Pololu Robotics & Electronics
Menu
My account Comments or questions? About Pololu Contact Ordering information Distributors

I/O question in programming


#44

The LV-168 does have an external 20 MHz oscillator and we configure it to use this external crystal.

I notice you are missing a zero in your clock frequency. Make sure you are setting F_CPU to 20000000 (that’s twenty million: 20,000,000 – you should have seven zeroes).

- Ben


#45

Hurray!!! That was it Ben. I had missed the zero. I now have Test for the first time. Can’t wait to get back to the original project now. Adam I will let you know how it goes.

Thanks so much guys!!!

Tony


#46

That’s great to hear! All you should need to change to go back to the old code (that displayed NSEW) is the F_CPU number in the first line of the Orangutan code (from 8000000 to 20000000).

You might still want to try bringing in the beacon and the radio separately, you could use this code to send “Test” over the radio, and try the NSEW code over the wire.

-Adam


#47

I hooked the radios up. I left the test code in both MCUs. No IR Beacon. I have each radio sharing power from corisponding MCU. Nothing comes across the LCD of the LV-168 unless I touch either antenna of the radio. Then the weird characters show up. That sounds like a ground issue to me, but not sure what to do about it. I hate to give the radios seperate power source. Oh, I also took out the common ground beteen the MCUs. Just to make sure I am still outputting the Test code I put back the common ground and move the wire and still get expected outcome of Test.

Thoughts?

Tony


#48

The transmitter should operate just fine from your 9V battery, but the receiver really wants to run off of 5V. You might try powering it off of one of the 5V I/O power ports of your Orangutan (you’ll need to connect the selection jumper, see the Orangutan LV-168 usage notes). The radio manufacturer doesn’t really specify how much current the receiver needs to run, but the Baby O’s step up regulator can handle 200mA, so it should be enough.

If this doesn’t do it another picture of your breadboard should be worth 1000 words.

-Adam


#49

Adam, here is a picture of you suggestion. I get constant scrolling.


New Pololu.com website
#50

Looks like you have everything wired up correctly. And on further reading, Sparkfun says that the receiver only uses about 3.5mA, so it should be fine running off of 5V from your LV-168 Orangutan.

Reading a little bit about this on the Sparkfun wireless forums, some people have problems with the transmitter overwhelming the receiver when they are placed very close together (i.e. on the same breadboard). First, try getting rid of your antenna wires (you can put them back if you want later, when the devices are a little more separated). You can also reduce the transmitter strength by powering it from the Baby O’s 5V regulator (connect the transmitter VCC pin to the Baby O’s VCC pin, which is unfortunately a little cumbersome to do on a breadboard since you don’t want to short VCC and M1).

I’m also finding posts from a lot of people who are dissatisfied with how sensitive to noise and interference these radios are. I find this very odd, since the Sparkfun product description practically glows about how fantastic and simple a cut-the-cord wireless solution these are. I know you said before you were trying to keep this project cheap, but you may want to think about investing in some better radios, like ones that do internal signal processing and error checking. I’ve used the LPRS easyRadio modules with great success, but they’re also about five times more expensive than these. Anyway, that’s for later. Before even thinking about that you should be able getting some good data flowing between these!

-Adam


#51

Adam,

I tryed connecting the trans. to the VCC of the Baby-O and that did not help. Not sure what made me think of this but I tryed what Ben suggested aways back.

for(i=0;i<100;i++){ 
         _delay_ms(10); 

to the top of the while(1) loop (before the USART_Trans(‘T’)) and see if that has any effect at all.

  • Ben

By golly it worked!!

Now for the next part of hooking back up the IR beacons and code for that.

Wish me luck,

Tony


#52

Adam & Ben,

The whole thing works perfectly!

Thanks so much for your consistent help in this project. Top notch support even at some crazy hours. I learned a bunch. This project is not actually for a robot, but I am anxious to start one. I have taken apart an old remote control car that has a good base for starting a robot so I will most likely need help on that project someday.

For now I am off to testing the 3 piece IR Beacon in the real environment, outdoors.

Thanks a million.

Cheers,
-Tony


#53

I’m very glad to hear you were able to get everything working!

