Intermittent signal quality at 0

[Troy Lass]

I know this is not a WeeWX issue but can I get some guidance...

Over the years I seem to get signal quality at 0, often.  I'm not moving the console, in fact I get the same thing on a console and a envoy in the same location.

The console is approximately 25 feet from the station and I can see the station in a clear line of sight from the console.

Below is an example of the signal quality...  The other readings are blank durning the 19:00 - 1:00 0% signal time frame...

It's strange to me that it doesn't appear random, and that the largest outage is from a time I was in bed, no phone, no microwave, all lights off, etc...

NOTE - I've just replace a bad data logger so you don't see previous data collection in the image.  I've been having this issue since I've been using the consoles, this is not new.  The times change but I seem to always get signal outages.

Thanks,

Troy

[Thomas Keffer]

You may have to ask Davis about this one. I would guess there could be problems with interference from neighbors, perhaps even another ISS.

Another possibility is a failing supercap or battery in your ISS.

But, you should ask Davis.

-tk

--
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 weewx-user+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/weewx-user/aec589cb-b955-4837-bfa8-384169f516a0%40googlegroups.com.

[Troy Lass]

Ok, I'll open a ticket with Davis....

Just a couple more data items...

I live in a rural area on 35 arces with few neighbors.  There are no power lines or other obvious concerns.  One of my neighbors does have a Davis station on iss1, which I can get a signal to and the station over 300 yards away.  I run my station on iss2 for that reason.

I did just put in a new battery in the last week and this has been an issue for years over multiple batteries.

If I hear anything interesting from Davis I'll follow-up on this thread.

Thanks,

Troy

On Tuesday, February 11, 2020 at 9:18:26 AM UTC-7, Thomas Keffer wrote:

You may have to ask Davis about this one. I would guess there could be problems with interference from neighbors, perhaps even another ISS.

Another possibility is a failing supercap or battery in your ISS.

But, you should ask Davis.

-tk

On Tue, Feb 11, 2020 at 7:48 AM Troy Lass <colo...@gmail.com> wrote:

I know this is not a WeeWX issue but can I get some guidance...

Over the years I seem to get signal quality at 0, often.  I'm not moving the console, in fact I get the same thing on a console and a envoy in the same location.

The console is approximately 25 feet from the station and I can see the station in a clear line of sight from the console.

Below is an example of the signal quality...  The other readings are blank durning the 19:00 - 1:00 0% signal time frame...

It's strange to me that it doesn't appear random, and that the largest outage is from a time I was in bed, no phone, no microwave, all lights off, etc...

NOTE - I've just replace a bad data logger so you don't see previous data collection in the image.  I've been having this issue since I've been using the consoles, this is not new.  The times change but I seem to always get signal outages.

Thanks,

Troy

--
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 weewx...@googlegroups.com.

[Greg Troxel]

Troy Lass <colo...@gmail.com> writes:

> Over the years I seem to get signal quality at 0, often. I'm not moving
> the console, in fact I get the same thing on a console and a envoy in the
> same location.
>
> The console is approximately 25 feet from the station and I can see the
> station in a clear line of sight from the console.

So 25 feet line of sight means that the path loss (difference between
power into the tx antenna and power coming out of the rx antenna, more
or less) is almost certainly not an issue. I mean that if the
transmitter is working, and there is no strong interfering signal, then
this should be solid.

I have a VP2 with the console about that distance away, but with maybe 2
walls in between. I see spikes from 100% down to maybe 99% mostly and
96% once over a month, maybe twice over a year.

Given that two base stations are seeing the same problems, there are
three hypotheses:

A) The transmitter is failing to transmit correctly

B) There is some interfering signal

C) Both receivers are defective in a similar way

D) Something else we don't understand (always there, but good to list)

In my experience, when a Davis ISS has transmitting issues due to power,
it will stop when it's cold, or dark, and then start again when it warms
up or gets sun. Your pattern doesn't fit that. So this doesn't smell
like A to be.

C seems very unlikely.

For B, this does not strike me as a bizarre situation. There is a
tremendous amount of badly-designed and badly-built electronics that
radiates energy where it should not.

I have a friend who had signal dropouts with an Acurite station in one
location, not closely matching yours, but similar in that they were
occasional and inexplicable. After the entired station was moved (well
over 10 km, to an essentially uncorrelated 433 MHz environment), there
were no dropouts. So I don't find your situation surprising.

