Problem Setting Up Airspy, Raspberry, Dump1090-fa

Hello All,

I just rebuilt my ADS-B rig. I had all of this working a few years ago but can’t quite get it this time. I am able to see a few airplanes but nearly as many as I used to see. I know Covid has had an impact on the numbers but I am only getting data from planes I can actually visually see. Range seems very limited.

My antenna, FlightAware standard issue, is located in a better location than when I last set it up. The only clue I have is that the Airpsy I was using for receiving ADS-B can’t be found. So I am using one that I was using for decoding a Russian Weather satellite feed. I have a memory of accessing the Airspy directly and adjusting the frequency it listened too. But that was three years ago so who knows what I remember about what? But I do wonder, how does Airspy device know what frequency I want to listen to? There must be a setting somewhere, so I figure.

Wrapping it up. Why can I track only close objects? What have I missed?

Raspberry Pi 4 running Bullseye. Airspy purchased in 2018 or maybe 2019. A booster that requires power. I have enabled the bias t. FA antenna. Six foot, premium cable, (spent some money on that), between antenna and Airspy. Raspberry is powered by POE so it is close to the antenna.

I have played with the gain settings. No luck. Any ideas?

John Ullom

Have a look at the stats of nearby stations to see how you compare - that will confirm (or not) that you have a receiver problem.

Any more info on the WX receiver? Make sure it doesn’t have any filtering.

Check your connectors to make sure a RP-SMA hasn’t snuck in somewhere.

What are you using for the Bias-T? Is it really working?

Has the antenna and/or feeder been out in the weather since it was working last?

The Airspy is using a wide frequency while ADS-B is on 1090 MHz only. As far as i understand the Airspy is almost “blind” if no LNA with a filter is in use.

What is your setup from Raspberry to the Antenna?
Should be

  • Raspberry
  • Airspy
  • filtered LNA
  • Antenna

(I did not mention cabling which also can have a high impact)
And the LNA requires also power, either via Bias-T or coming from an external source depending on the LNA used.

But let’s see when @wiedehopf is replying, an expert in Airspy setup.

There is neither enough info about the setup (which LNA) nor enough info about what’s going on (no graphs).
Nor is the airspy_adsb config given.

Then you need to check for terrain as usual: https://github.com/wiedehopf/tar1090#heywhatsthatcom-range-outline

Without the highlight i would have probably waited until there is more usable info in this thread.

The decoder that uses the airspy is an ADS-B decoder. ADS-B is always on 1090 MHz.
You could just as well ask how the decoder knows what ADS-B messages are :confused:

2 Likes

@ All My apologies for the poorly asked question and much appreciation for helping me clarify.

@ geckoVN
Thanks for input. I am going to remove the LNA and see if that is the issue. I doubt I ever needed it but as noted in my question, it worked before.

@ fox hunter
The question: * Raspberry – As noted in my question, pi 4, running Bullseye

The question: * Airspy – As noted in my question, bought in 2018 or 2019. I forgot to mention that is the Mini version

The question: * filtered LNA – Don’t know. But if you read my original question you would see that is the same unit that I had working in the past. Maybe it is broken. Not likely. I’ll take it out of the equation.

The question: * Antenna – As noted in my question, Standard Issue Flight Aware.

The question: *Cabling – As noted in my question, Six foot, premium, professionally made, worked in my original setup. Spent 50 bucks on it. The cable is not the problem

The question: *Bias-t – As noted in my question, enabled. Yes the LNA requires it. And as mentioned, it worked before.

@wiedehopf.

Your observation: There is not enough info about the setup – Set up is summarized at the end of my original question. Pi 4 / Bullseye / post 2017 Airspy / LNA / bias t enabled / FA antenna / Six foot, premium cable between antenna and Airspy / Pi powered by POE.

I am unsure of the model of the LNA, I enclosed it in some pvc for weather hardening purposes. I’ll get it out, get model, then remove from the setup so as to reduce complexity. The Airspy I am using is the mini version and was purchased in 2018 or 2019. I am using the default config file with only two changes. Added -b to the parameters and played with the gain setting. Nothing else.

