I have been observing the same behavior from my 18B20s I have 10 and they all do the same thing. I can get several readings from them and then the 185 (or 85 in some cases) shows up.
I did an experiment and found that if I ran owhttpd against the owserver and accessed the uncached temperature reading it was usually(more than 90%) correct. I wrote a test using the ownet java api against the owserver and found that it usually (more than 90%) *failed*. One strange behavior observed during the testing: If I let the java call loop and it was reading 185 and then access the owhttp temp my java test would temporarily read correctly for 1 iteration. Strange! I did have my adapter and sensors running on an old redhat EL 4 server and they seemed to work fine although I didn't run as exhaustive of testing. On a windows machine using the usb (tmex I think) driver and OW viewer I see no issues. On that old RH system I was also able to use the 1-wire public domain kit, on my new ubuntu setup the kit fails to properly initialize the usb adapter. So, this leads me to believe it is something to do with the system (libusb, kernel, etc.). I am actually not using the ground pin (I think pin 1) on the sensors which might be an issue. I was using it before and not getting any exceptions. I will do a test with pin 1 and 3 connected and see the results. Current Config: OWFS version= 2.6p5 libfuse-dev = none Ubuntu version= 7.04 Kernel version= 2.6.20-16-generic 1 wire adapter = DS9490 (ds2490) usb libusb = 0.1.12 Any new info? [EMAIL PROTECTED] wrote: > > > I was/have been getting the same 185F temp from the 18B20 in parasitic > mode IIRC. I think I was testing and seemed to notice > that it went > away when powering them vs parasitic. I haven't played with much > lately (been testing pH via the DS2450) but I had a script running that > cycled through about 2 to 5 sensors via uncached collecting the temp > and the same sensor would work for a couple iterations, then fail > returning 185F. I could pull it from the net and reconnect then > after > restarting script it did the same - work a couple times and die. I > think I had one that after running parasitic, then powering it worked > fine. I was wondering if maybe it was that the sensor didn't have > the > power to charge & read correctly due to the multiple successive > calls to the uncached? No real proof or hard evidence, just a > thought. > > OWFS version= 2.5p10 > libfuse-dev version= 2.5.3-2.1ubunt > Ubuntu version= 6.10 > Kernel version= 2.6.17-10-server > USB owfs adapter > > I don't know the OW net configuration that I had at the time of the > errors, I think they were mostly parasitic temp sensors of some flavor, > and the last one or two read would be the troublesome ones. If you > like I could try to recreate the situation as well. > > -Scott > > > Paul Alfille wrote: > Here's a quick and dirty test with USB, and a bunch of > DS18S20s: > amd64:/home/paul/owfs # owserver -u -p 3000 > amd64:/home/paul/owfs # for x in `owdir -s 3000` ; do owread -s 3000 > $x/temperature ; done > Data error on /bus.0/temperature > > Data error on /settings/temperature > Data error on /system/temperature > Data error on /statistics/temperature > 16.9375 > 16.9375 16.875 > 16.875 16.875 > 16.9375 16.875 > 16.875 16.875 > 16.9375 > 16.9375 16.875 > 16.9375 16.875 > 16.875 > 17 > 16.9375 16.8125 > 16.9375 16.9375 > 16.8125 > 16.875 16.875 > 16.875 16.875 > 16.9375 16.875 > 16.8125 > 16.8125 17.0625 > 16.9375 16.9375 > 17.0625 16.9375 > 16.9375 > 17 16.875 > 16.875 > 17 16.875 > 16.8125 > 16.875 > 17 > 16.875 16.875 > 16.875 > 16.875 > 85 > 16.875 16.75 > 16.9375 16.875 > 16.9375 16.8125 > 16.875 > 16.9375 > 17 > 16.875 > 17 16.875 > 16.875 > 17 16.875 17.0625 > 16.9375 > 16.875 16.9375 > 16.875 > 17 > 16.9375 > 17 > 16.8125 > 17 > 16.875 17.25 > 16.8125 16.875 > 16.9375 16.9375 > 16.9375 > 16.875 16.875 > 17.125 16.875 > 16.9375 16.875 > 16.8125 16.875 > 16.9375 16.8125 > 16.9375 > 16.875 > 17 > 17 17Data error on > /2C.8EF604000000/temperature > 16.875Data error on > /12.4AE41F000000/temperature > > Data error on /12.15DF1F000000/temperature > Data error on /12.F3EA1F000000/temperature > Data error on /01.0683C9090000/temperature > Data error on /01.8660BE090000/temperature > Data error on /01.4FA8CC080000/temperature > > Data error on /89.852BC300705E/temperature > Data error on /05.5ADD1E000000/temperature > Data error on /05.DD011F000000/temperature > > Looks like one "85" in the mix. > > Serial seems to work as well (separate test). > > > Paul Alfille > > > On 1/20/07, Paul Alfille < [EMAIL PROTECTED] > > wrote: > There > shouldn't be. That's how it's tested. > > Which adapter? Which program? Which platform? > > Paul Al;fille > > On 1/20/07, Alan Prather < > [EMAIL PROTECTED] > wrote: > Upgraded > to 2.6P0 version and am having trouble getting accurate readings from > my DS18S20 temperature sensors. These are using parasitic power > from > the > 1wire bus. I get accurate readings from these exact same > sensors using > the > > Dallas OneWireViewer tool, but in OWFS they are giving me a constant > reading > of 185 degreed Farenheit, or 85 Celsius if I choose that option. > > In backreading it looks like this was a problem previously, but not > sure if > > that problem persists. Like I said, the readings are accurate > using > the > Dallas 1-Wire Viewer tool, so I have to believe the sensors are > functional. > > Is there still a problem with this sensor model in OWFS? > > > Thx, > > Alan > > > ------------------------------------------------------------------------- > Take Surveys. Earn Cash. Influence the Future of IT > Join SourceForge.net's Techsay panel and you'll get the chance to share > your > > opinions on IT & business topics through brief surveys - and earn > cash > > http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=D EVDEV > > _______________________________________________ > Owfs-developers mailing list > [email protected] > > https://lists.sourceforge.net/lists/listinfo/owfs-developers > > > > > > > > > > ------------------------------------------------------------------------- > Take Surveys. Earn Cash. Influence the Future of IT > Join SourceForge.net's Techsay panel and you'll get the chance to share > your > opinions on IT & business topics through brief surveys - and earn cash > http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=D EVDEV > > > _______________________________________________ > Owfs-developers mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/owfs-developers > > > > > > > > ------------------------------------------------------------------------- > Take Surveys. Earn Cash. Influence the Future of IT > Join SourceForge.net's Techsay panel and you'll get the chance to share > your > opinions on IT & business topics through brief surveys - and earn cash > http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV > _______________________________________________ > Owfs-developers mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/owfs-developers > > ------------------------------------------------------------------------- This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ _______________________________________________ Owfs-developers mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/owfs-developers
