SparkFun Forums

[tah997]

G'day all,

has anyone had any issues using the sprintf() function when compiling with WinAVR. I try to compile with a float in the data format and WinAVR gives me a warning saying that the compiler was expecting a double instead it has been presented with a float, however it does compile. When i run the prog it doesn't give any value.

I can verify that the sprintf() function does work, cos when you change the appropriate data type values for an int, it works like a charm.

The program is for a simple voltmeter for an Atmega8. it uses ADC0 and the LCD is connected to PORTD.

Code: Select all

#include <stdio.h> 
#include <stdlib.h>
#include <avr/io.h>
#include "lcd.h"
#include <util/delay.h>


#define F_CPU 16000000UL  // 16 MHz

void delay_ms(uint16_t x); //General purpose delay


int main(void)
{

	float 	adc_data = 0,
			volts = 0;

	char 	lcd_buff[20];
	
	int j =0;
	

	DDRD = 0x00;	//makes all outputs
	PORTD = 0x00;


	/* initialize display, cursor off */
	lcd_init(LCD_DISP_ON);

	/* clear display and home cursor */
    lcd_clrscr();
		
	delay_ms(3000);
		
	ADCSRA = 0b11100111;
		
	ADMUX = 0b00000000; //0b 00000000
		
		
		while(1==1)
		{
				
				ADCSRA=0b11100111; //0b 10001110
				
				ADMUX=0b0000000;

				delay_ms(2000);
				
				adc_data = ADC;
				
				//j = ADC;
				
				//adc_data = adc_data*5/1023; //convert into volts
				
				//adc_data = adc_data - 1.650; //shift midpoint
				
				lcd_clrscr();
				lcd_gotoxy(4,1);
				
				sprintf(lcd_buff,"%1.4f",adc_data); 
				//sprintf(lcd_buff,"%4d",j);

				lcd_puts(lcd_buff);
				
				lcd_gotoxy(8,1);
				lcd_puts("mV");
		}

}


void delay_ms(uint16_t x)
{
  uint8_t y, z;
  for ( ; x > 0 ; x--){
    for ( y = 0 ; y < 90 ; y++){
      for ( z = 0 ; z < 6 ; z++){
        asm volatile ("nop");
      }
    }
  }
}

Any suggestions or info from people who have used sprintf() successfully with floats would be most appreciated.

Cheers,

Timmy

[seulater]

just for curiosity sake, remove the formating and try this and see if you get any different results.

sprintf(lcd_buff,"%f",adc_data);

[n1ist]

I don't have winavr on this machine so I can't check, but I seem to remember there's a setting in the makefile for which version of printf and sprintf to use. By default, it uses a simplified printf to save space. That version probably doesn't know about floats. You may have to add the math library too (-lm to LDFLAGS)

/mike

[tah997]

n1ist,

changing the printf() info in the makefile from simplified printf to the more advanced printf solved the issue. Works a charm now.

cheers,
Timmy

[stevech]

yes, the standard libraries don't support printf() for floats - to save space in flash.

[Jeff]

Hello, can you tell me how you changed the printf() to the more advanced to allow floats? I cannot find it in AVR Studio.

Thanks

[Liencouer]

from http://www.nongnu.org/avr-libc/user-man ... stdio.html might be what you want...

If the full functionality including the floating point conversions is required, the following options should be used:

-Wl,-u,vfprintf -lprintf_flt -lm

[stevech]

floats, doubles. Try to avoid and use ints and longs. Or longs with a fixed implicit decimal point, such as 1.54 scaled up by 100 to be an integer.