That still doesn’t add up. Cellphones (4G) work on 800Mhz and 1900Mhz. Even with a radio that crosstalks a lot that is still quite clear of the 1090Mhz range.
Well apart from logic and theory, this is what practically happened with me and so many others:
“Without external filter, the FA26+ blue prostick plus performed poor, and performance jumped up dramatically when external filter was added.”
1 Like
I’ll buy a filter and test.
Also are you running MRTG on the Pi? if so What MIBS are you collecting to get the data plotted out?
I have installed “Web Portal”. The “web portal” installs and uses rrdtools/collectd to create graphs.
The instructions to install web portal are here:
before buying additional filters - make a scan with rtl_power to see what’s in the air at your site: Filter not producing desired result - #11 by obj in addition you can play a little with gain setting in dump or piaware to see what gain matches best when flightaware antenna is connected. there is a script here in the forum that sets different gain values - but to be honest i’m more a fan of doing it manually that gave at least me faster and more precise results.
in my one to one tests with about 10 different antennas the flightaware (you have) was runner-up with decent performance!
cheers
tom
p.s. this is how my plots looked like
1 Like
So because of other circumstances I find myself unable to sleep as much as I’d like. So when I woke up after a couple hours of sleep this morning there was something bothering me. I did the math and a 1/2 wave length antenna for the 1090MHz frequency should be something like 137.5mm (speed of light divided by frequency divided in half). The FA antenna is 26 inches or 660.4mm not even close to the proper length for a 1/2 wave antenna. The math says that is a 4.8 wave length antenna and that is rather odd as an antenna length. so either the antenna element in the internals is not tuned to 1090MHz or there is a lot of empty space inside that shell. Right now I’m not willing to waste my FA antenna by cutting it apart to find out. Either way, I’m going to build a full wavelength antenna later on and compare it to the FA antenna and my 1/4 wavelength antenna.
@TomMuc - I’m going to work on generating those charts once I read up on the cli for rtl_power.
@obj So what length is it then? Since you are FA staff and have the element out I imagine you have the specs on it.
@obj can you also weigh in on the difference in performance between the antennas?
if you goal is the direct way to max the radio reception of your site - then what you are now thinking about is the most unlikely part of the story. do a noise scan with rtl_power → play with gain of dump1090 → check the antenna-run between fa-antenna and dongle → add a decent filter that matches the noise at your site.
My aim is always the highest performance from the systems I have.
@TomMuc - Read the docs on rtl_power, attempted to run it with this:
rtl_power -f -i 10 -e 1h 600M:1400M 2017-10-07_600-1400_001.csv
Also tried:
sudo rtl_power -f -i 10 -e 1h 600M:1400M 2017-10-07_600-1400_001.csv
I’m getting a Segmentation fault without the sudo and nothing at all with it. The great Oracle (Google) says that it is an issue with libusb in the raspbian however the version they site as resolving it is older than the version I am running.
Did you have any issues running this on your Pi?
jessie or stretch? fa-sd-card or raspbian image and all installed on your own? exact errormsg?
the setting i mostly use:
./src/rtl_power -f 800M:1400M:300K -i 10 -c 50% -g 50 -e 20m -F 9 >scan.csv
fa-sd-card image running jessie with apt-get update & upgrade run on it already. The latest version of libusb1.0 is already running. The exact error is:
Segmentation fault
There is nothing else there and rtl_power does not have a debug option. There is nothing in /var/log/syslog about an error with rtl_power
Running the command example above yields the following:
pi@piaware2:~$ sudo rtl_power -f 800M:1400M:300K -i 10 -c 50% -g 50 -e 20m -F 9 >scan.csv
Number of frequency hops: 429
Dongle bandwidth: 2797202Hz
Downsampling by: 1x
Cropping by: 50.00%
Total FFT bins: 6864
Logged FFT bins: 3432
FFT bin size: 174825.12Hz
Buffer size: 16384 bytes (2.93ms)
Reporting every 10 seconds
Found 1 device(s):
0: Realtek, RTL2832U, SN: 00001000
Using device 0: Generic RTL2832U
usb_claim_interface error -6
Failed to open rtlsdr device #0.
pi@piaware2:~$
sudo killall dump1090-fa
I wondered about that.
While I run this, what software did you use to map out the csv to jpg?
wget https://raw.githubusercontent.com/keenerd/rtl-sdr-misc/master/heatmap/heatmap.py
chmod +x heatmap.py
./heatmap.py scan.csv scan.png
