lots of traffic currently with > 100 aircraft, this is how the CPU utlization look at the moment of the Air!Squitter which i am playing around with today again:
The dump1090 instance which is still running on the second Raspberry (Model 4)with the antenna indoor is currently on a similar CPU level
Just out of curiosity, do you have root or sudo as well, or just user access?
Yes, full SSH access including root permissions, not only the web interface.
The screenshot of htop was taken after launching with sudo command (process 3509 in screenshot)
Jetvision offers an image for download in case you “overpace” your work on the device. As the SD card is non-removable, you can launch it in a special way connected to a PC. From there you can use Etcher to flash it as removable device like any other SD card
After having seen the internals ( thx anonymous person 
), it appears to be a bone-stock Orange Pi Zero 512, so if you take the lid off, it can be changed no problem.
How does this work exactly? I’d be curious to learn that trick as it should work for any SBC unless they have special firmware on something.
The FPGA is an Altera Cyclone IV, the wifi is an rtl8811cu plastered onto an M2 board and the GPS is a Ublox MAX-M8W-0-10. Interesting setup.
ADD: RE FPGA Cyclone IV ADS-B VHDL, this could be on interest for someone:
The RF hardware on a bladeRF is totally different though.
The FPGA-based decoder on the airsquitter is, AFAIK, just an evolution of the FPGA-based decoder on the Beast, Radarcape et al.
I haven’t tried it yet, but will let you know how it works once i need it.
From their Wiki you have to press the button at the front while powering it on which brings it into a “special mode”
The recovery image itself is approx 250 MB in size.
So, now i made a decision regarding keeping or not.
After some days of testing it seem to be that the device is not giving me an improvement to my current setup (Raspberry PI3B - FA Pro Plus Stick).
Comparing the number of Positions and Flights reported together with the hourly stats it delivers similar results. I think my setup is already delivering the best performance based on my location.
The only significant improvement would be the number of messages/frames per second which is up to 25% higher than on the Raspberry. But as long as the rest of the results is the same, it’s even for me too expensive.
Thanks to the return policy Jetvision already accepted my return without problems which i will do now.
Don’t get me wrong:
The device is running perfect, no question about that. Especially the setup is easy without any specific linux skills (they help in digging deeper into) delivering very good results.
Maybe even better on a different location or with one of their active antennas.
But spending another 180 bucks for their “Active Diapason” antenna will end me up in water and bread for the next three months 
Overall question if i can recommend it? It’s a yes if you want to have a “fire and forget” tool.
I just went through the same process with the Airspy R2. Extensive testing and results are about the same with the FA blue Pro Plus stick with Uputronics LNA vs Airspy with Uputronics.
I was curious to see if the Airspy would improve handling of attenuated signals from the forest around my house. While the Airspy did things a bit differently and kept planes with messages (waiting a long time for new positions) on my map longer, overall the FA stick with Uputronics actually performed about equally for max range and total positions, and slightly better for the balance of consistently tracking closer low flying aircraft, avg max range, and plane count. My range is limited by mountains in some areas so lowerish gain gives the best possible balance of coverage and total numbers. With Airspy, I couldn’t dial in the same balance with range at the edges and maintain solid tracking of low altitude planes 15 to 50 nmi away.
I like Airspy’s options. I tested all of them. Also tested with a band-pass filter between Uputronics and Airspy but that did more harm than good for total metrics but it did increase the dynamic range.
I’m not returning the Airspy though. Too much fun for other projects and my backup RPi and antenna toys. I’m sure I’ll give it another try some time. They seem to be continuously optimizing firmware and stuff. One thing that might help could be higher resolution gain adjustments. Like 1 or 1.5 db adjustments vs 3db. As you tweak some of the other Airspy adjustments, you occasionally want to make a gain adjustment to compensate.
Yes, it’s interesting to see how that all work together. For me it’s good to have an evidence that my Raspberry setup went to the maximum.
I still did not pack it up, the return policy says 14 days. Maybe i revert my decision.
When i start thinking where i came from a year ago with almost no knowledge about fligtg tracking i decided to start with one of the Raspberry starter kits. This contained a Raspberry 3B in a case, power supply. I decided to start with that as it conained already the blue stick with a simple indoor antenna . This i used a while with some simple tweaks (removed the tip and replaced it with a piece of copper cable in the correct length).
But as usual you want more. So i built two different DIY spiders.
After i confirmed myself that i am not the best antenna builder i got a cheap one from an enthusiast on ebay. This was optimized to exactly 1090 MHz. We then then moved in December and i got a second Raspberry because of a blind spot having the antenna outdoor (own roof in the way).
Later i decided to get the Jetvision Antenna which performs slightly better than the eBay Antenna. So that guy did a very good job. This antenna is now feeding my indoor Raspberry while the Jetvision is outdoor on my balcony (we have two, otherwise my wife would have killed me)
Honestly without this forum and all the experts i never would have gone that far.
For me it’s fascinating that the Air!Squitter (Radarcape) is simply plug&play with optimal results out of the box.
To be continued…
Finer gain adjustments aren’t possible.
The existing 0 to 21 scale works quite well i think.
1 point on the scal is 3 dB.
I can’t see how that could be.
Low altitude at 50 nmi is just gonna be very weak signal if any.
RPi Monitor installed on Air!Squitter. Seem to be that the CPU itself doesn’t have too much to do, even in busy times.
Decision made now, device will be returned to Jetvision.
It might be performing better in rough conditions, but for me it does not give me that improved reception which justifies the price of 300 Euro. I assume i am already at the max which i can get.
However the device is well built and can be set up easily without extended skills and delivers good results out of the box.
After the return i now have a brand new Raspberry Pi 4B with 2 GB RAM. This is the replacement of my outdoor Raspi 3B which does not have an active cooling and got pretty hot the last days.
The Raspi 4 is using an active Armor housing, yesterday the temp did not exceed 50°C where the Raspi 3 went up to 65° in the same climate the day before.
Just out of curiosity:
Is it possible that the RPi 4 performs a little better than a 3B from the reception?
I do have the feeling that the range is a little larger and i needed to reduce the gain by 2 steps to reduce the number of messages > 3dBFS.
Also the message rate seem to be a bit higher for the same amount of aircraft.
All settings were the same for both of them.
FA Pro Plus stick is connected via USB extension to one of the USB3 ports.
Or is this just by chance?
On the digital side it doesn’t matter if it’s a 3 or a 4; it’s still running the same code and that will produce the same result given the same input.
The different boards might have different noise characteristics in your setup.
Thanks for the explanation
Look at my graphs here and switch to one year.
You can see from the ADS-B CPU Utilisation that I installed the AirSpy at the end of September last year and then I switched from a Pi3B+ to a Pi4 in November. The Memory Utilisation graph is also a good indication of when I switched to the Pi4.
I also went from 12MHz to 20MHz so you can see exactly where the Message Rate went up, it ties in with the Pi4.
The number of aircraft seen and tracked also went up at the same time. I don’t think I made any other changes at that same time, my aerial had ‘been raised 7ft and in the clear’ was in August 2019.
So I think you may see an increase.
Yes, but you are using an Airspy where a Raspi4 is highly recommended for performance improvement.
I am using a FA stick (blue Pro Plus) where this should not be a problem.
Rpi3 to Rpi4 could mean more stable voltage for the SDR, that shouldn’t really have an effect but apparently it does.
Maybe it’s related to ripple or something, not sure.
People have also reported better performance using powered hubs.
So maybe it’s better voltage, less ripple on the voltage or it’s different RF noise characteristics.
Impossible to tell without having stupid amounts of test gear probably (and lots of knowledge about RF signals).
