I’m a newbie quickly getting hooked on tracking planes with ADS-B. I’d appreciate any suggestions to put me on the right path to optimize my setup. I’m currently feeding FlightAware and ADSBexhange.
My setup:
ADSBexchange 1090/978 N-Type antenna
1ft RG8 Coax cable
BingFu SMA to dual SMA splitter
FlightAware Pro Stick (Orange) for 1090 Mhz
ASDBexhange Orange R860 dongle for 978 Mhz
My antenna and RPi3+ are currently installed in my attic. I eventually want to install it outside on a mast that will be extend a few feet above my roof. Before I mount the antenna on a mast outside, I’d like to optimize my setup.
I’m thinking that I will mount the antenna outside and will continue to use the very short coax cable/SMA splitter to connect the antenna to the dongles and use longer active USB cables to connect the dongles to the RPi. I want to RPi to be mounted below the roofline so that it is easier for me to access and service it.
I’d appreciate your help on a few questions:
Would it be better to have two separate antennas going to the separate dongles or is ok to use the dual band antenna with the SMA splitter?
What data should I look at to determine whether I need to use filters and/or amplifiers for each of the frequencies?
Are there any other things I should be thinking about to optimize my setup?
Separate antennas for 1090 and 978 MHz would be better, but not hugely better. Probably the easiest thing to look at is “Nearby Sites” on your stats page to see what the best site is getting near you. (Although looking at your stats page, a “Nearby site” can be up to 100 miles away). Still might be useful though.
Thanks. In looking at the sites within 75 miles of me, I have the most flights, but not the most positions. I’m still trying to learn how all of this works. How could a site have fewer flights than me but more positions?
you could install graphs1090 in order to see the performance of the setup
Install it. let it run for 24 hrs and check if your setup is around 5% of strong messages in the top graph. If the percentage is higher, reduce gain to get closer to the 5% if it is lower then 5% you might increase the gain a little.
Here’s a discussion with some links in regard to gain:
Thanks. I’ve been playing around with adjusting the gain for the last couple of days. I seem to have reduced the range of my receiver by reducing the gain. A few hours ago, I set the gain back to the maximum (-10) and I installed Graphs1090. For the last couple of hours, it seems that my percentage of messages > -3dBFS is around 5%. I’ll check it after 24 hours.
My message rate seems low. I imagine the rate will increase when I move the antenna outside. But how can I tell if I need to add in filters or amplifiers to my setup?
The flights that the other site receives are visible for a longer period of time. Which could be due to having more range than you have. Could also be you are having drop outs at close range from having too strong of a signal. As tomvdhorst says, install graphs1090 in order to get more information on this.
Thanks. I’ll do some research. I might need help interpreting some of the graphs in Graphs1090.
Regarding the terrain around me, I live in a valley that is about 15 miles wide with mountains that are between 5,000 and 9,000 feet high, east and west of me. There is also a cellphone tower that is about 1 mile west of me. These factors are causing me to wonder about filters and amplification.
There is a small airport about 8 or 9 miles south of me. I’m interested in tracking the private jets that fly in and out of there. (There is also an international airport about 50 miles north of me. I’m interested in tracking those planes too) The planes from the small private airport don’t seem to show up in my feed as much as I had expected. I imagine they are broadcasting on 1090 Mhz. I recently added a dongle for 978 Mhz too. But, I’m only seeing 1 or 2 planes each day on the 978 Mhz frequency. I wonder if my gain is too high on 1090 and/or 978 Mhz to track the planes, using the small private airport, that a flying low and close?
In order to determine if you need a filter you might want to look at the weak signals in the ADSB signal level graph.
It seems your weakest signals are around -20 dB where you would normally have them around -30 dB.
Also your message rate is indeed low. I think the max gain setting is overpowering the dongles due to the airport nearby.
Anything within 3 NM should be considered as a strong signal.
