Well, this is weird. Now the CPU load is back down to 35% or so. Perhaps there’s a step function in the processing performance required that’s not linear. While the number of planes right now is less than this morning, it’s not 10x less.
Very curious. In any event, I’m hoping that the use of the Pi3 will alleviate the issue!
I have their original design. It’s a big honkin piece of metal 
Process usage (numbers vary widely, this is just a snapshot in time)
piaware 105.9%
dump1090 25.5%
fa-mlat 4.3%
faup1090 1.0%
top 0.7%
Not sure why it goes back and forth between moderate usage (piaware ~ 30-35%) and this.
Ok, now it’s back to “normal”.
piaware 4.3%
dump1090 27.8%
faup1090 8.3%
fa-mlat 13.8%
Again, it’s an RPi 2.
Anyone know how much resource the browser display (i.e., nnn.nnn.nnn.nnn:8080) takes up?
It is highly unusual for piaware to be at 100%, or really anything over 5%, and I’d be interested in knowing what it was doing just then (and which version it is)
Are you using 5Ghz wifi, or something else? How do you power your pi? Did you just leave it off in a remote low traffic area and cross your fingers, or do you have official “permission”? Considering trying this myself. My house has direct line of site to Mt. Ord and Four Peaks.
It’s at a regular radio communications site, one of many on the hill. APS is happy to sell us electrons. My share costs about $36 a year. There was no internet connection, so I’m using 5 GHz UBNTs to do backhaul.
Mt. Ord is loaded with radio sites, and I’m sure if you could find a ham interested in ADS-B and that had access to a site up there anything’s possible 
The Arizona Repeater Association has a large site up there and may have internet service as well. (w7ara.net). Some in that group are flying enthusiasts.
Four Peaks Brewery would be fine and make it even more worthy of regular visits , but Four Peaks mountains are a designated wilderness, so that would be unwise…
Cheers - Jon N7UV
I’ve had a chance to do more watching, and I’ve seen as strong as about -1.5 for a plane flying almost directly overhead my position on a course to land about 11 miles away. Usually, they’re at about 3000 feet this far out, but this plane was quite a bit lower. I don’t have the data immediately at hand, but that’s what I recall. Anyway, you know how the paths sort of change to a dashed pattern when the signal is not good? It did that for about 3 seconds as it was approaching my position. I did not note the RSSI value around that time, but I suspect the trail became dashed because of signal clipping. At that point, it was less than a mile away. I can tolerate clipping for brief periods on the very closest planes if it means I get better reach at the far end.
Regarding your comment about not so much range, I wonder if that’s related to your being in a desert and my having to look through trees near the horizon. I notice that planes approaching KAUS, which is about 11 miles away, are generally being received at about -2 to -4 almost until they touch down. After or around the time they touch down, the signal deteriorates rapidly until about -14 or so, usually. Note that if I stand on top of my roof, I cannot see the airport. There are trees and buildings in the way. My antenna is a few feet higher, but trees will still be in the way, I suggest that may be the cause of the differences you note between our spreads.
There are trees and buildings in the way. My antenna is a few feet higher, but trees will still be in the way
Try again in midwinter if the trees are deciduous. I see a summer/winter difference in the direction of a tree in the front yard.
Do you have heat sink installed on your raspberry pi?
Yes, but it wasn’t enough to stop the throttling. I added a case with a fan and it is now very happy.
What fan case are you using? I’m currently just using a room fan, which works well, but a built-in RP fan would be better.
This is the case I use
amazon.com/gp/product/B019S … UTF8&psc=1
Makerfire Raspberry Pi B+ Case Protective Case Enclosure Box with Mini Cooling Fan Heatsink Kit
I just noticed a flight reporting -49.5 dBFS. It’s unwavering. So I suspect it’s not accurate and possibly represents a bug. I don’t think I’ve ever seen a value below around -20 other than this one.
If I click on the Flightaware link for this plane I get a message that says invalid code received. ICAO shows as 40338f. It’s GLF4, judging by flightradar24.com
I never did either until I switched from AGC to manual gain.
Now I’m routinely seeing RSSI values approaching -30 dBFS for aircraft near the 200 NM ring. Overall, it seems my RSSI values have since increased quite a bit no matter what the range of the target aircraft.
Go figure.
I saw this as well and I think it was one of the newly anonymized MLAT planes. To me it looks like the MLAT feed for such planes does not include an RSSI value or else includes a value of -49.5 dBFS.
-49.5 dBFS is a raw signal level of zero, so yeah, you will see this for aircraft where you are receiving only synthetic messages (anonymized mlat, mlat during the period after the aircraft has left your reception range but before the mlat server stops feeding you results)
The particular plane involved passed almost directly over my location, and I received messages from about 200 nm away both coming and going. It displayed -49.5 the entire time.
When you say “anonymized mlat”, are you referring to aircraft that have been obfuscated to keep the information from the public?
See the mlat announcement post for details.
Hi, I have tried to reverse engineer your set-up, is this correct?
Would be very cool to see a bill-of-materials for this…
simpler and better … The Final Filter Shootout