- Ben


#54

Hello,

I haven’t been following this conversation that much, but I don’t think the IR beacon will work very well outdoors. Maybe at night.

- Jan


#55

I hoped that would not be the case. Is the problem in the IR trans. or IR receive or both. Is there different sensors I can add to the IR Beacon board to help? Could you offer any advice. I realy like the Pololu IR Beacon set up so I hate to move from it if the outdoors becomes a problem.

-Tony


#56

The IR beacon is based on transmitting IR and comparing the IR reception on the receivers facing in different directions. The LEDs are already set to be very bright to get 20’ range indoors, but if it’s bright enough outside, the receivers just won’t see the transmitters. There’s not much you can do about that, but you can at least try to cover the receivers so that they are not in direct sunlight. Your results will probably vary with the time of day and weather.

- Jan


#57

Jan,

Check out this encloser seller/site. hammondmfg.com/dwg2c.htm
Would the blue or red help.

-Tony


#58

It looks like the red one might help.

- Jan


#59

Hey nexisnet,

wondering if you could spend a minute or two looking at this code. You will see some of it is code you had helped me on. Recall I needed the IR beacon to be wireless and that part works great in this code I am posting. However now I am looking to get the EZ1 to come across wireless and have piced in code that Ben helped me on when I used the EZ1 with ADC.

My Receive code:

[code]#define F_CPU 20000000//CPU clock
#define BAUD 2400//baud rate for UART
#define MYUBRR (F_CPU/16/BAUD-1)//baud rate variable for UART hardware

#include <avr/io.h>
#include <avr/interrupt.h>
#include “device.h”
#include “lcd.h”

unsigned char newBeaconState=0;

ISR(USART_RX_vect){//USART Byte reieved
newBeaconState=1;
}

void USART_Init(unsigned int ubrr){//Initialize USART hardware & settings for Serial Radio
UBRR0H=(unsigned char)(ubrr>>8);//set buad rate
UBRR0L=(unsigned char) ubrr;
UCSR0B=(1<<RXEN0);//enable reciever
UCSR0B|=(1<<RXCIE0);//enable recieve complete interrupt
UCSR0C=(3<<UCSZ00);//Set frame format for 8bit with 1 stop
}

void delay_ms(unsigned int time_ms) // delay for time_ms milliseconds
{
while (time_ms–)
delay_ms(1); // defined in <util/delay.h>, argument must be < 13
}

int main(){
lcd_init(); // this function must be called before any other LCD command

USART_Init(MYUBRR);//Initialize serial USART 
sei();//enable global interrupts 

while (1){ // loop forever

  long sum = 0;
  unsigned int avg;{
  
  
       UDR0 |= ( 1 << PC4 );      // start conversion
     while (UDR0&(1<<PC4))      // wait while converting
        ;
     sum += PC4;                     // add in conversion result
  }
  avg = sum >> 9;      // a.k.a. avg = sum / 2^9 = avg / 512
                 //  typically, bitshifting is faster than division
   

  if(newBeaconState){ 
     newBeaconState=0; 
     lcd_gotoxy(0, 1); // go to the start of LCD line 2 

     

     if(!(UDR0&(1<<PC1))){//east detected, do something 
        lcd_string("<--     "); // write a character 
     
     } 

     if(!(UDR0&(1<<PC2))){//south detected, do something 
        lcd_string("   <>   "); // write a character 
     
     } 

     if(!(UDR0&(1<<PC3))){//west detected, do something 
        lcd_string("     -->"); // write a character 
     }
	 
	 if(!(UDR0&(1<<PC4))){//EZ1 detected, do something 
  		lcd_gotoxy(0, 0);
		lcd_string("DIST:");
  		lcd_gotoxy(6, 0);            // go to the start of LCD line 2
  		lcd_int(avg * 4.89 / 117.60);   // display average ADC as an integer 
     }

	 else{ 
        lcd_string("NOSIGNAL"); // write a space
     } 
  
  }

}

return 0;
}
[/code]

My Transmit code:

[code]#define F_CPU 20000000//CPU clock
#define BAUD 2400//baud rate for UART
#define MYUBRR (F_CPU/16/BAUD-1)//baud rate variable for UART hardware