Your wrote: – You could just as well ask how the decoder knows what ADS-B messages are

I could ask that but that but that would be a dumb question. I get how Dump1090 works. My question, poorly asked, was:

“The only clue I have is that the Airpsy I was using for receiving ADS-B can’t be found. So I am using one that I was using for decoding a Russian Weather satellite feed. I have a memory of accessing the Airspy directly and adjusting the frequency it listened too.”

Now maybe my recollection is wrong. Not hard to believe. But what I was asking was if I needed to change anything about the Airspy itself, so as to make it work with Dump1090. I bought three of them, from you I think, at the same time. One was used for Air Traffic Chatter, One for the Russian weather satellite. The third was for ADS-B. I had all three setups working in the past. I lost the Airspy Mini that was part of the ADS-B setup. So, I repurposed one of the other two.

A better way to have asked would have been:

How does the Airspy Mini device know which Frequency it is supposed to listen too? Does Piaware control the setting of the Airspy? Are there any settings? If it was the standard RTL dongle you guys sell, I know it would be configured for 1090MHz. But the Airspy Mini, per my understanding, is a far more flexible device. I am not on expert on RF tech. Don’t want to be. But my intuition tells me that something needs to inform the Airspy as to which frequency it should tune in to. Unfortunately, my intuition and a quarter will only get me a gumball.

Finally, let me stress, all of the equipment I am using, I successfully used before. Other than mounting the antenna in higher location, it is only three feet away from where I had it working before. I could send graphs and config files but I don’t get how that will help in the context of my question.

Thanks for all the help!

John Ullom

Piaware knows nothing about the airspy. airspy_adsb does all the configuration of the device.

Arguably, less flexible. But supports a higher sampling rate and has a better ADC.

Without a quality filtered LNA like the rtl-sdr triple filtered or uputronics ADS-B LNA the airspy performance won’t hold up compared to the specialized rtl-sdr version by flightaware which have an integrated LNA / filter.

So you installed this? https://github.com/wiedehopf/airspy-conf#airspy-conf

airspy_adsb is talking to the airspy so … dump1090 has nothing to do with interfacing with the hardware.
dump1090 just gets the raw messages with timestamps and runs network only.

dump1090 most commonly uses rtl-sdr devices which it tunes to 1090 MHz.
Let’s just say this: any ADS-B decoder with support for certain hardware would tune that hardware to 1090 MHz. If it wouldn’t it would be a defective decoder.
Anyhow … if it wasn’t tuned to 1090 MHz, you wouldn’t receive anything so … yeah i really don’t get the question. Regardless this should answer it even if i don’t get it?

Well the graphs showing the previous and current performance / signal strengths can indicate certain issues by just looking at them.
Also the range graph / message rate graph is just much more objective than people saying “performance is worse”.
Signal graph can show some obvious issues in some cases.

You could always try an rtl-sdr with biastee enabled to compare to the airspy.
(if you have an rtl-sdr model that has biastee hardware support, many don’t)
“same hardware” is mostly the antenna then.
Anyhow try the old hardware, see if you can reproduce the good reception.

The airspy_adsb configuration file is relevant but if it’s mostly default that’s fine.
Though still good to see because i don’t know which default you used … there is old info in some place.

1 Like

Thank you obj. I had it working before. I used the setup scripts provided at git. No cookie. Might try rebuilding the Pi. I’ll keep hacking at it.

Thank you for helping me understand.

1: I have to do some work to confirm it but assuming it was available in 2019, I am positive I bought the uputronics ADS-B LNA: – 1090MHz ADS-B Filtered Preamp I am suspicious because that is the one item that might cause everything to sort of work. I will remove from the setup and retest. If I choose to replace, which would you go for?

2: Yes. I used those scripts at the link you provided.,

3: I only have graphs and such from two days of not much working. Lost the original install a year ago when I used the Pi for an emergency project.

