Hi, I recently set up a PiAware instance, but am having issues getting results I think I should be getting. I am using the FlightAware Plus (blue dongle).
First, the history. While configuring and waiting to install the antenna on the roof, I had the antenna direct connected to the SDR, which was direct connected to the Pi in a south-facing window. I routinely got hits in the 200-250 mile range to the south (not much in the north, obviously). I then moved the antenna to the roof, still direct connected to the SDR. I ran a USB cable from the roof to the Pi.
I was again getting 200+ mi hits, but only to the north- I realized that the very apex of the roof was blocking my line of sight to the south. So I moved it again:
(under a basket to keep it out of the direct sun, but the antenna has full 360 line of sight).
Once I moved it, I encountered the issues. My ADS-B and MLAT positions dropped significantly. May 8th was the day the antenna was first moved to the roof, and May 24th it was repositioned, you can see the dropoff here:
You can see the steep transition from ADS-B and MLAT to āotherā signal types.
For fun, I ran a spectrum scan, I donāt know if that will help:
Does anything here jump out as being a cause for the degradation of ADS-B signal processing? I have the Plus, so there should be a filter helping things alongā¦ any thoughts on improvements? Thanks
Odd, are there any warnings on the stats page? what does skyview show?
My first idea would be that the messages are clipped and therefore not decoded and counted as other. David Baker explained in a post here what constitutes other, but I keep forgetting it and cannot find it right now.
With regard to heat management, I think a black basket might be counterproductive, a white might be a better choice.
Also, I am not sure how long the FA dongle or a survive outside with minimal protection. Corrosion is the enemy of the PCB.
You can see the high number of āotherā hits, and even on ADS-B you can see that not all the ADS-B message have positions attached with them (note 4 aircraft seen with ADS-B, but only positions for 2).
My only thought on the matter (and why I did the spectrum scan) was that maybe there is just so much other junk in the local spectrum that it is overriding the ADS-B traffic. Maybe the blockages of the house when it was in the window, and the roof apex before it was moved was enough to block some of those other signals? The FA dongle has a filter, but Iām wondering if somehow that isnāt enough? https://www.amazon.com/gp/product/B01M7REJJW/ref=oh_aui_detailpage_o08_s01?ie=UTF8&psc=1 is my purchase, and Amazon.com is the antenna Iām using.
Would an additional filter help? Turning the gain down to almost nothing? Iām hesitant to buy more stuff; this ācheapā hobby has started to balloon with expenses
And good comment on the basket- the sides have holes for cross ventilation, and Iāve weatherproofed the SDR pretty well, so most moisture should be kept out. Also, its hanging suspended inside the basket, so that its not directly touching the hot surface.
ā¦ itās black so that it blends with the roof because Iām hiding it from the HOA ā¦ you can apparently have a satellite dish, but not a 6" antenna.
I would say it can possibly be your antenna set up. try the same experiment from your desk or where ever you work from and if similar results occur then change the antenna. I assume that your DUMP1090 settings have not been messed with since your install?
Judging by the screen shot I am approximately 120 miles north of you and I know that you should be easily receiving a good amount of east-west and north and south traffic.
I went back and re-read the whole post. Judging by a few factors, mainly the HOA. My condolences on that. Now judging by the rooftop picture, Aparently you have a tall attic. My suggestion is get this;
or get this; https://www.amazon.com/gp/product/B00WZL6WPO/ref=oh_aui_detailpage_o03_s00?ie=UTF8&psc=1
and install it hanging from the highest beam in your attic.
I have the same small ADSB antenna you have pictured and it has been sitting in a drawer for over a year.
Yes the external antennas I had pictured go from 50 USD to 150 USD but that option will be far suited than what you have.
As for costā¦it took me a while nickel and dimeing my set up because the hobby is my last priority.
If Iām reading it right you now have the antenna connected to the Pro Stick Plus directly and both mounted on the roof, and then a long USB cable connecting the Pro Stick Plus to the RPi which is inside the house?
Could it be that by moving it you have reached or exceeded the maximum length for the USB cable which would preserve quality signals? Or perhaps it is now closer to something electrical which could interfere with those signals on the cable? USB cable runs should be limited to a maximum of 5m.
