Ack!  I hate when this happens, but nevermind...  apparently I needed a 
stronger pull-up than 4.7k (currently using 2.35k which works).

BTW, the read-slot is not required, as I have found out now.

Cheers,
-Neil.



On Tuesday 28 June 2005 02:41 pm, PicDude scribbled:
> Hi,
>
> I am successfully able to read temperature from a DB18B20 (using a
> PIC16F872) using the Skip ROM command, and I've just finished writing a
> Search-ROM routine that "seems like" it works -- I get back the correct
> number of devices on the wire (1 or 2 so far), and a dump of the output
> shows the family code "28" in the correct place.  I have yet to implement
> the CRC check.
>
> What I was doing next was to replace the Skip-ROM command with a Match-ROM
> command but am getting an error.  The datasheet for the DS18B20 (
> http://pdfserv.maxim-ic.com/en/ds/DS18B20.pdf ) has a flowchart (Fig 11)
> that shows each of the 64 bits being sent after the Match-ROM command, but
> after each bit, it shows a diamond-shaped decision box that verifies each
> bit.  If not a proper match, it shows the Master re-starting the
> transaction.  Is this required?  If the master re-starts based on this
> decision, then the master should be comparing this.  It can't compare
> something unless it retrieves the bit back from the 18B20 at that time, so
> it implies that a read-slot is generated.  But I don't see any other info
> anywhere else that indicates that a read-slot needs to be generated after
> each bit, and generating this read-slot will make a big difference here.
>
> Anyone here know what's really supposed to happen?
>
> Cheers,
> -Neil.


-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist