Vantage VP2 lost contact with rain/temp/humid module (but not with separate wind module)

[martinu]

My Vantage Pro 2 setup has two transmitters: one in the rainfall/temperature/humidity module and another in the wind speed/direction module which is in a different location. Intermittently over the last 6 hours the David console has lost contact with the rainfall/temperature/humidity module ("--" values are displayed on its screen) for periods of about 30 minutes. This has led to periods when the external temperature has flatlined and it has missed periods of heavy rainfall. In between, everything appears to have been normal.

All three episodes have righted themselves without any manual intervention.

Firstly, would you expect that the Cumulus console site (port 8898) should display either the "Data stopped" or "Sensor contact" flags if one of the transmitter modules has lost contact but the other is still being received?

Secondly, has anyone experienced this communications fault? Everything has been working perfectly for about 7 months since I've had the VP2. There is line of sight reception from sensor to console over a distance about about 150 metres, with just a double-skinned brick wall of the house in the way. I can't see that anything has changed, as regards position or alignment of console and sensor aerials or obstructions between the two.


The error stats are currently:

DavisTotalPacketsReceived 14786
DavisTotalPacketsMissed 5522
DavisMaxInARow 1850
DavisNumCRCerrors 1574
DavisNumberOfResynchs 11

having been

DavisTotalPacketsReceived 14786
DavisTotalPacketsMissed 5474
DavisMaxInARow 1850
DavisNumCRCerrors 1573
DavisFirmwareVersion 3.88
DavisNumberOfResynchs 11

about 20 minutes earlier (so total packets missed and num CRC errors have increased).

The sensor batteries on all the channels are reported to be "OK".

Are there any other diagnostics I can gather? Should I reboot the sensor module and/or the console (and if so, how)? Is it likely that there could be interference from any other RF devices - has anyone ever found anything that interferes with the reception?



This the temperature readings for the last 12 hours. It shows three flatlline periods from 0934-1030, 1203-1244 and 1424-1507. The effect on humidity and rainfall is less obvious because these have not changed much, but the console showed "--" for both temp out and hum out during these times. There have been no other episodes during the past 48 hours which is the period over which I generate graphs. The every-10-minutes Jul21log.txt data doesn't show anything like it in the preceding days and weeks, though I realise that an episode would have to last for several 10-minute periods to be apparent from that data.

[mcrossley]

Hmm, the data stopped only kicks in if ALL the primary sensors flat-line. In your case the wind was still reporting so it never triggered. The data stopped granularity could probably do with improving here.

Sensor contact lost does not apply to Davis VP2 stations, it is for FineOffset stations and Davis WLL only at present.

It does look like interference to me. You are getting re-synchronisations which indicates it has lost contact completely and has to start the listen-to-find process again. There are also lots of CRC errors, which would indicate that packets are being received, but they are corrupted - that says interference to me.

[beteljuice]

Or perhaps ...

... and it has missed periods of heavy rainfall.

... the rain was heavy enough to reduce effective transmission distance.

Was / is the problem always associated with rain ?

[martinu]

Hmm, the data stopped only kicks in if ALL the primary sensors flat-line. In your case the wind was still reporting so it never triggered. The data stopped granularity could probably do with improving here.

Sensor contact lost does not apply to Davis VP2 stations, it is for FineOffset stations and Davis WLL only at present.

It does look like interference to me. You are getting re-synchronisations which indicates it has lost contact completely and has to start the listen-to-find process again. There are also lots of CRC errors, which would indicate that packets are being received, but they are corrupted - that says interference to me.

Yes, maybe Data Stopped should look at combinations of sensors which are all on the same transmitter, rather than looking at all sensors. And maybe look for this in combination with tell-tale signs like rapidly increasing CRFC and resyncs.

It's not happened again. I do wonder about interference, though whatever it was had not happened in the previous 6 months that I've had the VP2. Weird. I'll probably never know...

[martinu]

Or perhaps ...

... and it has missed periods of heavy rainfall.

... the rain was heavy enough to reduce effective transmission distance.

Was / is the problem always associated with rain ?

Interesting suggestion. I've looked at the rainfall rate over that period of time, and it doesn't correspond. There had been heavy and prolonged rain (for 3 hours, peaking at 20 mm/hour) that finished several hours before the first outage, and there was very light rain that started about 5 mins after another outage began (*). And the third time, it hadn't rained for an hour or so before. And there wasn't fog/mist - ie water in the atmosphere.

The signal path is about 30 metres across level grass, with a double-skin brick wall and a TV cabinet (TV, DVD player, Raspberry Pi for logging Cumulus data etc) in the way. But nothing's changed: it's been the same for 6 months.


There have been periods of really torrential rain (which the VP2 console really does report as "it's raining cats and dogs"!) in the past - up 150 mm/hour - but data comms wasn't affected.


(*) It was the fact that this rainfall wasn't being recorded that first alerted me to the problem.

[mcrossley]

I'd take a look on your console radio stats and see what the receive signal strength is for your ISS (page 39 of my manual). Just moving the console a few cm can change it.