[ Log In ]
Image of the Atmega324p

Atmega324P

$8.50
Qty:
Serial to USB converter with Micro USB cable

USB to Serial Converter

$10.95
Qty:
Thumbnail: Crystal Oscillator 18.432 MHz for UART

18.432 MHz Crystal Oscillator 18pf 30ppm

$0.94
Qty:
Thumbnail: 22 pF Capacitor

22 pF Multilayer Ceramic Capacitor

$0.43
Qty:
Thumbnail: Quartz crystal oscillator - 16 MHz

16 MHz Crystal Oscillator 20 pF Through Hole

$0.75
Qty:
Thumbnail: 4x4 keypad top view.

4x4 Keypad with Adhesive Backing

$3.80
Qty:
Thumbnail: quad buffer line driver 74HC126E

Quad Buffer Line Driver (Through Hole)

$0.69
Qty:
USB AVR programmer

USB AVR Programmer

$9.95
Qty:
3 pin slide switch

SPDT Slide Switch 3 pin 30V

$1.49
Qty:
Handheld auto range multimeter

Handheld Auto Ranging Digital Multimeter

$17.95
Qty:
Skip Navigation Links

Getting the Full 10-bits from the ADC (Analog to Digital Converter)

This is where we UP the anty with Mrs. ADC. Do you have the faith? Do you think she will be able to handle that much cash?!? She's pretty stressed as it is! She is sensitive to sound, she's dishonest, and she makes mistakes all the time. Let's see how she does.

Actually, she wasn't that bad with the 256 currency exchange mode. But that's where it ends. The 10-bit 1024 exchange mode is a bit more of an issue with Mrs. ADC. She gets real nervous with that much currency! Noice from Breadboardville, Mr. Pot and King Core is making her crazy. There is cacophony all around her.

This time, Mr. Cap is really necessary! Fat Cap would do a good job, but Mr. Fat Cap is no match for the upheaval going on at Breadboardville. Breadboardville is a town with gusts of wind that can wash away rain voltage currency. If there are waves of this untamed currency in the air, Breadboardville is sure to pick it up. See the next chapter to see if Mr. Cap can launder his voltage currency well enough to overcome this turbulence! We will also meet Mr. Gravity! The tipsy fellow down the metal street.

So, how do we capture the 10-bit number currencies? We only need to look at another register. For the 8-bit number, we were looking at the ADCH (Analog to Digital Conversion Result High) and when the ADLAR in ADMUX is set, then the ADCL is left justified and the ADCL will contain a number between 0-256. When we get ADCL (Analog to Digital Conversion Result Low) involved, then we can capture the two extra bits that exist in the 10-bit number.

By using the ADCH for just receiving 0-255, we are essentially skipping every 4th number in the actual ADC. By including the ADCL, we are able to capture the extra 4 numbers in-between each of the ADCH numbers.

Please don't glaze over the following information!!! If you don't understand shifting operations (>> or <<), then this will really help you. Read it carefully!

First, we utilize a 16-bit variable to hold the 10-bit number (in the program, I call it "theTenBitResults"):

So, theTenbitResults starts off like this:

bit15 bit14 bit13 bit12 bit11 bit10 bit9 bit8 bit7 bit6 bit5 bit4 bit3 bit2 bit1 bit0
With ADLAR = 1:

ADCL starts off with these bits (remember 2 bits have 4 possibilities 0, 1, 2 and 3):

bit1 bit0 ---- ---- ---- ---- ---- ----

Two of the 10 total bits.

And ADCH starts off with these bits (remember that 8 bits have 256 possibilities):

bit9 bit8 bit7 bit6 bit5 bit4 bit3 bit2

The remaining 8 of the total 10 bits.

Shifting ADCL to the right 6 places using this: ADCL >> 6, we would get:

---- ---- ---- ---- ---- ---- bit1 bit0

If by stating theTenBitResult |= ADCL >> 6; theTenbitResults would look like this:

---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- bit1 bit0

The ADCL is shifted to the right by 6 places (moving it all the way to the right)

Now we just need to get the ADCH into the theTenbitResults variable. All we need to do is make room for the two bits of the ADCL, so by shifting ADCH two places to the left and still applying the ADCL, we get: theTenBitResult |= ADCH << 2 | ADCL >> 6

---- ---- ---- ---- ---- ---- bit9 bit8 bit7 bit6 bit5 bit4 bit3 bit2 bit1 bit0

And then Bob's your Uncle, if you like left adjusted stuff. Let see how to do it right adjusted.

Don't worry about the 6 un-used spaces in the 16-bit number. We really don't have any other options.

With ADLAR = 0:

ADCL starts off with these bits:

bit7 bit6 bit5 bit4 bit3 bit2 bit1 bit0

Eight of the 10 total bits (the 8 low bits).

And ADCH starts off with these bits:

---- ---- ---- ---- ---- ---- bit9 bit8

The remaining 2 of the total 10 bits (the two high bits).

Shifting ADCH to the left 8 places using this: ADCH << 8, we would get:

bit9 bit8 ---- ---- ---- ---- ---- ---- ---- ----

If by stating theTenBitResult |= ADCH << 8; theTenbitResults would look like this:

---- ---- ---- ---- ---- ---- bit9 bit8 ---- ---- ---- ---- ---- ---- ---- ----

The ADCH is shifted to the ledft by 8 places top make room for the low 8 bits.

All we need to do now is to place the ADCL into the variable since the lower 8 bits would just naturally be set into the correct places: theTenBitResult |= ADCH << 8 | ADCL

---- ---- ---- ---- ---- ---- bit9 bit8 bit7 bit6 bit5 bit4 bit3 bit2 bit1 bit0

And there you go!

Lets see how this program may appear (I selected the latter of the two approaches with ADLAR = 0):

#include <avr/io.h>
#include <avr/interrupt.h>
#include "MrLCD.h"
int main(void)
{
InitializeMrLCD();
Send_A_StringToMrLCDWithLocation(1,1,"ADC Result:");
ADCSRA |= 1<<ADPS2;
ADMUX |= (1<<REFS0) | (1<<REFS1);
ADCSRA |= 1<<ADIE;
ADCSRA |= 1<<ADEN;

sei();

ADCSRA |= 1<<ADSC;

while (1)
{
}
}
ISR(ADC_vect)
{
uint8_t theLowADC = ADCL;
uint16_t theTenBitResults = ADCH<<8 | theLowADC;
Send_An_IntegerToMrLCD(13,1,theTenBitResults, 4);

ADCSRA |= 1<<ADSC;
}