Vantage Pro 2 and bad temperature/humidity sensors

[Daniel Pifko]

Hi all,

I posted about this a while back (early 2007?) though I can't find the
reference in the Old Archives to link to. So you'll just have to take
my word for it ;-).

The punchline:

I'd like to find out where wview decides to mark an archive record bad
if the remote temperature reading is out of whack and change it so the
wind speed/direction reading has to be out of whack to mark a record
as bad. I can't seem to find this. Can someone point it out, please?
Unlike many other installations, the wind speed and direction are most
important to me and the temp/humidity is secondary. I'd be glad to
have one of my helpers work on the code if we can at least identify
where the appropriate function is.

The background:

I run the weather station at http://www.flyfunston.org/newwebcam/ .
The weather system includes a Linux box running wview 3.2.1 and a
Davis Vantage Pro2 6152 with wireless sensor unit for wind, temp,
humidity and rain.

The temperature / humidity sensor, a solid state unit, stopped working
one day. The wind speed and direction sensors kept on working and the
console kept displaying those readings. However, wview uses the
temperature reading to determine the health of the weather station.
As a result, wview started getting errors along these lines:

Feb 8 16:50:04 comet wviewd[1932]: <4222328750> : BAD archive record
for 2008-02-08 16:50
Feb 8 16:50:04 comet wviewd[1932]: <4222328750> : Copying last good
archive record for this interval...
Feb 8 16:50:04 comet wviewd[1932]: <4222328750> : storing record for
2008-02-08 16:50

and the max wind reading and all charts stopped working (e.g.
http://www.flyfunston.org/img/Daily.htm ).

Mark, you kindly jumped in and pointed out that a temp/humid sensor
failing is a rare event.

It turns out that's not the case. I am now on my 3rd temp/humid
sensor and it just failed. Davis has identified a similar problem
with all VP2's located anywhere near a body of water, salt or fresh.
The temp/humid sensor just fails all of a sudden. My last two worked
for only a few months. The last replacement was supposed to have
protected the sensor better. It didn't. The new one they're sending
me uses another hardware revision to try to address the problem. I
hope it works.

In any case, I hope someone can help me out.

Thanks,

Daniel

[IslandLife]

Almost concurrently with reading your post, my station is starting to
report the same "BAD archive record"

The computer (Mac OS X 10.4) is hooked up to a Davis Weatherlink, and
every 5 - 6 hours or so it will start reporting the bad data. Upon
restarting wview, temperature, wind speed, etc. are all 0 (zero). It
takes a computer restart to reset the system. I do not think that any
of the sensors are bad (the digital tabletop readout reports fine).
I've changed the backup batteries in the Weatherlink, changed the USB
port on the computer, to no avail. This anomaly started about the
same time the CWOP server mess started. System log is as follows:


Feb 10 07:24:32 Cube sudo: Paul : TTY=ttyp1 ; PWD=/Users/Paul ;
USER=root ; COMMAND=/sbin/SystemStarter start wview
Feb 10 07:24:32 Cube radmrouted[417]: <1430061> : started on radlib
system 1, workdir /var/wview
Feb 10 07:24:33 Cube wviewd[421]: <1431113> : station interface:
serial ...
Feb 10 07:24:35 Cube wvalarmd[426]: <1432127> : /etc/wview/
wvalarm.conf does not exist - exiting...\n
Feb 10 07:24:35 Cube wviewftpd[435]: <1432233> : FTP: INIT: no
relative transfer path given - defaulting to your FTP login
directory...
Feb 10 07:24:35 Cube wviewftpd[435]: <1432235> : FTP: INIT: 6 rules
added
Feb 10 07:24:35 Cube wviewftpd[435]: <1432235> : starting ftp timer
for 1 mins 40 secs
Feb 10 07:24:35 Cube wviewsshd[437]: <1432245> : /etc/wview/
wviewssh.conf does not exist - exiting...\n
Feb 10 07:25:06 Cube wvcwopd[429]: <1463133> : radSocketClientCreate:
in progress connect failed: Connection refused
Feb 10 07:35:05 Cube wvcwopd[429]: <2062224> : radSocketClientCreate:
in progress connect failed: Connection refused
Feb 10 07:50:05 Cube wvcwopd[429]: <2962581> : radSocketClientCreate:
in progress connect failed: Connection refused
Feb 10 10:51:04 Cube wvhttpd[432]: <13821543> : WUNDERGROUND-error:
connect() timed out!
Feb 10 10:51:04 Cube wvcwopd[429]: <13821550> : radSocketClientCreate:
gethostbyname failed: Invalid argument
Feb 10 11:12:09 Cube ntpd[147]: time set -0.223500 s
Feb 10 12:35:04 Cube wviewd[421]: <20061308> : BAD archive record for
2008-02-10 12:35
Feb 10 12:35:04 Cube wviewd[421]: <20061308> : Copying last good

archive record for this interval...

Feb 10 12:40:04 Cube wviewd[421]: <20361308> : BAD archive record for
2008-02-10 12:40
Feb 10 12:40:04 Cube wviewd[421]: <20361308> : Copying last good

archive record for this interval...

Feb 10 12:45:04 Cube wviewd[421]: <20661309> : BAD archive record for
2008-02-10 12:45
Feb 10 12:45:04 Cube wviewd[421]: <20661309> : Copying last good

archive record for this interval...

Feb 10 12:50:04 Cube wviewd[421]: <20961358> : BAD archive record for
2008-02-10 12:50
Feb 10 12:50:04 Cube wviewd[421]: <20961358> : Copying last good

archive record for this interval...

[IslandLife]

"is hooked up to a Davis Weatherlink"

Oops, I guess it's called a Weather Envoy. It's the USB version, and
I've been using it for +/- 3 years now.

[mteel]

Line 653 of stations/vantagePro/vproInterface.c

Mark

[Daniel Pifko]

Thanks. I'll have a look.