#include <util/delay.h>
#include <avr/io.h>

void USART_Init(unsigned int ubrr){//Initialize USART hardware & settings for Serial Radio
UBRR0H=(unsigned char)(ubrr>>8);//set buad rate
UBRR0L=(unsigned char) ubrr;
UCSR0B=(1<<TXEN0);//enable transmitter
UCSR0C=(3<<UCSZ00);//Set frame format for 8bit with 1 stop
}

void USART_Trans (unsigned char data){//Transmit a byte of data over USART
while(!(UCSR0A&(1<<UDRE0)));//wait for transmition to complete
UDR0=data;
}

int main(){
unsigned char i;
DDRC&=~((1<<PC0)|(1<<PC1)|(1<<PC2)|(1<<PC3)|(1<<PC4));//Configure PortC IO

USART_Init(MYUBRR);//Initialize serial USART

while(1){
for(i=0;i<512;i++){
_delay_ms(10);

  USART_Trans(PINC);//Transmit state of PINC 
   
  } 

}

return 0;
}
[/code]


#60

Hey Tmapes, the code looks good to me, is it working well?

I’m a little confused though, the sonar ADC is on the receiver side, are you thinking of moving it to the transmitter (baby-o) side?

-Adam


#61

Correct. Could you help me in achieving that? The IR beacon works great in this code and displays correctly using PC0,1,2,3. I want to use PC4 on the transmitter (baby-o) side and display DIST: on the recieve side.

Tony


#62

After actually working through your code I noticed a few mistakes. You seemed to be mixing up UART and ADC registers. Just so we’re on the same page, I brought Ben’s sonar reading code back in to your receiver code. Also reading the UDR0 register multiple times was causing a few things to get loopy. I added a variable called beaconState to holds the value of the incoming serial byte, so we only have to read the register once.

So, here’s a starting point with the big Orangutan receiving just the beacon state from the baby O over the radio, and directly reading and displaying (in approximate inches) the Sonar analog output on PC4:

[code]#define F_CPU 20000000//CPU clock
#define BAUD 2400//baud rate for UART
#define MYUBRR (F_CPU/16/BAUD-1)//baud rate variable for UART hardware

#include <avr/io.h>
#include <avr/interrupt.h>
#include “device.h”
#include “lcd.h”

unsigned char newBeaconState=0;
unsigned char beaconState;

ISR(USART_RX_vect){//USART Byte reieved
newBeaconState=1;
beaconState=UDR0;
}

void USART_Init(unsigned int ubrr){//Initialize USART hardware & settings for Serial Radio
UBRR0H=(unsigned char)(ubrr>>8);//set buad rate
UBRR0L=(unsigned char) ubrr;
UCSR0B=(1<<RXEN0);//enable reciever
UCSR0B|=(1<<RXCIE0);//enable recieve complete interrupt
UCSR0C=(3<<UCSZ00);//Set frame format for 8bit with 1 stop
}

void delay_ms(unsigned int time_ms) // delay for time_ms milliseconds
{
while (time_ms–)
delay_ms(1); // defined in <util/delay.h>, argument must be < 13
}

int main(){
lcd_init(); // this function must be called before any other LCD command
USART_Init(MYUBRR);//Initialize serial USART

// ***** ADC INITIALIZATION *****
ADCSRA = 0x87; // bit 7 set: ADC enabled
// bit 6 clear: don’t start conversion
// bit 5 clear: disable autotrigger
// bit 4: ADC interrupt flag
// bit 3 clear: disable ADC interrupt
// bits 0-2 set: ADC clock prescaler is 128

ADMUX = 0x04; // bit 7 and 6 clear: voltage ref is Vref pin
// bit 5 set: right-adjust result (10-bit ADC)
// bit 4 not implemented
// bits 0-3: ADC channel (channel 4)

sei();//enable global interrupts

while (1){ // loop forever

  long sum = 0;
  unsigned int avg, i;

  // Here we accumulate 512 conversion results to get an average ADC.
  // According to the mega168 datasheet, it takes approximately 13
  // ADC clock cycles to perform a conversion.  We have configured
  // the ADC run at IO clock / 128 = 20 MHz / 128 = 156 kHz, which
  // means it has a conversion rate of around 10 kHz.  As a result,
  // it should take around 50 ms to accumulate 500 ADC samples.
  for (i = 0; i < 512; i++){
     ADCSRA |= ( 1 << ADSC );         // start conversion
     while ( ADCSRA & ( 1 << ADSC ))      // wait while converting
        ;
     sum += ADC;                     // add in conversion result
  }
  avg = sum >> 9;      // a.k.a. avg = sum / 2^9 = avg / 512
                 //  typically, bitshifting is faster than division

	lcd_gotoxy(0, 0);
	lcd_string("DIST:");
	lcd_int(avg>>1);   // display average ADC as an integer, divided by 2 for aprox inches

  if(newBeaconState){
	 newBeaconState=0;


	 lcd_gotoxy(0, 1); // go to the start of LCD line 2

	 

	 if(!(beaconState&(1<<PC1))){//east detected, do something
		lcd_string("<--     "); // write a character
	 
	 }

	 if(!(beaconState&(1<<PC2))){//south detected, do something
		lcd_string("   <>   "); // write a character
	 
	 }

	 if(!(beaconState&(1<<PC3))){//west detected, do something
		lcd_string("     -->"); // write a character
	 }


   else{
		lcd_string("NOSIGNAL"); // write a space
	 }
 
 }

}

return 0;
}[/code]
I just tried this out on my original Orangutan (with a 20MHz clock, just like your LV Orangutan) and it seems to work fine.

Now, to transmit all this information you’re going to need more than just one byte of data. I moved the sonar ADC code over to the transmitter program (still on PC4), and rewrote it a little to send bytes in sets of 3. First is a ‘start’ byte, with a value of 255 (the maximum number one byte can hold). No other byte will be 255, so the receiver knows that it should begin listening for new beacon and sonar information after seeing a 255 byte. Next the transmitter sends the beacon state, but instead of sending all of PORTC, it sends (PINC&0x0F), which is only the lower four bits. This way the byte value can only be 0 to 127, and can never be 255 (i.e. a false start byte). Last, the transmitter sends the sonar reading, in approximate inches, between 0 and 254.

Code for the Baby Orangutan Transmitter:

[code]#define F_CPU 20000000//CPU clock
#define BAUD 2400//baud rate for UART
#define MYUBRR (F_CPU/16/BAUD-1)//baud rate variable for UART hardware

#include <util/delay.h>
#include <avr/io.h>

void USART_Init(unsigned int ubrr){//Initialize USART hardware & settings for Serial Radio
UBRR0H=(unsigned char)(ubrr>>8);//set buad rate
UBRR0L=(unsigned char) ubrr;
UCSR0B=(1<<TXEN0);//enable transmitter
UCSR0C=(3<<UCSZ00);//Set frame format for 8bit with 1 stop
}

void USART_Trans (unsigned char data){//Transmit a byte of data over USART
while(!(UCSR0A&(1<<UDRE0)));//wait for transmition to complete
UDR0=data;
}

int main(){
DDRC&=~((1<<PC0)|(1<<PC1)|(1<<PC2)|(1<<PC3));//Configure PortC IO

USART_Init(MYUBRR);//Initialize serial USART

// ***** ADC INITIALIZATION *****
ADCSRA = 0x87; // bit 7 set: ADC enabled
// bit 6 clear: don’t start conversion
// bit 5 clear: disable autotrigger
// bit 4: ADC interrupt flag
// bit 3 clear: disable ADC interrupt
// bits 0-2 set: ADC clock prescaler is 128

ADMUX = 0x04; // bit 7 and 6 clear: voltage ref is Vref pin
// bit 5 set: right-adjust result (10-bit ADC)
// bit 4 not implemented
// bits 0-3: ADC channel (channel 4)

while(1){
	  long sum = 0;
      unsigned int i;
	  unsigned char dist;