This would result in a sudden drop in data quality for dump1090 which appears to be consistent with your symptoms. If you think it might be this, for diagnostic purposes try temporarily moving the RPi closer to the roof to bring the cable length back down to what it was when you had good signals, and use a mains extension to get power closer to the RPi, then see if the problem goes away again.
Also make sure youāre using a quality 2.5A PSU for the RPi such as the official branded one.
Thank you all for your great suggestions. I think Iāve found the solution. Coincidentally @abcd567 , thanks to your post at Tweaks - #17 by abcd567 I ran the optimize-gain.py script, and with a small tweak (added some additional low-range gain values) I found something interesting- there were two ideal gain settings, one high (42) and one low (15), giving me a camel hump graph with two peaks, versus a normal bell distribution. Using the higher value, I can see aircraft with RSSI values around 29 or so. The lower setting lets me see aircraft with RSSI around 3 or so. It looks like I canāt see both sets of aircraft at the same time.
But back to the good news- setting the gain to 42 (it was 25 before), I can see a large amount of aircraft again! Iām only halfway through the day (and not even to the busy part) and I already have 4x the number of positions I did previously.
@glennblum That is an impressive screeshot. Which one of those antennas (if any) are you using? What gain value do you have set? Do you think heat shielding on the roof will offest the increased antenna gain when compared to the smaller outside antenna?
@chrislfa Yes, excellent points, and I (think I) have overcome those already. You described the situation accurately. I have a 6" USB cable from the Pi to a powered USB hub, and then have a 10m active USB cable plugged from the hub to the SDR on the roof. (30m is the max you can go with an active cable, requiring a power source like a hub).
And yes, my first power source was only 1.2A, I found out that issue the hard way already
So my last two questions:
I still only get ~40msgs/sec ā¦ I was VERY surprised by @glennblumās 250/sec metric there. Maybe thatās just because he has a bigger antenna? Or should I really be seeing more than 40/sec? Hereās my current map:
Is there anyway to catch both the small RSSI and large RSSI aircraft with one dongle? Or would I need a second dongle tuned to the lower gain value?
Thanks again to all - this has been a big help to me, and I hope my documenting my setup as much as possible here, it can help others as well. I am VERY open to additional suggestions in tuning the software, or trying new things.
Itās great that you sussed out the problem. This thread remains here and future visitors will likely fiind it as they search for similar problems.
Have you tried setting gain to -10 (minus 10)? That activates automatic gain which usually ramps up gain higher than the 49.6 you can set manually. You may find that despite the nearby unwanted signals the receiver is capable of cutting through that and grabbing the most in-band signals at the highest gain. If thatās too much, try a manual gain of 49.6 and see how that goes.
Iām getting 400-1000 messages per second with an external antenna indoors and Iāve set mine manually to 49.6 which has given me the best performance for number of aircraft and range, with positions actually down a bit in some places (which I interpret as the positions it can see being of higher quality)
Once you know how the max gains are performing you have a handy baseline for tweaking further if needed.
Since I received an RPi3+ this week, it made the other RPi3 (non +) available to be re-purposed. I installed it next to the main station, in the garage. This one is a poor manās ADS-B station. A homebrew QuickSpider antenna, an $8 generic dongle, no pre-amp, no filter.
Itās interesting to see the performance differences. While the total number of planes seen is lower on the poor manās station, itās not by much. The more interesting, is that there are times when it sees twice as many planes as the main stationā¦
Here is how I have my feeder set up.
Roof Mounted antenna tip-pod and my antenna is up about 18 feet above my ground level. Fed with RG6 cable, 500 feet worth with proper grounding and protection into my Pro stick plus PiAware. Gain value set to AGC. Because the antenna and the pro stick plus filter is designed for approx 1090 mhz so I let the AGC to be set.
Why the RG 6 cable? Yes the 75 ohm miss match is acceptable to me since I have a of the cable and plenty of type āfā connectors. My non-feeder is run by my computer with a RTL-SDR.com V3 dongle, LMA wide band amp, Flightaware 1090 filter, same cable and home made cantenna. The performance is not as good as my feeder but good enough for me.
I say tweaking the gain settings is a must to find the sweet spot.
I would suggest to replace/ upgrade your antenna.
