Wednesday, 19 October 2011

PIC Timer 0 Tutorial


PIC Timer 0


















TIMER0 is usually 8 –bit, 16 – bit or 32 – bit and can do the following functions:
  • "Speedometer" counts the events when the external input clock (pulses which are counted) by the pin RA4/T0CKI.
  • Counts and generated the interrupt or trigger any output of the required time (which can perform by the internal pulses are counted) is the internal clock of the PIC Microcontroller is (system clock divided by 4).
The prescaler can be used by the timer 0 to divide the input by 2, 4, 8, 16, 32, 64, 128, or 256.
Example:
When the prescaler is set to divide by 2, there is 1 output pulse for every 2 pulses which are input to the prescaler.
If set to 256, there will be 1 output pulse for every 256 pulses inputted.
So, the overflow time of TMR0 can be made longer by setting the value of the prescaler.






Prescaler can be configured from bits and T0PS0 T0PS1 and T0PS2 register T0CON. The Prescaler, which is at the center of timer, can be used for either the TMR0 or WDT. The figure above shows the prescaler connected to TMR0.
OPTION REGISTER



T

Initializing the OPTION_REG register
The following is an example how we can initialize the OPTION_REG:
PSA=0; // Prescaler is assigned to the Timer0 module
PS0=1; // Prescaler rate bits
PS1=1; // are set to “111”
PS2=1; // which means divide by 256
TOSE=0; // rising edge
TOCS=0; // Internal instruction cycle clock

The PSA bit (bit 3) of the OPTION_REG determines to which of the two the prescaler is connected. The prescaler is a programmable counter whose count ratio is determined by OPTION_REG bits PS0, PS1, PS2. 255(FFh) to 0(00h) an interrupt overflow occurs and T0IF bit (bit 2) of the INTCON register is set (becomes "1").
The hardware is designed such that when both the GIE (bit 7) and TOIE(bit 5) of the INTCON register are Hi ("1") the interruption occurs and the PC goes to address 004h, to start program operation from there. By writing the value of the TIMER 0 register by writing the value in TMR0.
TIMER0 starts to work in 8-bit mode it count from zero to 255 (which is 2^ 8 = 256 8 bit). When timer count reaches the maximum value of 256 the counter will reset and starts counting from zero again.

Some controller has separate TOCON instead of OPTON_REG
           



TMR0ON: Control bit in the delivery of the interim TIMER0.
T08BIT: bit selection of a working system temporary TIMER0, 8-bit or 16-bit:
T0CS: source selection bit of the time a temporary TIMER0:
T0SE: choose the type of triggering lower edge or upper edge:
PSA: bit choose to use the pre-factor measurement prescaler:
T0PS2; T0PS1; T0PS0: bits choose the value of prescaler.














  • TMR0L Register: It will store lower 8 bit value of TIMER0.
  • TMR0H Register: It will store Higher (8 bit value if it is 16 bit higher 2 bit if 10 bit) of TIMER0.
PIC Timer 0: Calculation example (INTERNAL CLOCK SOURCE)

Here is an example of the typical calculations for creating an 18ms interrupt repeat rate using PIC Timer 0.

Selecting a prescaler ratio of 1:128 gives the following interrupt period (with Fosc/4(ie. System frequency divide by 4 gives the timer frequency) or 4MHz/4 = 1MHz) and using the maximum overflow from Timer 0.

Calculation:

1/ (1MHz/128/256) = 32.768ms
Formula for calculating time    T=

128 – prescaler value
256 – Maximum timer 0 overflow value (2^8 = 256)

//for maximum timer overflow of 256 (ie. 2^8 = 256) it takes 32.768ms time.(ie. Maximum time to generate interrupt)

Obviously this is longer than we need (we require 18ms) but we can cut it down by changing the overflow point (that is by changing the value of 256 it results change in interrupt period).

To do this you need the period of the frequency input to Timer 0 which is(ie. Pulse required for timer to increment the values form 0 -255):

Calculation:

1/ (1MHz/128) = 128us

128us = frequency of pulse that required for timer 0 to increment from 0 to 255
Formula for calculating time    T=

128 – prescaler value


This is the period of time to increment from each count (0 - 255) in Timer 0.