4: My config file is stored here: /etc/default/airspy_adsb and here are the contents sans comments:
GAIN= auto
OPTIONS= -v -t 90 -f 1 -w 5 -P 8 -C 60 -E 20 -b
SAMPLE_RATE= 12
NET= -l 47787:beast -c 127.0.0.1:30004:beast
STATS= -S /run/airspy_adsb/stats.json
#don’t change:
G=-g
M=-m
AFFINITY=“0-7”

5: “Regardless this should answer it even if i don’t get it?” Yep. Question answered.

Maybe.
What dongle are you running now?
The Airspy Mini has a bias-T built in. Most dongles do not, hence my question" “Is it really working?”

The uputronics is fine really especially if you want to go with the builtin biastee as the rtl-sdr LNA needs more current.

Before you do anything further, do the easy thing: https://github.com/wiedehopf/tar1090#heywhatsthatcom-range-outline

This will give you an idea of what range you can get and which range you are currently getting.

Also you can provide a screenshot and we have some objective measure of how bad it is.

The config just posted has -b for the bias tee on.
LNA without power is such a hefty road block … you’d see it in signal strength.

One more reason to see some graphs / more data. (https://github.com/wiedehopf/graphs1090#graphs1090)

… but does the dongle have a bias-T to be turned on?
The ‘-b’ is in the config for the Airspy Mini, which the OP can’t find, so is using a substitute dongle of unknown provenance.

The OPs site … looks like it has a “hefty road block” to me


One last point: I’m more of a ‘hardware guy’ than a ‘software guy’, so I consider “5V on a multimeter” a lot more convincing the a “-b” in a config file.

https://flightaware.com/adsb/stats/user/skygizmo

Sounded to me like it was “not the same airspy” he used for the old station, but rather a different airspy.

I suppose it’s possible that either the biastee isn’t working or the LNA is just dead for some reason.
Really hard to tell.
Could also be some connection issue, one of the connectors on the LNA could have broken off the PCB (rare but not super rare for the uputronics LNAs).
Or the circuit could be fried.

If the range is super limited, eliminating the LNA from the signal chain is a good first step indeed.
Issue so fare was that no numbers were quoted so it’s really hard to judge without any numbers.
I suppose “only planes i can visually see” is very limited but i didn’t see that initially.

5 nmi max range would have been a much more useful statement that’s easier to decipher when quickly reading something.

1 Like

Ah - ok, that would make sense.
If that’s the case, then yes, it should all be working as it was.

Good news! It was the LNA. I should have taken it out of the chain before posting. I opened up the enclosure I had it encased in. Corrosion. So much for my weather proofing.

Might order a new LNA. Not sure if I need one but they don’t cost much. I’ll leave this topic open incase anybody has some thoughts. I’ll close it in a day or two.

I should have reduced the complexity before resorting to asking for help. I’ll post some data in a few hours in hoes somebody can tell me if they thing I can do better.

Great forum. Thanks to all!

3 Likes

Gee - I wonder what time you pulled the amp!!
image

A toothbrush, some alcohol and you may be able to bring it back from the dead.

Yep. Kind of feel dumb for not checking before posting.

As I am not much of signals guy, what benefit would I derive from having an LNA? My goal is to track local traffic right down to the runway. I seem to be getting a good signal. Is there any reason to try for better?

Thanks for hanging in there with me to all.

If you’re fine and don’t mind the reduced range … no reason to have an LNA.

LNA might also not change if you get taxiing aircraft as usually you get it or you don’t, nothing in between (weak signal).

Your stats are now looking quit impressive.
Looks like you has a terrain problem to your east, but you are now the highest performing site in your area, not to mention 300 positions at 400km+.

I suspect you’d start loosing strong (local) signals if you put an LNA back in service.
How high is your antenna? You may get stronger ground signals with more height (height is always good).

Bunch of hills half a mile to my east. I can see stuff that is high up out over the central valley. Hyper local is what I am interested. So no LNA for now. I do get a kick out of seeing aircraft pop up 250 miles out over the Pacific.

As far as height goes, I am pushing it as it is. I have tolerant neighbors. Don’t care to find out what their limit is, I know I am breaking a few codes!

Thanks for help and advice.