I’ve had a Prostick Plus (blue unit) for about 3 years and to date it’s performed well. I had to take my ADS-B station offline earlier this year and recently (start of September) recommissioned it with a new high gain antenna.
Everything was working wonderfully until shortly after midnight on the 5th of October, when suddenly (like a cliff edge!), I lost about 12-15dB off my signal level and a corresponding drop in range and tracked flights:
12pm 4th October - 12pm 5th October
2-9 October
I’ve been walking through logs, etc to see if there’s anything that could explain/correlate with the sudden change in signal level/performance. Whilst I do have adaptive dynamic range configured, this definitely did not change. It’s been running at Step 28 both historically and currently so it’s not to blame. Regardless, I did try manually changing the gain to Step 29 (max) to see if it would match previous performance and whilst this did raise the average signal level, it did not come close to matching previous performance - something is still very much amiss (and it also doesn’t explain the sudden drop in performance with no change to adaptive gain).
Is it possible the amplifier in my Prostick plus has died, as that would certainly explain the cliff-edge change in performance?? Has anybody else seen this behaviour with an older Prostick Plus? I have tried resetting the Prostick Plus via USB (it’s hard to access as it’s in the loft) and done a fully reboot of the Pi, both of which made no difference. I’ve not yet tried a full power down so that will be the next step, but I’m really not anticipating this will help, given the full USB reset did not.
Check if all the connectors are tightened correctly. Even a loose connector can give such an effect. If the amplifier would have died the stick wouldn’t work even with lower gains. Did you move the position of the antenna ?
Thanks Tom, there’s been no change to anything physical at all, including antenna position.
I’ll climb up to the loft to check the connections between the ProStick and the antenna feed line, although nothing has been disturbed since install. The antenna itself is at the top of a 7 metre mast and very hard to access to check that connection. However it’s an N-connection, encased in self-amalgamating tape so it would be bizarre if this has somehow worked lose.
If you check it you can start eliminating causes of the issue. Physical connections are sometimes overlooked and once checked we can look further into the software side of things
Thanks again for the reply. I’ll check the connections on the weekend when I can arrange physical access and go from there. Fingers crossed it’s something simple!
I had a similar problem. Dug into the data files, looked everywhere. Then noticed the antenna had slipped of the pole and was hanging down at an angle in the attic. Remounted the antenna with better tightening of the bolts, and amazing what a difference. Those comments above about checking the obvious are right.
When you power-cycle the Pi, give it ten min. to cool down before rebooting. If it’s a soft PSU, it’ll probably work well for awhile before dropping to its current level.
Thanks Jon, it was indeed raining at the time the signal drop occurred and water ingress was one of the first places my mind went too.
However it’s been quite wet in general ever since I got the setup back online at the end of August, so if it is water ingress, I’m surprised it’s taken this long to manifest. Also I wrapped the entire union between the coax and antenna (N connection) in self-amalgamating tape as an added precaution (even though N-connectors are supposed to be weather-proof). So unless the antenna itself is leaking at the base, I’m really not sure how water could have gotten in. The antenna is one of these units and they seem very well constructed.
Your suggestion is definitely a possibility though but I’m seriously hoping it’s something further down the line as pulling the antenna down is going to be quite a job :
Very interesting thought… I’d not given a lot of thought to the PSU. Would I/Should I expect to see “undervoltage detected” warnings being logged if the PSU was on the way out though? (I’ve not had any undervolt warnings logged to date).
FYI, the Pi is powered via PoE. The PSU itself is rated @ 48V/2Amps which gets stepped back to 5V at the remote end.
I just wanted to post a follow-up to this incase anybody comes across this thread in the future.
Over the weekend, I finally managed to get good weather AND some spare time, so I pulled the mast down, and took down my antenna to see if I could get to the bottom of this issue.
So on checking it all over, the connectors are all clean and dry, the antenna shows no signs of water ingress and appears well sealed, and is sound. I re-mounted the antenna further down on my mast on a cross-bar (to get it away from the mast), and used a shorter cable. When I put it back online on Saturday, it showed a very small improvement compared to previously (keeping in mind it’s lost about 8 feet of height).
However on Sunday, it was like a new setup, performing much closer to my performance graphs above before the issue started in October last year (especially taking into account the fact the antenna is lower down than it was before).
Here’s the perf graphs from the last 7 days (the data gap on Saturday is when I was doing the physical work).
You can see how on the Sunday after i did the work (on Saturday,) suddenly my “Messages > 3dBFS” is back to where it was before the initial fault and the number of aircraft I’m tracking is way up. As is the range and Signal level (with the spread between 1st to 3rd quartile closer to where it was before the issue).
So the only thing I’ve changed is a) the cable and b) the location. I’ve dismantled the original cable, and can find no fault with it (both cables are from the same lot of coax). Anyway, at least things are working better now and I’m going to leave the antenna in this lower position for now, as it’s much easier to get to if it goes bad again. I just wish I could have found a more conclusive answer though! Very odd indeed!
So I am assuming you installed your own connectors on the same lot of coax that you originally had up. Do you have a antenna analyzer to check the completed cable? As you may already know, a single strand of ground braid can cause serious issues with the impedance match of your antenna system.
I’m new to this FlightAware software so I am wondering how you get access to those ADS-B Signal Level graphics? Please and thank you.
Did you check the cable for impedance? Could be that one of the connectors was failing ? I don’t suspect the cable as such but the actual connection points.
Did you have soldered connections, screwconnections ? Just thinking along…
The graphics are a package called graphs1090. The author is wiedehopf, you can find it on GitHub. It’s a valuable tool for analyzing performance of your setup and troubleshooting
Thanks Craig, unfortunately I don’t possess or have access to an antenna analyser, but I’m sure that may have revealed something.
You are quite correct, I made both cables myself (cut from the same length of RG-213) and both cables checked out perfectly from a DC perspective (all I can check with the limited test gear I have). Without more comprehensive test gear, I suspect it will be difficult to find answers without a lot of trial and error!
Thanks for the thoughts Tom - you’ve touched on similar ideas to Craig. Unfortunately I don’t have test equipment for measuring impedance, but the cables did check out from a DC perspective (shield/core not shorted, good end to end conductivity, etc). I fitted the connectors myself, with the centre pin soldered (not crimped!).
I disassembled the N-connector of the original (suspect) cable when I pulled the setup down, to check for any signs of water ingress, etc, and it all looked mint.
I haven’t ruled out the possibility the antenna itself is to blame as it may have an internal mechanical fault and the movement of pulling it down and putting it back up may have rejigged it’s internals.
What’s really weird is that last winter here, the temp dropped a few times to <-5.5ºC, at which point (almost exactly,) my gain returned, just like the original good times. It was like whatever was causing the fault, went away at extreme low temps. As soon as the outside temperature went above -5.5ºC, the gain dropped back down again just like it did in my original graphs.
Temperature changes influencing reception, hmmm, that smells like a fracture of some print trace line or a dodgy connection indeed. Increasing temperature causes expansion of metals so that might be a possible cause. Low temperature has the opposite effect so that would enable the connection again. If it is a complete segment failing that might explain the drop in range and Rssi.
Depending on the antenna, isn’t there a way to disassemble it ?
The Pirimoni antenna should be accessible via the bottom N connector if you are willing to investigate.
You just have to seal it again in order to get is waterproof again.
But that’s up to you to decide.
:
). These are the units I’m using from TP-Link: 