256 * 128us = 32.768ms

So by controlling the overflow point you can set the overall interrupt period.
The required period is 18ms so some calculations:

18ms/128us = 140.625 (nearest integer value is 141)

This is the number of counts required after which the interrupt is generated. To use it Timer 0 it is loaded in the following manner:

TMR0 = 256-143; // need 141 but Timer 0 looses 2 at load.

From this point on every 128us is counted by Timer 0 and it will overflow after 141 counts (or 18ms)

Timer 0 overflow time = 141 * 128us = 18ms

PIC Timer 0: Calculation example (IEXTERNAL CLOCK SOURCE)
What is the output frequency - Fout, when the external oscillator is 100kHz and Count=8?
Calculation:
First, let’s assume that the frequency division by the Prescaler will be 1:256. Second, let’s set TMR0=0. Thus:
TIMER0 formula
Formula to calculate Fout for Timer0
MikroC Timer 0 Internal Counter Mode
Experimental Setup and Software
After reviewing the theory, lets think about doing some experiments with the Timer0 module. First, we will create an approximate 1 sec delay using the Timer0 module as a timer. The PIC16F688 runs at 4 MHz clock frequency, so the duration of a machine cycle is 1 μs. To generate 1 sec delay interval, the timer should count 1000000 machine cycles. If we use the prescaler value of 256, the required count will be reduced to 3906. Now, if we preload the TMR0 with 39, it will overflow after 256-39 = 217 counts. This gives the required number of overflows to make 3906 counts = 3906/217 = 18. With this setting, after every 18 overflows of TMR0 register (preloaded with 39), an approximate 1 sec interval is elapsed. The software below implements this to flash an LED with an approximate 1 sec interval.
 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
/*
1 sec timer using TIMER0
Internal Oscillator @ 4MHz, MCLR Enabled, PWRT Enabled, WDT OFF
*/
// Define LED connection
sbit LED at RC0_bit;
unsigned short Num;
// Interrupt service routine
void interrupt() {
  Num ++;           // Interrupt causes Num to be incremented by 1
  if(Num == 18) {
   LED = ~LED;       // Toggle LED every sec
   Num = 0;
  }
  TMR0 = 39;        // TMR0 returns to its initial value
  INTCON.T0IF = 0;  // Bit T0IF is cleared so that the interrupt could reoccur
}
void main() {
 CMCON0 = 0x07;    // Disable Comparators
 ANSEL = 0x00;     // Disable analog channels
 TRISC = 0x00;     // PORTC O/P
 LED = 0;
 Num = 0;
 OPTION_REG = 0x07; // Prescaler (1:256) is assigned to the timer TMR0
 TMR0 = 39;          // Timer T0 counts from 39 to 255
 INTCON = 0xA0;     // Enable interrupt TMR0 and Global Interrupts
 do {
 // You main programs goes here
 } while(1);  // infinite loop
}
The circuit setup is the same as Flashing an LED. The LED is driven through RC0 pin.






Note: The 1 sec delay that we just created is not very accurate because of the following reasons:
  • In this example, the PIC16F688 microcontroller was operated with the 4 MHz internal RC oscillator, which is not highly accurate and stable. For better accuracy, an external crystal oscillator must be used.
  • The timer program was written in a high-level language, so we don’t know exactly how the program looked like after being compiled. For an accurate hardware timer, the programming must be done with assembly language, and any instruction that can affect the TMR0 read/write operation, should be accounted as it may introduce additional delay.
MikroC Timer 0 External Counter Mode
 











The Timer0 module will be used as a counter to count the external clock pulses through RA2/T0CKI pin. The external clock source will be derived from the mains AC supply. The mains AC is 120V, 60 Hz sine wave signal. It will be first stepped down to 9 V, 60 Hz signal using an AC wall adapter. Before applying it to the T0CKI pin, it must be rectified and the peak voltage should be chopped down to +5V. The following circuit will convert the 9 V, 60 Hz AC voltage to an approximate +5 V square wave suitable for the T0CKI input.











