（16） ADC conversion experiment

         This section is mainly about reviewing ADC The corresponding content of , Our chapter adopts a AD chip xpt2046 To read the change of external voltage , Display the digital quantity of voltage on the nixie tube .

About ADC：

         We all know that there are digital quantities inside the MCU , Namely 1 perhaps 0, And our current and voltage are analog quantities when they are transmitted , That is, the analog quantity is likely to be a dynamic continuous voltage , If we directly put this voltage on the MCU , Through our internal circuit transformation , More than how much voltage value is 1, Less than how much voltage is 0, But this does not indicate the specific size of the voltage value , So there is ADC To convert analog voltage value into multi digit digital voltage value , The more bits of voltage you convert , Explain that the more accurate your transformation . So for a ADC For example, there are several important indicators .1 It's resolution : Resolution is how many bits I use to express the output voltage ,ADC The resolution of is generally 8 Bits or 12 position , So 8 Bit and 12 Bit examples are converted at the same time 10v Voltage example , When we are 12 When a ：

        We say its resolution is 12 position , Each scale of it is 2.4mV, And when we are 8 When a ：

        At this time, we say its resolution is 8 position , Each scale of it is 39mV, It clearly verifies the previous conclusion ,12 Bit resolution is much higher .2 Is the conversion error , The conversion error we are talking about is half of our minimum scale , such as 12 Bit conversion error in conversion 10V The conversion error of time is 1.2mV.3 Is the conversion rate , That is, the number of conversions per second , And finish once A/D Time required for conversion , Is the reciprocal of the conversion rate .

About ADC transformation

        ADC The conversion is mainly divided into three steps , They are sampling , Quantization and coding . Sampling is mainly about explaining the sampling principle , The sampling principle mainly means that the sampling frequency must be of the signal frequency 2 times , But the sampling frequency should not be too large , Too large will lead to too many sampling points , The sampling time is too long , The ideal sampling frequency should be the signal frequency 3 To 5 times . And about quantification , We all know that changes in numbers are not continuous , So the size of any numerical quantity , Are expressed as integer multiples of a minimum unit , Therefore, when using digital quantity to express voltage , It must also be converted into an integral multiple of the smallest unit , This process is quantification . We express each sampling value of a continuous voltage as an integral multiple of the smallest unit , And the integer multiple value is encoded in binary , Then it's coding , This binary signal is us AD Output value .

Hardware ：

        XPT2046 It is a four wire resistive touch screen , It contains 12 Bit resolution AD converter ,XPT2046 There are many functions , But in this section, we mainly focus on the use of voltage detection , my 51 There is a sliding rheostat on the MCU , We can change the input voltage by adjusting the voltage value of the sliding rheostat , The figure is shown below ：

         Among them AIN0 That is, with our XPT2046 The input pin of is connected , About XPT2046 The hardware diagram of is as follows

               X,Y,Z,VBAT,AUX Enter after selecting through the on-chip control register ADC,ADC It can be configured as single ended or differential mode , choice VBAT,AUX perhaps X,Z Should be configured in single ended mode . As a touch screen application , You should choose differential mode , We can get it through the manual. We need to put A2,A1,A0 Configure to 001, Whole XPT2046 The sequence diagram of is as follows ：

                 Among them 7 The meaning of is as follows :

         We choose bit 7 by 1

         From the above list, we can know that the configuration signal we want to send is 0x94.

Software ：

        It's about XPT2046 The procedure is as follows , Although our voltage value is 5V, But we don't need to calculate the actual value of the voltage , We just need to get the actual number of copies , As for the follow-up treatment, we won't deal with it for the time being . It's about XPT2046 The reading and writing procedures are as follows ：

#include "xpt2046.h"
#include <intrins.h>
void delay_us(unsigned int xus){
	while(xus--);
}
unsigned int spi_read(){
	int i;
	unsigned int dat = 0;
	DCLK = 0;
	for(i = 0;i < 12;i++){
		dat <<= 1;
		
		DCLK = 1;
		DCLK = 0;
		dat |= DOUT;
			
	}
	return dat;	
}

void spi_write(unsigned char dat){
	int i;
	DCLK = 0;
	for(i = 0;i < 8;i++){
		DIN= (dat >> 7) & 0x01;
		dat <<= 1;
		DCLK = 0;
		DCLK = 1;
	}	
}

unsigned int read_data(unsigned char cmd){
	unsigned int val;
	int i;
	
	DCLK = 0;
	CS= 0;
	spi_write(cmd);
	for(i = 0;i < 6;i++);
	DCLK = 1;
	_nop_();
	_nop_();
	DCLK = 0;
	_nop_();
	_nop_();
	val = spi_read();
	CS = 1;
	return val;	 
}

        The procedure about the nixie tube is as follows ：

#include "key.h"
unsigned char smgduan[10]={0x3f,0x06,0x5b,0x4f,0x66,0x6d,0x7d,0x07,0x7f,0x6f};
void key_scan(unsigned int val){
	int i;
	
	unsigned char table[4];
	for(i = 0;i < 4;i++){
		table[i] = val % 10;
		val /= 10;
	}
	for(i = 0;i < 4;i++){
		switch(i){
			case 0:
				LSC = 0; LSB = 0; LSA = 0;
				break;
			case 1:
				LSC = 0; LSB = 0; LSA = 1;
				break;
			case 2:
				LSC = 0; LSB = 1; LSA = 0;
				break;
			case 3:
				LSC = 0;  LSB = 1; LSA = 1;
				break; 
		}
		P0 = smgduan[table[i]];
		delay_us(100);
	}
	
}

        The procedure of the main function is as follows ：

#include "reg52.h"
#include "xpt2046.h"
#include "key.h"
void main(){
	unsigned char cmd =0x94;
	unsigned int val;
	int i = 0;
	while(1){
		if(i==50){
			val = read_data(cmd);
			i = 0;
		}
		key_scan(val);		
		i++;		
	}
}

The code of the two header functions is as follows ：

#ifndef _XPT_2046_H_
#define _XPT_2046_H_
#include "reg52.h"
sbit DIN  = P3^4;
sbit DOUT = P3^7;
sbit CS   = P3^5;
sbit DCLK = P3^6;
unsigned int spi_read();
void delay_us(unsigned int xus);
unsigned int read_data(unsigned char cmd);
#endif

#ifndef _KEY_H_
#define _KEY_H_
#include "reg52.h"
#include "xpt2046.h"
sbit LSA = P2^2;
sbit LSB = P2^3;
sbit LSC = P2^4;
void key_scan(unsigned int val);
#endif

This completes XPT2046 Programming for .