Author Topic: Temperature errors due to packet loss?  (Read 421 times)

0 Members and 1 Guest are viewing this topic.

Offline miraculon

  • Sunrise Side Weather
  • Forecaster
  • *****
  • Posts: 3439
  • KE8DAF
    • Sunrise Side Weather in Rogers City MI USA
Temperature errors due to packet loss?
« on: March 25, 2019, 09:04:33 AM »
I was hoping that some of you Davis packet experts could provide some insight.

After running for months with 90%+ good packets per the VUE diagnostic screen, good packets have been in the 45%-63% range very recently. My VUE console that shows the local marina VP2 data via a long range repeater is sometimes showing a temperature discrepancy vs. what is reported via Weatherlink.com. (I recently fixed the WLIP upload issue over there). I see the exact same temperatures via the repeater on all my consoles, Envoys, and meteohub/bridge devices.

Normally the temperature has agreed quite well as I recall, especially during high packet rate success.

I know that during poor packet reception that I miss wind gusts, but this makes sense because the samples are being missed. For temperature, this is usually fairly static and I would not expect to see errors. The temperature has disagreed by as much as 2F recently.

I took a VUE console over to the marina parking lot and configured it for Station #1 and watched the data from weatherlink.com and this console. They matched exactly.

What could the long-distance repeater be doing to introduce a temperature error? I have never seen this behavior before. (although I wouldn't have known from last October to this past week due to the loss of WLIP function)

I suspected and have observed signals in the adjacent 800MHz band (cell and emergency services) on my RF Explorer spectrum analyzer. I do have a BPF with the 902-928 MHz pass-band at the pre-amp feeding the repeater input.

Weatherlink: https://www.weatherlink.com/embeddablePage/show/372a56e9f20d41bab0877649cb397b2b/slim
Via repeater: http://rogerscityweather.com/marinavue/marina.html

Although this morning, it seems to be closer. Good packets are 63%. It was worst with good packets at around 45%.

Greg H.


Blitzortung Stations #706 and #1682
CoCoRaHS: MI-PI-1
CWOP: CW4114 and KE8DAF-13
WU: KMIROGER7
Amateur Radio Callsign: KE8DAF

Online PaulMy

  • Forecaster
  • *****
  • Posts: 4513
    • KomokaWeather
Re: Temperature errors due to packet loss?
« Reply #1 on: March 25, 2019, 09:48:36 AM »
Hi Greg,
The difference between the 2 is quite significant.  At 09:30 a.m. on the WL.com (looks like 1 min interval) it was 26.9F and 49% compared to 29.3F and 42% on the repeater link which looks like a 5 min interval. Wind also off quite a bit.
Again at the 09:35 and 9:40 time the temp and hum on the WL.com follows the previous update times- 26.9 and 26.8D and 49% and 50%, and the same for the repeater 29.7 and 29.5F and 43% and 44% so neither is fluctuating much.

I can't help on reception issues or packet losses, but could there be another Davis reporting through the repeater?

Enjoy,
Paul
« Last Edit: March 25, 2019, 09:50:54 AM by PaulMy »
Davis Vantage Pro 2 Plus 24-FARS Wireless
Envoy/USB DataLogger/Cumulus v.1.9.4   -   Vue Console/WiFiLogger/CumulusMX
Komoka, ON  Canada
www.komokaweather.ca   www.komokaweather.com   www.komokaweather.com/weather28/  
www.komokaweather.com/pws/index.php
Blitzortung Station #1076 www.lightningmaps.org/blitzortung7/america/index.php?bo_page=statistics&bo_show=network&lang=en&bo_station_id=969


Offline miraculon

  • Sunrise Side Weather
  • Forecaster
  • *****
  • Posts: 3439
  • KE8DAF
    • Sunrise Side Weather in Rogers City MI USA
Re: Temperature errors due to packet loss?
« Reply #2 on: March 25, 2019, 10:55:15 AM »
Quote
could there be another Davis reporting through the repeater?

I don't think so. When I reconfigured the VUE console for Station #1 (no repeater), and drove home from the Marina, it lost signal for good. (as expected)
If there was another #1 around, I would expect to see something from a another ISS.
Once home, I reconfigured this VUE to correspond with my own station transmitters.
I will re-try this again and see if a non-repeater signal exists.

The temperature is changing rapidly this morning, and this could be a sampling problem like wind gusts I suppose. Although Saturday, the temperature seemed to be fairly stable. Good packets are at 63% this morning still. There is some junk on the 800MHz band according to the RF Explorer, although I don't hear activity on the 800MHz trunking scanner.

Just as a side note, the local temperatures from my own transmitters is fairly close to the Marina VUE via repeater. Hmmm...  :-k I have deliberately avoided using Sta #1 on any of my own stations.

Greg H.


Blitzortung Stations #706 and #1682
CoCoRaHS: MI-PI-1
CWOP: CW4114 and KE8DAF-13
WU: KMIROGER7
Amateur Radio Callsign: KE8DAF

Offline miraculon

  • Sunrise Side Weather
  • Forecaster
  • *****
  • Posts: 3439
  • KE8DAF
    • Sunrise Side Weather in Rogers City MI USA
Re: Temperature errors due to packet loss?
« Reply #3 on: March 27, 2019, 08:44:17 AM »
The "good packet" rate has improved to 63%.

The temperatures are agreeing fairly well this morning.

I wonder if the recent CME event was causing some RF disruption with the repeater signal. I had a lot of issues involving Smarthings, noise on VHF, poor WiFi, etc. the last few days that seem to have cleared up.

Maybe the temperature rate of change was causing a missed sample situation similar to what I see with wind. That is the only explanation that makes sense to me. I would think that data corruption would result in wild results affecting all parameters, which I did not see.

Greg H.


Blitzortung Stations #706 and #1682
CoCoRaHS: MI-PI-1
CWOP: CW4114 and KE8DAF-13
WU: KMIROGER7
Amateur Radio Callsign: KE8DAF

Offline johnd

  • Forecaster
  • *****
  • Posts: 3299
    • www.weatherstations.co.uk
Re: Temperature errors due to packet loss?
« Reply #4 on: March 27, 2019, 10:41:47 AM »
Only quick comment I'd make is to be careful about assuming that eg 63% reception equates to 63% within say every 10-min period. Reception could be dropping out completely for 30-60 mins at a time and then recovering to 90% for a couple of hours. It's important to look at the graph of wind reception vs time to get a feel for the pattern of reception.
Prodata Weather Systems
Prodata's dedicated Davis EnviroMonitor website
UK Davis Premier Dealer - All Davis stations, accessories and spares
Littleport, Ely, Cambs UK

Offline miraculon

  • Sunrise Side Weather
  • Forecaster
  • *****
  • Posts: 3439
  • KE8DAF
    • Sunrise Side Weather in Rogers City MI USA
Re: Temperature errors due to packet loss?
« Reply #5 on: April 22, 2019, 04:45:35 PM »
My good packet rate is now in the mid-high 90's percent range. It must have been interference, either man-made or solar in nature.

One day, even the marina's own console was not picking up the ISS, only baro and indoor readings worked. That went away and fixed itself.

It is all working quite well again, inexplicably.

Greg H.


Blitzortung Stations #706 and #1682
CoCoRaHS: MI-PI-1
CWOP: CW4114 and KE8DAF-13
WU: KMIROGER7
Amateur Radio Callsign: KE8DAF