Hi all, like the title says, I finally got a good N-male to F-female adaptor, and I swapped out my Quick Spider with the Flight Aware 1/2 wave antenna. To my surprise, I’m getting more range somewhat (100nm to 125nm max) but markedly fewer planes (29 vs 51) and messages/sec (151 vs 250+!) . Is this expected behavior for the difference between the two designs? Meaning, does the FA antenna require more stuff like a filter to perform as well as an unfiltered Quick Spider?
Setup is:
Both antennae side by side in attic, ~12ft AGL.
FA-Ant > N to F > Regular RG6 coax (approx 5 feet) > F to MCX pigtail adaptor > NooElec SDR > RPi
Quick Spider > Regular RG6 coax (approx 4 feet) > same F to MCX pigtail adaptor > NooElec SDR.
All settings on Dump1090-fa and the SDR drivers are the same.
I do plan to move the FA antenna outside where it belongs, later in this hobby, but I wanted at least parity in performance under the same conditions before I do so.
The short answer is that you may be encountering overload of your SDR. The higher gain of the FA antenna may be enough to cause it, as if connected correctly, it has more gain than the spider. You may have to experiment with gain setting in your Pi.
No doubt someone will be along with a longer response soon.
Further to belzybob’s reply it’s definitely worth exploring that. The gainmap script I wrote might help visualise in-band noise and nearby noise at different gains. You’ll need to use the gain values supported by your SDR, or just use integers and let it auto adjust to the nearest supported values. I had something similar; I added a Uptronics filtered preamp which helped pull in signals from much further out but I had to lower the gain a notch to quieten down in-band noise in order to ‘hear’ those signals properly.
Either use gainmap script as proposed by @chrislfa in his above post.
OR
Try these gain settings manually: 50, 45, 40, 35, 30, 25, 20
.
After finding best gain setting, test for presence/intensity of interfering Cell/Mobile signals. These are also a reason for poor performance, and performance can be substantially improved by adding a filter.
The method to find Cell/Mobile signals in your area is given in the this thread: Do I Need A Filter?
Thanks guys. I’m stil wrapping my head around all of this. It seems that the Freq space between 850Mhz & 885MHz is extremely “strong” for lack of a better word, compared to the little blip at 1090MHz. What I don’t get is why this even matters? Who cares about those freqs when I only want to decode messages around 1090Mhz? My guess is that we’re looking at the SDR’s gain level because the 1090Mhz messages are now weaker, because perhaps the AGC is taking into account other freqs like 850Mhz?
So the only thing that makes sense to me is that the Quick Spider is actually an /excellent/ design, because it naturally rejects (or captures less of) freqs outside the designed range. (In essence a natural filter) Hence, the SDR dongle’s AGC will work better because it’s not ‘hearing’ the super loud 850-890Mhz messages. When I connected the FA Antenna, 850-890 got a lot “louder” because of the design of the antenna. This made my SDR downshift its gain (because of AGC), making the 1090Mhz messages “quieter” and therefore harder to discern. The Longer dipole length is capturing weaker 1090mhz mesages from father away, but they’re not amplified enough. A band-pass filter would ‘fool’ the AGC enough to focus on the band I want (1090) to work much better, or I can literally go back to a piece of coax that I mangled (Quick Spider) and get better performance, because I don’t want to learn about yet another connector (SMA) for the stupid filter, and buy expensive, inferior coax to connect to the SMA band pass filter.
Am I close? Far off? Just frustrated with buying adaptors on Amazon?
Even though a spider antenna is tuned to 1090mhz, it’s still going to see the cell phone stuff at 850-890mhz and fool the AGC. If you live in a major metro area, even an untuned coat hanger is going to see more cell tower energy that what you’re looking for on 1090.
I live in a place where I see lots of junk at 910-980mhz. A filter is worth every penny I paid for it.
I’m with you, and that makes sense, SMburn, but what doesn’t make sense to me is why a filter is needed at all for the FA antenna (essentially a 1/2 wave dipole) and NOT needed for the Quick Spider, which I already admitted in another thread that I built poorly. There must be some physics/software explanation. And I guess one of the biggest questions I have is how to tell what AGC is currently using, and if I crank it up, will the SDR/Dump1090 completely ignore the “uninteresting” freqs, even if they’re very ‘loud’
In spite of how any antenna is tuned, it’s far from perfect. It might be spot on at 1/4 wavelength, or whatever. “Resonate” is a relative term, re: my coat hanger comment. But a simple 1/4 or 1/2 wave vertical is also going to resonate at oddball 5/8ths, 3/8ths, etc on some other frequency.
There’s a reason why your cell phone works even when it’s in your pocket, even with a tiny antenna. It’s being blasted with RF at a much higher level than anything on 1090.
In the case of an FA antenna vs. a spider antenna, there’s simply more metal up in the air with an FA antenna to get blasted with cell tower stuff, which may very well be fooling your AGC.
@Cetrian
I wont discuss “why”, but it is proved by practical experience of a very large number of users that with a filter+proper gain setting, the FA antenna outperfoms the 1/4 wavelength antennas such as Spider and Cantenna.
Off Designed-Frequency Behavior of 2 Types of Antennas
Frequency Sweep (From 100 MHz to 2500 MHz) of a High Gain Wire Collinear Antenna and Low Gain 1/4 Wavelength Ground Plane Antenna (Quick Spider).
The frequency sweep results of a high gain wire collinear antenna show that it not only performs best at its designed frequency, but also performs fairly good at or near sub-multiples and multiple frequencies.
For single frequency reception, such as ADSB, this characteristic of collinear has disadvantage that the RF signals picked from all these multiple & sub-multiple frequencies necessitates a tuner circuit or a band-pass filter at front end (i.e. between antenna and receiver).
The frequency sweep results of a 1/4 wavelength ground plane antenna show that it gives good performance at and near its designed frequency only, and at other far away frequencies its performance is poor. Hence the 1/4 wavelength ground plane antenna picks very little harmonic and sub-harmonic frequency signals.
SWEEP 1 of 2:
HIGH GAIN WIRE COLLINEAR ANTENNA WITH COIL & DECOUPLING SLEEVE
Wow, abcd567, you never fail to impress with the detail of your posts!
So, the short answer is that yes, the physics of the way the antennas resonate and reflect different frequencies necessitates a band-pass filter on the FA antenna, mainly because of the way the SDR handles gain and dynamic range. Not unlike how a larger lens on a camera might let in more light of all frequencies, but reduce the detail on a certain color unless other mitigating factors (aperture, film speed, etc.) are used. The only way this would not be true for the FA antenna is if there were no other sources of RF other than the ADS-B signals in your area.
Now I’m on to find a band pass filter with F-connectors
