I'm posting this to the list, since it is of genreal interest and others =
From: dfischer@... [mailto:dfischer@...]
Sent: Thu 8/11/2005 5:21 AM
To: Alfille, Paul H.,M.D.
Subject: OWFS and 2760
Hi Paul -
Thanks again for the great project.
I have some questions about the data I am seeing. If this is better =
in the forum, let me know.
I have a the nice start of an OWFS system to monitor and control a small
kiln. I am currently "talking" to a type K thermocouple (AAG TAI8560, =
the DS2760). I see it, but I am having trouble finding information to
interpret what I am getting. I have scanned the OWFS developer forum =
nothing popped out at me. I also looked over the DS2760 spec. sheet, =
looks like the raw data comes off at a lower level than what you get =
> You are supposed to read the typeK directory for temperature. The =
should be properly adjusted, including cold-junction temperature etc.
I am using the perl interface example, with OWFS running as a server on =
3006. I also have a second test perl script running that goes after =
/temperature data (family 30) and the /typeK/temperature data.
Can you please answer a few questions?
1. It looks like OWFS provides an entire device file tree even if only =
device is on the bus? The data for these other (empty) devices that is
shown must be the initialization values?
> Under "structure" the format of the possible data is shown. It isn't =
needed for normal operations. but only to give information for a more =
> The 2760 also shows a directory for every possible type of =
is no way to know which type is actually connected, so all are shown. =
real difference is the thermocouple constants used for the calculation.
2. Is there any documentation that explains the layout I see for the =
values? Your device descriptions give explanations of the categories, =
no examples of the data part. As an example, PIO is described under =
2760 section, but what does y,000000,000001,wo,000001, mean?
--- etc. ---
> Again, this is the "structure" data. Rather arcane and not required =
> y,000000,000001,wo,000001, -->
> y "yes-no" (boolean) value
> 000000 index in array (if more than one like memory pages, -1 for all)
> 000001 number of elements (1 if not an array)
> rw read-write value (as opposed to ro=3Dread-only and wo=3Dwriteonly)
> 000001 number of bytes
Controls the PIO pin allowing external switching.
Writing "1" turns the PIO pin on (conducting). Writing "0" makes the =
non-conducting. The logical state of the voltage can be read with the =
property. This will reflect the current voltage at the pin, not the =
sent to PIO"
3. I am mainly interested in the temperature values from the family 30
section (only device I am currently running). Does this mean that the =
junction temp. is 12C and the thermocouple temp. is 12C? I had an
independent device read a temp. of 24C in the same spot as the 1-wire
reading -- which would imply I simply add the cold junction temp. and =
typeK temp. -- but I have a feeling this is just a coincidence(?)
> The cold junction temperature (internal thermister) is indeed 12C
> If the typeK/temperature isn't 24C, then there may be a problem with =
calculation. I don't have the device, so we'll have to work this through
together, but I'm very interested in getting it to work properly.
I will be throwing some higher temps. at the TC to see how the readings
change -- but I think there must be an easier way to understand the =
output than trial and error experimentation.
temperature: f,000000,000001,ro,000012, <--- cold junction in C?
temperature: f,000000,000001,ro,000012, <--- thermocouple =
> No no no. Just 12 bytes length (for the temperature character string).
Thanks for any help.