms > 0) {
          		yield();
          		while ( ms > 0 && (micros() - start) >= 1000) {
          			ms--;
          			start += 1000;
          		}
          	}
          }
          
          /* Delay for the given number of microseconds.  Assumes a 1, 8, 12, 16, 20 or 24 MHz clock. */
          void delayMicroseconds(unsigned int us)
          {
          	// call = 4 cycles + 2 to 4 cycles to init us(2 for constant delay, 4 for variable)
          
          	// calling avrlib's delay_us() function with low values (e.g. 1 or
          	// 2 microseconds) gives delays longer than desired.
          	//delay_us(us);
          #if F_CPU >= 24000000L
          	// for the 24 MHz clock for the aventurous ones, trying to overclock
          
          	// zero delay fix

(rephrasing:)

void delayMicroseconds(unsigned int us)
{
    us = calculate_number_of_loops_based_on_F_CPU(us);
 
    __asm__ __volatile__ (
        "1: sbiw %0,1" "\n\t" // 2 cycles
        "brne 1b" : "=w" (us) : "0" (us) // 2 cycles
    );
}

GrOnThOs · April 4, 2018, 10:33pm

The first post has only two calls to millis - you need to pay attention to detail.
It's very important if you want to get anywhere in software.

Of course. six times in the second post but what i want to say is that anything more than one call is unnecessary and increase the code execut time.

notsolowki · April 5, 2018, 6:32am

amazing that you dont even need to run the code to debug it. i have been running it like this and its been working but ill take your word with the delaymicroseconds and im only going to be calling it every 30 seconds anyways i just didnt know if it was a bad idea. ive been running it without delaymicroseconds or any delay besides the 30 seconds delay and its been working. does it really need the delay? what was the point of using it?

unsigned long WaterWait = 30000;
unsigned long Waterdelay = 0;

int trig = 12;
int echo = 11;

void setup()
{
  
Serial.begin(9600);
 pinMode(trig, OUTPUT);
 pinMode(echo, INPUT); 
}
void loop()
{
  long t = 0, h = 0, hp = 0;
  digitalWrite(trig, LOW);
 if (millis() - Waterdelay > WaterWait) {
  digitalWrite(trig, HIGH);
  digitalWrite(trig, LOW);
  t = pulseIn(echo, HIGH);
  h = t / 45;
  h = h - 5;  // offset correction
  h = 35 - h;  // water height, 0 - 50 cm
  hp = 2.7 * h;  // distance in %, 0-100 %
  Waterdelay = millis();
  Serial.print(F("The Water Level Is At "));
  Serial.print(hp);
  Serial.println(F("%"));
 }
}

system · April 5, 2018, 8:47am

The point of not using delay is that it allows your sketch to do other things instead of locking up the processor, twiddling its metaphoric thumbs.
This is fine on the order of tens of milliseconds to seconds or even hours, but at the level of small numbers of microseconds, it doesn't make much sense.

Topic		Replies	Views
Avoiding Delay (blinkwithoutdelay) help Programming	23	4834	May 5, 2021
delay inside blink without delay Programming	13	1897	May 5, 2021
Non-blocking ultrasonic sensor without interrupts and timers? Programming	10	5498	May 5, 2021
Sonar Sensor Trouble incorporating millis General Guidance	57	4745	May 5, 2021
PING))) Per Arduino Italiano	14	4947	May 6, 2021

delayMicroseconds to millis

Related topics