The output square wave will be connected to the RA2/T0CKI pin of the PIC microcontroller. The TMR0 will start from 0 and count the incoming pulses for 1 sec. The result will be displayed on the LCD screen. The number of pulses per second is the frequency of the incoming signal, which is 50 Hz. The circuit diagram for the microcontroller and LCD part is shown below.



















/*
Timer0 as a counter
Internal Oscillator @ 4MHz, MCLR Enabled, PWRT Enabled, WDT OFF
*/
// LCD module connections
sbit LCD_RS at RC4_bit;
sbit LCD_EN at RC5_bit;
sbit LCD_D4 at RC0_bit;
sbit LCD_D5 at RC1_bit;
sbit LCD_D6 at RC2_bit;
sbit LCD_D7 at RC3_bit;
sbit LCD_RS_Direction at TRISC4_bit;
sbit LCD_EN_Direction at TRISC5_bit;
sbit LCD_D4_Direction at TRISC0_bit;
sbit LCD_D5_Direction at TRISC1_bit;
sbit LCD_D6_Direction at TRISC2_bit;
sbit LCD_D7_Direction at TRISC3_bit;
// End LCD module connections
// Define Messages
char message1[] = "Frequency=   Hz";
char *freq = "00";
void Display_Freq(unsigned int freq2write) {
 freq[0] = (freq2write/10)%10 + 48;    // Extract tens digit
 freq[1] =  freq2write%10     + 48;    // Extract ones digit
 // Display Frequency on LCD
 Lcd_Out(1, 11, freq);
}
void main() {
 CMCON0 = 0x07;    // Disable Comparators
 ANSEL = 0x00;     // Disable analog channels
 TRISC = 0x00;     // PORTC O/P
 TRISA = 0b00001100; // RA2/T0CKI input, RA3 is I/P only
 OPTION_REG = 0b00101000; // Prescaler (1:1), TOCS =1 for counter mode
 Lcd_Init();                 // Initialize LCD
 Lcd_Cmd(_LCD_CLEAR);        // CLEAR display
 Lcd_Cmd(_LCD_CURSOR_OFF);   // Cursor off
 Lcd_Out(1,1,message1);      // Write message1 in 1st row
 do {
  TMR0=0;
  Delay_ms(1000);  // Delay 1 Sec
  Display_Freq(TMR0);
 } while(1);  // Infinite loop
}





























The Watch Dog Timer's (WDT)
The function of the watchdog timer (WDT) is to prevent incorrect operation of the software. (Run away: It executes instructions which are not part of the program. / Loop: It executes the same part repeatedly.) The WDT function isn't always necessary. If there is a mistake in the program you can usually recognize there is a malfunction by the way it operates (it doesn't perform the way you expect it to). If the WDT resets the PIC you may not be able to understand what caused the program to malfunction. It may be better not to use the watchdog timer.
To stop the operation of the watchdog timer, reset the WDTE bit of the configuration word (2007h) of the program memory to "0". In PIC (the PICMicro used in this experiment), the watchdog timer has a nominal time-out period of 18 ms, which can be extended up to (18msec x 128 = 2,304msec) by using a prescaler with a division ratio of up to 1:128. The prescaler rate is selected through PS0-PS2 bits of the OPTION register. Note that the PSA bit must also be set to assign the prescaler to WDT, otherwise it will be used by Timer0.






















                         In normal operation, if the watchdog timer is enabled, a WDT reset instruction (CLRWDT) is placed in the main loop of the program, where it would normally be expected to be executed often enough to prevent the WDT overflow. If the program hangs, and the CLRWDT instruction is not executed in time, the program counter is reset to 0000 so that the program restarts.
However, if the PIC microcontroller is in Sleep mode, a WDT time-out will not reset the device, but just causes it to wake up (known as WDT wake-up) and the microcontroller continues program execution from the instruction following the Sleep instruction.
Note: When a Sleep instruction is executed, the watchdog timer is cleared. But the WDT will keep running if it has been enabled.
Example: 1
Experimental Setup
The setup for this experiment would be as shown in the circuit diagram below. The PIC16F628A microcontroller runs at 4.0 MHz clock using an external crystal. An LED is connected to RB0 port pin, which glows for .5 sec after every 4.3 second delay. During the first half of the delay (2.3 sec), the microcontroller is put to Sleep mode and a WDT time-out wakes it up. The remaining 2 sec delay is created through a software routine using NOP instruction. An ammeter is connected in series between the power supply voltage and the microconroller circuit to monitor the current consumption of the circuit. The MCLR pin is pulled high through a 10 K resistor.
















