Topic: Success at receiving VP2 signals via Davis 7654 Long-Distance Repeater

[rdsman]

Greg:

It needs to be between the antenna and the amp...

Ray

[miraculon]

Since I have all the parts, I thought that I would lash up the mini-circuits amp and see what I get.
It was a very humid morning, foggy and condensation all over everything. A good worst case test for my wet foliage issue.

At first I put the BPF at the input and the 50 Ohm terminator and tee at the output of the mini-circuits amp. That is a ZFL-1000GH as you will recall from the last exciting episode..

I ran a N-male to SMA-male jumper to the BPF, and a SMA-male to RP-TNC at the Davis RF-IN cable. The mini-circuits amp has SMA-female connectors for both input and output.

Nada. OK, maybe things are just too wet and the signal is so low that the gain can't make up for it.

I tried with and without the BPF, the 2dB attenuator instead of the terminator with no luck. I came down to the basement to watch the "compass rose" RSSI indicator and it was "all dashes".

Finally out of desperation, I removed everything that has any kind of attenuation. Removed the BPF with its 2.5dB insertion loss, and put the 50Ohm terminator back on. Still no luck. Red blinking status LED out on the repeater, no synch on the console.

I then remembered that there was some gain dependence of the ZFL-1000GH with power supply voltage. I was running it on a regulated 13.5V supply. The gain difference at 925.40MHz between 12V and 15V is substantial, 27.20dB at 12V and 31.88dB at 15V. I brought up the Marlin P. Jones bench supply and set it to 15V. Watched it for a while and saw some different patterns on the status LED, pausing off for a while. All of a sudden green blinkage occurred!

Apparently, the gain is critical and the stack-up of the 2.5dB in the BPF and probably another good 2dB due to the supply voltage hurt the cause significantly. Adding the 2dB attenuator was insult to injury.

So, I am still mulling on whether to make the final leap with the 17dBi antenna or not, but I am going to run this gain configuration once my enclosure (a Davis transmitter back housing and spare plain hinged cover) and the 15V power supply arrives from Mouser.

Here are the "DIP 4 Test Mode" RSSI readings in the compass rose. According to Davis the desired range is 30-60 (DAU, Davis Arbitrary Units). This is the true "as received" signal strength (normalized to the Davis funny units). Note that I ran the no-amp case afterwards, so any improving conditions due to the foliage would not bias the test favoring the amplifier. (it was clearing and drying up during the afternoon)


With the 14dBi antenna only:
Ranging from 22-26 typically around 25 DAU. (out of 60)

With the full treatment, 15V supply, ZFL-1000GH, 50Ohm terminator, no BPF:
Ranging from 38-40 typically 40 DAU. I did see a brief dip to 32 once, but wasn't typical.

After I found the right "recipe", it did 'move the needle' by about 10-15 DAU. While it remains to be seen whether this really takes care of most of the "wet foliage" issue, it seems that it should give some more margin and reduce the probability of loss of signal.  Still not sure about the bigger antenna, the 17dBi is about twice as long. I'll leave space for it's larger bracket, just in case.

Greg H.

[johnd]

FWIW my benchmark for acceptable reception long-term is 27-28 as a minimum (though this doesn't really leave any margin for any temporarily adverse reception conditions). Hence the preference for a VP2 RSSI starting with a 3 - if you can get to eg 32-33 then that seems to manage to sustain good reception over a variety of conditions.

[miraculon]

FWIW my benchmark for acceptable reception long-term is 27-28 as a minimum (though this doesn't really leave any margin for any temporarily adverse reception conditions). Hence the preference for a VP2 RSSI starting with a 3 - if you can get to eg 32-33 then that seems to manage to sustain good reception over a variety of conditions.

Thanks, that gives me a good idea. I think that I am on the right track.
The 15V supply is on order from Mouser and the hinged sensor housing back is on order from Scaled Instruments.
I think that I will leave space for the 17dBi antenna bracket, but hold off on it and see how it goes with the just the mini-circuits amplifier and the existing 14dBi antenna.

Greg H.

[miraculon]

Permanently installed the mini-circuits amplifier yesterday. The RSSI is the same as it was during my testing last week with the MPJA power supply. I got a linear regulated 15V supply from Mouser to run it. It is kind of ironic that I have a solar-powered heater but the preamp ahead of it is now AC powered. (well, actually DC)

The percentage of good packets still isn't stellar, even with the 10-15 unit increase in RSSI (#4 repeater switch on, watching the wind degree display). Maybe some packets are getting clobbered by cordless phones, smart meters, etc on the way here over the 2,000ft (610m) distance.

Here are some close up pictures just before I installed it directly above the repeater on the mast.

Greg H.

[miraculon]

Well during heavy rain yesterday afternoon, I had loss of signal even with the preamp. At least the Marina accumulated 0.06" and triggered the "It's Raining Cats and Dogs" message before the receivers here lost the data. This morning at about 4 AM it came back alive again. It seems that the preamp made an improvement, recovering sooner than prior but it isn't a cure all.

According to the linked paper previously, the attenuation due to wet foliage is significant. I'll just live with it and hope for better results once the leaves are off the trees. It is what it is.

Just a heads up to anyone setting up a long-range repeater system: avoid trees in the Line-Of-Sight if you possibly can. Unfortunately, I can't avoid it in my case.

Greg H.

[miraculon]

I have been running several days since the leaves dried up and we have had fair weather. The percentage of good packets is now at 56% and is fairly steady, with one exception yesterday afternoon. The console behavior looks "normal" despite the low good packet rate.

I decided to monitor the "VP reception" on the "Solar Graph" chart in Weather Display for this station/console. I am using a DSI-01 on this VUE console, so I could not use the Weatherlink plot function since there is no stored data. WD works fine for this without popping for yet another logger...

What is interesting is that from 6:24P to 6:27PM yesterday afternoon the VP reception went to 92%! My RSSI is close to "40" when I have the #4 test DIP switch set in the repeater, so it should be a pretty good signal normally. (without wet leaves). I was kind of surprised that my percentage of good packets was so low.

This "event" yesterday has me thinking that there is an interference source that was briefly shut off for a few minutes.

I'll attach the screen capture of the WD graph of it. Really weird. Consumers (electric utility) has been installing smart meters all over town, but the reception issues may have existed previously.

Greg H.

[johnd]

What is interesting is that from 6:24P to 6:27PM yesterday afternoon the VP reception went to 92%! My RSSI is close to "40" when I have the #4 test DIP switch set in the repeater, so it should be a pretty good signal normally. (without wet leaves). I was kind of surprised that my percentage of good packets was so low.

Reception variations are often a bit of a mystery, at least in the absence of decent wireless monitoring kit. Interference could potentially be a plausible  explanation. Which wireless channel is being used?

[miraculon]

The marina is on Channel #1, I am using Repeater A.

I went for a walk with my RF Explorer. The only interesting area was in the park adjacent to the marina. I found a fairly broadband noise at about 907MHz, at least compared to the known Davis FHSS signals. Behind some shrubbery, I found a utility panel with a smart meter. I was very close to it and didn't realize it at first. It is well within the beam pattern of the Yagi.

There must be other smart meters in the Line-Of-Sight as well, though. There are many gadgets in the ISM band (902-928MHz here in the States) that could be interference sources. I found that my Z-wave switches out in the garage were susceptible to a wireless thermometer transmissions.

Greg H.