Dallas/Maxim One Wire Library

[zeke 693]

I don't think you want to turn this place into PICList. That would just be :crazy:

[bluehash 1,581]

I apologize guys but oPossum knows

I was just kidding, feel free to post code from other architectures.

[gordon 229]

Not that I mind, but maybe the Other Microcontrollers section would be a better fit.

[nuetron 64]

Pardon my ignorance, but what is "Dallas/Maxim One Wire"? Is it an asynchronous single-line protocol?

[zeke 693]

Pardon my ignorance, but what is "Dallas/Maxim One Wire"? Is it an asynchronous single-line protocol?

 

Yes. It's a two wire bus that has a time based protocol. Maxim/Dallas is the only manufacturer in the world.

 

Here's an excellent summary of it.

 

Most people use their temperature sensors in their HVAC projects.

 

But, it's also used in a serious manner. I have a client who has based their entire business on these temperature sensors. They are one of the largest users of one wire technology in the world.

[nuetron 64]

How difficult would it be to modify One Wire for iRed, as in this schematic?

[oPossum 1,083]

MSP430s with USCI support IRDA. That would be a much better way than trying to adapt one wire to IR.

[krzyk2 11]

Thanks to pointers from Rickta59 here's code tuned to work in msp430-gcc.

I've used it in my project to read temperature from DS sensor and it works correctly.

dow.zip

[Skunky Lee 0]

Hi!

 

So I'm having a bit of a problem trying to figure out this code, and I'm wondering if anyone could help me out a bit.

 

Basically, I'm trying to use G2231 to read a DS18B20. I'm using pin 1.6 and 1 MHz factory calibrated clock.

The thing is, when I send down the initial check owex(0,0), I get correct behaviour (tested on 2 different sensors). 0xFFFF if the sensor is plugged in and 0 if it's not.

 

Then, I try to send down the Skip Rom command (0xCC) in order to verify the communication. It fails, for it always returns a 0. Curious enough, though, if I send down the ReadRom command (0x33), I do get the echo back (0x33). I'm using the sensor in the pull-up resistor configuration.

 

I get the feeling I'm missing something basic here. It says in the OP that if I don't get an echo when using 0xCC command, there's a problem on my bus, but I dont actually get what that would mean. I don't have access to an oscilloscope and the voltages check out fine. I'd be grateful for any help.

 

Here's my main code:

void main(void)
{

volatile int p;
volatile int test;

WDTCTL = WDTPW + WDTHOLD; // Stop watchdog timer

DCOCTL = 0;
BCSCTL1 = CALBC1_1MHZ;
DCOCTL = CALDCO_1MHZ;


	LCD_init();

	one_wire_setup(&P1DIR, &P1IN, BIT6, 1);

	LCD_gotoXY(0,0);

	p = owex(0,0);

	if(p == 0xFFFF){
		LCD_writeString("D found.");
		LCD_gotoXY(0,1);
		test = owex(0xCC, 8);
		LCD_writeString(convertADC(test,0));
	}else{
		LCD_writeString("D not found.");
	}

}

 

So basically, if there's a sensor plugged in, I get this on the lcd:

Device found.

0

 

If there isn't, I get:

Device not found

[oPossum 1,083]

Make sure the jumper for the P1.6 LED has been removed - that can cause problems.

 

A bus error would typically be when the bus is not pulled up when is should be - something is erroneously keeping it low.

 

Does my sample code work properly?

[Skunky Lee 0]

Hi!

 

Thanks for the help. Sorry I didn't answer sooner but something unexpected came up.

 

Yeah, the jumper on 1.6 was removed, I forgot to mention that. Anyway, I'll try running your sample code tomorrow and try to see what comes up.

 

Thank you very much for your help so far

 

[EDIT]

Hah! Found the problem.

 

I had a "lightbulb" moment and checked to see if I didn't accidentally mistake my DS18B20 for a transistor or something else in TO-92 casing. Funny enough, it seems that my DS18B20 isn't DS18B20 at all, but DS18B20P, which is a Parasite-mode only chip.

It'll be a while before I can get my hands on a normal one. Will report back then.

[DecebaL 0]

I have a problem.

Please help, how can I use multiple DS1820 s on same 1-wire bus?

Regards: Imre Huba

[cubeberg 540]

I'm working on implementing this for the IV-18 Clock booster and I'm not sure if I've got things running correctly or if I'm misinterpreting the output

 

Before the print_temp function executes, the values for TF and TC are around 3304 and 6459.

This is what I'm seeing in the console:

 

One Wire Test
FF 33 28 53 92 BF 03 00 00 B2 
206.5000 403.6875 

 

I'm seeing numbers increase when I touch the sensor and back down when I take my finger off and the serial is always consistent.

 

Any suggestions?

[oPossum 1,083]

What is the full part # of the sensor you are using?

[cubeberg 540]

DS18B20+. It was a sample from Maxim. I double-checked what's on the case against what I have in stock - it's definitely not the parasitic model. Using 3.6v supply and a 4.7k pull-up resistor.