Software
The following program is written in C and compiled with mikroC Pro for PIC. The OPTION register is configured to assign prescaler to WDT. The prescaler ratio of 1:128 creates the WDT time-out duration of approximately 2.3 sec. An additional software delay of 2 sec is created using the Delay_ms() library routine. The amount of current consumption during both the delay intervals is displayed on the digital ammeter.
/*    Understanding sleep mode in PIC microcontrollers
  * Test configuration:
     MCU:             PIC16F628A
     Oscillator:      XT, 4.0000 MHz
*/
sbit LED at RB0_bit;       // LED is connected to PORTB.0
 
void main() {
TRISB = 0b00000000;
TRISA = 0b00100000;
LED = 0;
OPTION_REG = 0b00001111;   // Assign 1:128 prescaler to WDT
do {
  asm sleep;               // Sleep mode, WDT is cleared,
                           // and time out occurs at Approx. 2.3 Sec
                           // Current is minimum in Sleep mode
  LED = 1;                 // After wake-up the LED is turned on
  Delay_ms(500);
  LED = 0;
  asm CLRWDT;              // Clear WDT
  Delay_ms(2000);          // Measure current here and compare with Sleep mode
 }while(1);                // current
}
 


Example : 2
              This example illustrates how the watch-dog timer should not be used. A command used for resetting this timer is intentionally left out in the main program loop, thus enabling the microcontroller to be reset. As a result, the microcontroller will be reset all the time, which is reflected as PORT RB.0~RB.7 LED blinking.
              Button S1 and (pull-up) resistance with value of 4.7 K will provide the manual reset mode to the microcontroller.
         
/*  Test configuration:
'    MCU:                        PIC16F628A
'    Test.Board:                 WB-106 Breadboard 2420 dots
'    SW:                         MikroC PRO for PIC 2010 (version v4.15)
'  Configuration Word
'    Oscillator:                 INTOSC:I/O on RA.6, I/O on RA.7
'    Watchdog Timer:             ON
'    Power up Timer:             Disabled
'    Master Clear Enable:        Enabled
'    Browun Out Detect:          Enabled
'    Low Voltage Program:        Disabled
'    Data EE Read Protect:       Disabled
'    Code Protect:               OFF
*/
void main()                     // Main;
{
 OPTION_REG = 0x0E;              // Prescaler is assigned to timer wdt
                                                    // (1:4[0x0A];1:8[0x0B];1:16[0x0C];1:32[0x0D];
                                                    // 1:64[0x0E];1:128[0x0F]);
 asm CLRWDT;                          // Assembly command to reset WDT timer;
 TRISB = 0x00;                          // All port B pins are configured as output;
 PORTB = 0x0F;                        // Initial value of the PORTB register;
 Delay_ms(600);                        // 600ms delay;
 PORTB = 0xF0;                       // PORTA B value different from initial;
 while(1);                                   // Endless loop. Program remains here until WDT
                                                  // timer reset the microcontroller;
 }                                               // End.