So I would start with lowering the gain and then see what the message rate and the weak signal rate are reporting.
Next time post the whole graph including gain that makes the analysis easier.
Thank you. I stumbled upon the heywhtasthat.com website a couple of days ago. The area indicated by that website roughly matches the area I’m currently able to track airplanes in.
Do any of these graphs suggest that I need any filters, amplifiers or other adjustments to my setup?
Can you help me understand why the -24.5 dB level indicates that a filter is needed? How should I interpret the noise level in that graph? Does the noise level compared to the signal level indicate anything?
You normally would want the weak signal graph close or below the noise line. Since the noise line is below the weakest line and with little variation it indicates that you have a noise source in your area or nearby that prevents the signals getting to the -30 dB level.
I would recommend using 2 separate antennas vs running them through a splitter. Removing the splitter would get rid of the splitter signal loses and the added physical connections that add reflections and can be an entrance point for noise. Is your 1 ft piece of coax the proper 50 ohm impedance? I would recommend using a longer piece of coax as the 12 inches puts the 1090mhz wavelength of 11 inches right in close proximity to the coax connectors. My experience also says never use 90 degree coax connections as they are proven to possibly severely degrade the RF signal quality. I myself have proven this effect using some very expensive quality RF meters. There is such a thing as communication cables being “too short”.
Thank you. It makes sense to avoid using a splitter. I will get a separate antenna for the 1090 Mhz frequency and will probably get a new dongle that has a built-in filter.
When I mount this outside, I plan to have the antenna about 10 feet above the roofline with the electronics in a weather resistant box. I plan to mount the box below the roofline on the north side of my house so that it is not directly in the sun
Does anyone have any thoughts if it makes a difference to use a 10 ft coaxial cable to connect the antenna to the dongle or should I use a shorter coaxial cable and use an (active) USB cable to connect the dongle to the RPi?
The blue antenna interface with the built in filter is a good idea. I run that direct connected to the RPi and have a 15 meter 50 ohm coax from FlightAware that I run down from the antenna to the antenna input at the Rpi.
Thanks. I’ve been thinking about using a coaxial cable that is at least 10 ft long to connect the antenna to the dongle and directly connecting the dongle to the RPi. But, I’m wondering if there might be any net benefit for using a shorter coaxial cable to connect the antenna to the dongle and using a USB cable to connect the dongle to the RPi that would be in a separate enclosure. I’m thinking the benefits might be lower signal loss by using a shorter coaxial cable, less chance of the dongle and RPi interfering with each other because they would be separated by some distance in two different enclosures and possibly fewer problems with heat because all of the components would not all be in the same enclosure. Any thoughts on this?
Yeah, there’s no “loss” per se in a USB cable (other than some IR drop), whereas there is a signal loss in a coax cable at 1090 MHz. I use a USB cable between dongle and RPi of 15 feet with no problems.
Personally, I would put an LNA at the antenna and keep the pi somewhere accessible so that you can change the SD card if you need to re-image. With the LNA at the antenna you won’t really need to worry about the loss from your feeder cable as the system noise figure will largely be determined by the LNA.
If you can find a suitable LNA to cover both 978 and 1090MHz then you will be able to use a single antenna for both UAT and ADSB and place your splitter at the pi end of the co-ax. The LNA will then compensate for the losses in both the feeder cable and the splitter.
Thanks. Accessibility of the Pi is something that I’ve been thinking about. That’s why I’ve been inquiring about the use of active USB cables to separate the dongles from the Pi. This would allow me to have a short coaxial cable from the antenna and have the Pi in an enclosure lower on the mast so that I could reach the Pi from the ground.
It seems to make sense to avoid using on antenna with a splitter, so I plan on getting separate antenna for 1090 and 978 Mhz.
I thought the FlightAware Prostick and Prostick Plus have LNAs built into the dongles. Is there a benefit to having an additional LNA, such as the RTL-SDR.com LNA, at the antenna?
