From what I can tell, the dallas 1 wire temperature probes require a signal to be sent to start converting a current reading into a temperature and writing it to the scratchpad. Then another signal is to be sent to read the current value from the scratchpad.
The converting is the expensive part, requiring as much as 750 ms.
I have several of these probes on one bus. I've written some code which attempts to go through each probe, tell it to convert its temp, wait 750 ms, then go back through each probe and sample the scratch pad.
This doesn't seem to work, though, as I get bogus temperatures frequently.
If anyone has more experience with these probes and can explain my problems, please elaborate
Thanks!