      // Here we accumulate 512 conversion results to get an average ADC.
      // According to the mega168 datasheet, it takes approximately 13
      // ADC clock cycles to perform a conversion.  We have configured
      // the ADC run at IO clock / 128 = 20 MHz / 128 = 156 kHz, which
      // means it has a conversion rate of around 10 kHz.  As a result,
      // it should take around 50 ms to accumulate 500 ADC samples.
      for (i = 0; i < 512; i++){
         ADCSRA |= ( 1 << ADSC );         // start conversion
         while ( ADCSRA & ( 1 << ADSC ))      // wait while converting
            ;
         sum += ADC;                     // add in conversion result
      }
      dist = sum >> 10;      // a.k.a. dist = sum / 2^9 = avg / 512, divided again by 2 for aprox inches
                     //  typically, bitshifting is faster than division
	  if(dist==255){
		dist=254;//only start byte can be 255
	  }
	USART_Trans(255);//transmit start byte
	USART_Trans(PINC&0x0F);//Transmit lower four bits of pinc (beacon state)
	USART_Trans(dist);
	_delay_ms(10);
}

return 0;

}[/code]
Finally, here is some code for the big Orangutan receiver. It waits until it receives a start byte, then stores the next two bytes in an array. When it has received the two data bytes, it displays them on the screen.

Code for the Big Orangutan Receiver:

[code]#define F_CPU 20000000//CPU clock
#define BAUD 2400//baud rate for UART
#define MYUBRR (F_CPU/16/BAUD-1)//baud rate variable for UART hardware

#include <avr/io.h>
#include <avr/interrupt.h>
#include “device.h”
#include “lcd.h”

unsigned char newBeaconState=0,beaconIndex=0,UDRGrab;
unsigned char buff[2]={0,0};

ISR(USART_RX_vect){//USART Byte reieved
UDRGrab=UDR0;//read the byte right away
if(UDRGrab==255){//new beacon state
beaconIndex=0;//reset the index
return;//exit and wait for more bytes
}else if(beaconIndex<2){
buff[beaconIndex]=UDRGrab;//store the byte
beaconIndex++;//increase the index
if(beaconIndex==2){//received both bytes
newBeaconState=1;
}
}
}

void USART_Init(unsigned int ubrr){//Initialize USART hardware & settings for Serial Radio
UBRR0H=(unsigned char)(ubrr>>8);//set buad rate
UBRR0L=(unsigned char) ubrr;
UCSR0B=(1<<RXEN0);//enable reciever
UCSR0B|=(1<<RXCIE0);//enable recieve complete interrupt
UCSR0C=(3<<UCSZ00);//Set frame format for 8bit with 1 stop
}

void delay_ms(unsigned int time_ms){ // delay for time_ms milliseconds
while (time_ms–)
delay_ms(1); // defined in <util/delay.h>, argument must be < 13
}

int main(){
lcd_init(); // this function must be called before any other LCD command
USART_Init(MYUBRR);//Initialize serial USART

sei();//enable global interrupts

while (1){         // loop forever
	  if(newBeaconState){
		 newBeaconState=0;

		lcd_gotoxy(0, 0);
		lcd_string("DIST:");
		lcd_int(buff[1]);   // display average ADC as an integer

		 lcd_gotoxy(0, 1); // go to the start of LCD line 2

	 
		 if(!(buff[0]&(1<<PC1))){//east detected, do something
			lcd_string("<--     "); // write a character
	 
		 }

		 if(!(buff[0]&(1<<PC2))){//south detected, do something
			lcd_string("   <>   "); // write a character
	 
		 }

		 if(!(buff[0]&(1<<PC3))){//west detected, do something
			lcd_string("     -->"); // write a character
		 }


	   else{
			lcd_string("NOSIGNAL"); // write a space
		 }
 
	 }

}

return 0;
}[/code]
There are some big code changes in here, so feel free to ask questions if any of this doesn’t make sense. Also, I generally try not to post code I haven’t completely tested, but I only have one Orangutan that isn’t embedded in a project right now, so I tested the two programs separately. I’m fairly sure they’ll work together though, but I guess you’ll have to answer that question.

-Adam


#63

Adam,

loaded the Orangutans with your new code. The result on the LCD is odd. Information is oreiented correctly however, data just flips and does not seem to be working with the sensors, both beacon and EZ1. Rather a random pattern of DIST: numbers and my directional arrows are flipping all around not connected to the beacon. I have proven the radio is working because if I turn off the baby the data on the big stops. FYI, the input for the EZ1 of course is my hand waving in front of it and the beacon input is the paired second beacon moving in front of the beacon attached to the baby.

Have I explained this enough for you, if not let me know and I can take another stab at it.

Tony