The Davis signal (assuming VP2) hops in frequency, to avoid narrowband
interference. I am unclear on the VP(1) signal. So it looks like, if
interference, that it's fairly broad band. In the US, this is in the
915 Mhz range. This link may be useful, and there may be a maintained
version somewhere:

https://github.com/bemasher/rtldavis

This is also interesting:

https://madscientistlabs.blogspot.com/2010/12/davis-weather-station-hacking.html
https://madscientistlabs.blogspot.com/2014/02/build-your-own-davis-weather-station_17.html

Overall, I suggest that you get an RTL-SDR dongle and use gqrx to look
at what's going on in the radio spectrum. You should be able to see the
transmit pulses, and I believe that broadband interference will appear
very differently from weak transmit pulses. I would recommend that
you get a dongle with a metal case, an SMA connector, and a TCXO.

I've used ones like:
https://www.nooelec.com/store/sdr/sdr-receivers.html?sdr_usb_ic=34
to receive 433 MHz signals from eg. acurite sensors, but have not tried
to listen to Davis.

Another thing to try is to unplug everything in your house that you can
deal with being unplugged overnight and see if that helps. You may have
a defective switched-mode power supply wall wart or semothing.

[Thomas Keffer]

Thanks, Greg. Nice to hear from an expert!

--
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 weewx-user+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/weewx-user/rmi8sl9yq0z.fsf%40s1.lexort.com.

[Troy Lass]

Greg,

Are you still willing to lend some guidance here?  I purchased a Nooelec Nesdr smart dongle and have HDSDR up and working with it.  I have it set to 916MHz.  There's a lot of settings here and it took me a while to figure out how to get local FM radio so if you could help me with the setting to start with that would get me started.  I did have a ham license years ago, KB0OEJ but I let it lapse and need to take the tests again.  That was about 25 years ago so that knowledge has past...  lol

Attached is a screen capture of HDSDR...  Note that right now I'm getting a good signal but I continue to have the signal loss issue.

I'd like to identify the davis signal and watch for interference...

Thanks,

Troy

[rich T]

Troy

Here are the 51 frequencies that Davis uses.  The frequencies are not in order, since they hop around.

902419338, 902921088, 903422839, 903924589, 904426340, 904928090, // US freq per 20190326
905429841, 905931591, 906433342, 906935092, 907436843, 907938593,
908440344, 908942094, 909443845, 909945595, 910447346, 910949096,
911450847, 911952597, 912454348, 912956099, 913457849, 913959599,
914461350, 914963100, 915464850, 915966601, 916468351, 916970102,
917471852, 917973603, 918475353, 918977104, 919478854, 919980605,
920482355, 920984106, 921485856, 921987607, 922489357, 922991108,
923492858, 923994609, 924496359, 924998110, 925499860, 926001611,
926503361, 927005112, 927506862,

For more information, see https://github.com/lheijst/weewx-rtldavis

[Troy Lass]

Rich, 

Thanks for the info, that will keep me busy for a while trying to figure out what they're doing there....  lol

I do have a new pi4 waiting a use so I play with that but was hoping I could use HDSDR so I could install on a laptop and carry around the property.

I'll read through this in the next day or so....

Thanks!

--
You received this message because you are subscribed to a topic in the Google Groups "weewx-user" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/weewx-user/24enqqdWiMI/unsubscribe.
To unsubscribe from this group and all its topics, send an email to weewx-user+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/weewx-user/b99f1f68-06cd-45c9-b8e7-c36207b425a9%40googlegroups.com.

[Greg Troxel]

See rich T's list of frequencies. As I understand it, the Davis
transmitter sends each burst at one of those, and then moves to the next
for the next burst. The point is to move around so that any
interference that is localized in frequency will not interfere with many
transmission in a row.

Given this slow hopping, there are multiple: interesting points for your
situation:

*) Since you have extended periods at zero loss, probably whatever is
causing trouble is relatively broad, affecting the entire 902-928 band.

*) There are several freqencies near 915, so if you are listening to 915
+/- 1 MHz, some transmission should show up. At 2.5s interval (from dim
memory, and I amy well be off), and 51 frequencies, that's 127.5s. So a
waterfall around 915 should be useful.


So, I'd be inclined to run in waterfall mode and adjust gain so that the
background noise is considered relatively not so high.
One tricky part is that rtl-sdr is often run in automatic gain control
mode, so you don't really see more total energy on receive, but
shifting from broadband to hotspots and quieter areas.

In your plot, the strong line at the tuned frequency is I believe a
normal artifact and not concerning. There is some structure in what I'm
seeing, but nothing super strong.

I do not know how long the Davis bursts last. There are a few things
that could be that, if they are short.

I would suggest that you 1) look at the waterfall for a full 127.5
seconds, so see if you see any vertical streaks and 2) keep it running,
and maybe set up an alarm on low receive% from davis, and then look, and
see if it looks different that how it looks when things are ok.