[Solved] Cycling up and down and up and down and up and down

I’m building a new feeder with a Pi 3B+, Airspy, bias-tee injector and an RTL-SDR LNA.

It’s on my bench at the moment so I don’t expect to see a huge amount but I expect it to be constant.

Instead, I’m seeing around 50-60 aircraft for a few minutes and then it drops down to fifteen or so where it stays like it for a few minutes and then goes back up again.

Has anyone experienced this sort of thing before? I’m loath to spend the time mounting this at the top of the mast if it’s going to carry on doing it. I’ve never seen this before.

aircraft

Can you check

sudo journalctl --no-pager -e airspy_adsb

Also the temperature and cpu usage graphs, maybe memory graph.

That is really odd. Nothing like what I have seen over the past few years.

I suspect that something is jamming your reception for a few minutes at a time.
Does your range reduce during the periods where your aircraft numbers are low?

If could be that the airspy app is also losing connection for a short time, then restarting.

Yes, the range drops right down. My other receiver in the loft is perfectly OK though, no sign of any local interference on that one.


ads-b%20cpu
temperature
memory

If I run that command, I get an error…
Failed to add match 'airspy_adsb': Invalid argument

This looks better sorry.

May 14 18:15:59 new_external_tracker systemd[1]: Started Airspy ADS-B receiver.
May 14 18:15:59 new_external_tracker airspy_adsb[334]: Listening for asavr clients on port 47806
May 14 18:15:59 new_external_tracker airspy_adsb[334]: Listening for beast clients on port 47787
May 14 18:15:59 new_external_tracker airspy_adsb[334]: Acquired Airspy device with serial 26A464DC28751693
May 14 18:15:59 new_external_tracker airspy_adsb[334]: Decoding started at 12 MSPS
May 14 18:16:11 new_external_tracker airspy_adsb[334]: Push client connected to localhost:30004 (beast)
May 14 18:16:11 new_external_tracker airspy_adsb[334]: Push client disconnected from localhost:30004 (beast)
May 14 18:16:24 new_external_tracker airspy_adsb[334]: Push client connected to localhost:30004 (beast)

It’s been through a couple of up and down cycles since reboot and that output hasn’t changed.

I’d recommend to test in another room.
Some signal source interfering seems like the most likely explanation.

No clue what would be so consistent though, a mystery to me.

Maybe just plug the airspy into windows and look at the waterfall around 1090 MHz?

This is strange.
It shouldn’t disconnect at all unless you restart dump1090-fa.
Looked normal at first glance but now i’m curious.

sudo journalctl -f -u airspy_adsb

Can you leave that open and see if anything happens during a cycle?

After connecting the push client there should be no messages.

Well roger me sideways. I’ve moved it to the back bedroom and it’s fine. There must be some very localised QRM in my shack but the only thing I can think of in here is the UniFi access point and why it would be in those cycles, I have no idea.

There is plenty of equipment in here but nothing I can think of apart from the AP which is at microwave level but even that’s the wrong frequency to interfere with 1090MHz.

3 Likes

… just for completeness, here’s the stats now after a couple of hours where the difference is very clear.

resolved

2 Likes

Is the coax well shielded? Is the RPI in a metal case?

There’s six inches of coax between the aerial the the LNA then the bias-tee is directly connected to the LNA and then there’s another six inches of coax tween the bias-tee and the Airspy.

The Pi isn’t shielded, it’s all mounted together in a plastic box. It’s the same box I used previously outside.

Whatever it is is very localised. The receiver in the loft is fine and the one I had mounted outside was perfectly fine.

1 Like

Doesn’t really matter when the antenna is indoors in a room full of electronics.

You could probably open a waterfall at 1090 MHz and move the antenna to find the source.
But i suspect you’re not that curious :slight_smile: (Not sure if i would be either)

That periodic up and down is very mysterious though.

Looks like something is happening every 10 minutes with an offset of 5 minutes from the full hour.

That’s pretty much how I clocked it.

You’re right, I don’t really care where it’s coming from. I know it’s there and that’s good enough for me, especially as the Airspy is well fixed in the box and would be a pain to remove. Also, I don’t have a Windows computer I could easily use as the one I have here with VRS and PRTG is buried in the back of a cupboard, well out of the way.

1 Like