Dallas 1wire using the DS2482 I2C to 1-wire chip

[MikeW]

Has anyone got this device to work ?
I am tearing my hair out on this.
I know my I2C bus is working, since I have other devices on the bus.
I have had to go back to software I2C to use it ( the hardware I2C works fine with other devices).
I am using the Dallas AN3684 software routines, and cant get the OWSearch to find any 1wire devices.


my 1wire DS18B20's work fine using the 1 wire bit-banged routines.

maybe I am trying to be too clever in off-loading the DOW protocol

Mike

[herselmann]

Which one? There are two different ones. I have sampled some just last week but don't know when i will have a chance to play with them and thus be able to give you advice.

I would start by testing the search code with a bit-bang interface. From what I have read from the datasheet, you should be able to replace the bottom 'layer' with the i2c bridge master chip at a latter stage leaving all the other functions the same.

That being said, from what I can remember from the command table in the datasheet, it seemed like the chip has some commands specific to the search function.

Did you remember your Pull-ups?

I recall something about two's compliment when writing to some registers, but might be confusing this with other datasheets...

If the manufacturer has to dedicate a whole app note to a device then it must mean that it is more difficult that expected...

Here is a link for AVR code:
http://www.koders.com/c/fid9F966AF5A29CD2065933ED960EF0CEDC16DE5A96.aspx

EDIT: This might also help: http://www.ccsinfo.com/forum/viewtopic.php?p=64807
_________________
There are only 10 kinds of people, those who understand binary and those who don't.

[MikeW]

DS2482-100, the single channel device.

I can talk to the I2C part of the DS2482.
I have all necessary pullups.

I am having problems with the find devices routines,.... it cant find any.

re two's complement, I think you mean that the config register only uses the bottom 4 bits, but the top 4 bits must be written as the complement., ie 0x0F, or 0x1E
seems very strange.

my 1wire bit banging routines work a treat.
since I dont have a pic io pin connected to the 1wire bus, it is not simple to do a search rom using bit banging.
time for wire and a soldering iron methinks.

the AVR routines dont have a find rom function implemented

Mike

[MikeW]

the AVR routines look promising, I will keep you updated

Mike

[MikeW]

this looks useful also

www.scu.edu/engineering/cse/ugrad/upload/COEN-2007-PROJECT-07.pdf

[herselmann]

Indeed, very interesting, just need more time to read...

Here is the other half of the AVR code, no more functions - just usefull defines: http://www.koders.com/c/fid5F08D29A19BB041FCD6E144C5374144AAAACF927.aspx

This is a debug output of some sort, could also be of some help: http://www.mail-archive.com/[email protected]/msg01542.html

I had a second look at the datasheet, and it contains the complete search function, can't see why this code would not want to work. Have you tried tech support? Post or PM me some code and I will see if I can have a look at it.
_________________
There are only 10 kinds of people, those who understand binary and those who don't.

[MikeW]

I sent a pm to Kenny, after searching the forum on DS2482.

He sent me complete code.

I just started a new project, loaded his code, and away it went, detected my DS18B20.

Wonderful !!!

Many thanks Kenny.

I wont post his code without his consent.

Mike