Monday, 26 September 2011

Counting type ADC

ADC tutorial (counting type)

ADC Math
0-5V Analog I/P (This is the maximum voltage we can give to PIC)

0-1023 Digital Count (The equailent digital count 2n n stands for number of bits example 28 = 2x2x2x2x2x2x2x2x2x2 = 1023)

Resolution = (5-0)/(1023-0) = 0.00489 V/Count  (or)  5/1020(or) 2n -1

To measure the 5 volt using counting type ADC

I/P voltage = Va = 0.00489 (V/Count)  * Digital Count (we know it takes 1023 count to convert 5 volt to equalent binary value from 0000000000 to 1111111111)

Consider the below example

                       
In the above circuit we want to measure 20 volt but our PIC will handle maximum voltage of 5Volt. So we want to do some voltage divider network
            Consider the above equation and find the resistance value here the roll of zener is it will ensure only the voltage is around 5.1 volt.
All the things are same except input voltage its 20V instead of 5V
so we want to do some mathematical equation in Va (Input voltage)

So Va (I/P voltage) = 4(to multiply to convert become 20)*Va = 4* Digital Count * 0.00489(resolution) = 0.01956 * Digital Count
10 bit ADC Binary and equailent analog count

0000000000  (binary minimum count)      0.00489 V/Count 
0000000001  (binary minimum count)      0.00489 V/Count
0000000010  (binary minimum count)      0.00978 V/Count (0.00489+0.00489)
.
.
.
.
.
.
.
.
.
1111111111 (binary minimum count)       1023

Make the thing clear 0.00489 (V/Count) x 1023 = 5.0024 (app 5V)

To avoid floating point, use I/P voltage = 196*Digital Count. This number is a long integer.
Example, suppose
Vin = 13.6V. Then,
Va = 0.25*Vin = 3.4V
=> Digital Count = 3.4/0.00489 = 695
=> Calculated I/P Voltage = 196*695 = 136220 = 13.6V  (First 3 digits of 6 digit product)

Sources of error
The above math looks pretty easy but when you implement it, you will not errors in output measurements  because the above calculations are based on following ideal conditions:
  • Vcc supply voltage to PIC12F683 is exactly 5V,
  • R1 and R2 are exact 1.3K and 3.9K respectively

So, let's revise the math above with real figures. I measured the supply voltage to PIC and it is 5.02V. R1 and R2 are measured to be 1267 and 3890 Ohms. So this gives,
 0 - 5.02 V Analog I/P ---> 0-1023 Digital Count
=> Resolution = (5.02-0)/(1023-0) = 0.004907 V/Count
Va = 1267*Vin/(1267+3890) = 0.2457*Vin
=> I/P voltage = 4.07*Va = 4.07* Digital Count * 0.004907 
                       = 0.01997 * Digital Count 
                       = 0.02*Digital Count (Approx.)

To avoid floating point, use I/P voltage = 2*Digital Count. No need for Long integer.
Example, suppose 
Vin = 4.6V. Then, 
Va = 0.2457*Vin = 1.13V
=> Digital Count = 1.13/0.004907 = 230
=> Calculated I/P Voltage = 2*230 = 0460 = 04.6V  (First 3 digits of 4 digit product)


Supose u want to make a volt meter it's maximum range will be 5 Volt(as u say)
Ur ADC converter has 10 Bit resulotion it mean in it has maximum value in decimal 1023.

It means as u apply maximum voltage(as u say 5 volt) to ADC it will give u 1023 in decimal.
So the formula is
Voltage = Applied voltage * maximum range / ADC resulotion

Voltage > measured voltage (we don't know it's value)

Applied Voltage > This is our applied voltage which we apply on ADC leg it rage will be 0 ~ 1023.

Maximum range > for which range u want to scale it( as u say 5 volt)

ADC resulotion > 10 Bit it means it maximum value in binary = (1111111111) or in Decimal 1023.

Now we start,
u want to scale it for 5 Volt( it mean it's maximum reading will 5 volt)
The formula will be
Voltage = Applied voltage * 5 / 1023
for example u apply 5 volt to ADC.
then ADC will convert this value into binary so it will be (1111111111) or if we see it in decimal then it will be 1023
It means
Applied Voltage = 1023
So we put this value in our formula
Voltage = 1023 * 5 / 1023
So the answer will be
Voltage = 5

Now an other example
Supose u apply 2.5 volt to ADC
So it will conver it into binary (0111111111) or it will eqaul to 511 in decimal.
Applied Voltage = 511
So we put this value in our formula
Voltage = 511 * 5 / 1023
So the answer will be
Voltage = 2.497