On Thursday, July 21, 2022 at 12:46:24 PM UTC-4 [email protected] wrote: > Many Davis stations show dropped reception at precisely midnight. It usually only > stays that way for only a single archive record, but I recall one instance on my station > where it fell into the 50% range precisely at midnight, stayed there for 3 days, then > went back up to 99%, again precisely at midnight.
i suppose there might be another davis in the area, but this envoy has not yet reconnected to the iss that it is supposed to use. it is a brand new system (5 months old!), so kinda doubtful that it is a battery/supercap issue. good to know about the rxCheckPercent - i thought that was a direct measure of rf integrity, but if it is based on anemometer samples then anemometer problems could make you think you have rf problems. there has been no new hardware installed by us, but there is tons of boat traffic all around the station, so rf interference is definitely possible. On Thursday, July 21, 2022 at 1:18:47 PM UTC-4 vince wrote: > Is that thing actually underwater ? OutTemp of 61.8 and humidity=100 ? ha! water temperature here rarely gets over 55F, and that would only be in a very shallow bay with little flow. > Some wild hardware guesses: > - did it 'lose connection' or are the readings just whacko ? > - does part of a Davis ISS suite fail high (ie, outHumidity=100) and others fail low (zeroes or no data) > when something shorts or loses its ground in the ISS ? > - did you get hit by a wave (or bird or gust) and things are just unplugged or loose or wet ? definitely no wave or critter damage. we got pictures of the station late yesterday, and the iss is pristine. don't know what values the iss reports when sensors fail. years ago (before davis changed to a better humidity sensor) i had a davis station on which all of the iss sensors pegged at an arbitrary value when the humidity sensor died. had to send that one in to davis for repair - they did a full overhaul for $100, plus shipping. i think that station was 7 years old at the time. it is still going strong now. i just hope that this new station is not having a similar kind of failure. > Software stuff: > - how are you getting tide data into your db ? a separate weewx instance using the weewx-maxbotix driver. maxbotix is an ultrasonic sensor. we sample at 1Hz for loop data (sensor can do 6Hz) and use 5 minute average for archive. this is a prototype station that will hopefully be used to guide the builds for community-supported tide-only stations. currently there are volunteers going to sites at high tide and measure sticks. think cocoras, but for tide height. slogging through the mud at 02:00 to catch a high tide measurement is only fun the first two or three times. not twice a day. fwiw, hohonu makes a tide station that uses the same maxbotix sensors. > - I see failures to upload to influxdb starting shortly before midnight but not at 00:00:00 precisely ? Might it be an influxdb or disk issue ? > - are you still seeing failure to upload to influxdb errors ? > - what do the influxdb logs look like ? A loss of sensor data shouldn't cause failure to upload to influxdb (should it?) > - what's your influxdb setup look like ? Are you using telegraf at all ? it turns out that this is a bug in the influx uploader. it improperly handles empty/None/string values. i think i fixed it now in the weewx-influx uploader. not using telegraf. influx2 has explicit support for None, but for various reasons i cannot (yet) move to influx2. love the fluxql features in influx2, understand why they put lots of grafana-like capabilities into a default influx2 install, not crazy about all the influx2/grafana overlap. > Wilder guess: > - do you really have a soil sensor on channel 2 ? > - can you remotely change ISS+Envoy channel to something non-default ? yes, there is a soil station on channel 2. currently it has one temperature probe on it - that measures water temperature in the tidal zone. plan is to add a few more at different depths, but the one sensor is not working yet. the installer might have connected the sensor to the wrong terminal. my takeaways: - use a spare davis vantage console to see if it will see the iss - try a wired connection between the envoy and iss - try unsetting then resetting the iss binding in the envoy - try changing the iss channel identifier, then bind the envoy to that - verify that the water temperature sensor is connected to the correct terminals in the soil station - get an sdr dongle to the site and do some rf scanning thank you tom, greg, and vince! we now have some things to try before sending the iss back to davis. -- You received this message because you are subscribed to the Google Groups "weewx-user" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/weewx-user/9658cf7d-c78c-4980-b389-ce8429e5b453n%40googlegroups.com.
