Raspberry Pi type B + DVB-T Dongle to feed FR24

[Stephen P]

Cheers peterhr, actually deleted ALL of dump 1090 and tried for " git clone git://github.com/MalcolmRobb/dump1090.git " hoping it would pick up updated version. Got dump1090 back and uploading without missing a beat "apparently". Google is a beautiful thing if you know what to do with it. Ill have a crack at the script, its only downtime on my radar which I've become addicted to. Thanks again

[peterhr]

Cheers peterhr, actually deleted ALL of dump 1090 and tried for " git clone git://github.com/MalcolmRobb/dump1090.git " hoping it would pick up updated version. Got dump1090 back and uploading without missing a beat "apparently". Google is a beautiful thing if you know what to do with it. Ill have a crack at the script, its only downtime on my radar which I've become addicted to. Thanks again

Excellent. The only reason for moving it to one side is it could be moved back if there was a problem.

[Harrie45]

but my planes have been doing strange things.
Thanks

Can you be more specific? Ground level?

[Stephen P]

Harrie45, in the air they have been jumping around and showing behind plot a bit but that sorta was explained reading forum.
I am really close to the airport, really close, and the planes run off runway. Similar to the explanation of last speed and heading to be received, continuing being used after touchdown.(i can see the smoke of the tyres so i know its down). My aim is to try and fix this issue of 0 speed, 0 altitude. Seems like you have it solved so am keen to check if it changes things from my end. We do have ground vehicles here but they don't show like Brisbane. Maybe they are not required to transmit here?

[peterhr]

Harrie45, in the air they have been jumping around and showing behind plot a bit but that sorta was explained reading forum.
I am really close to the airport, really close, and the planes run off runway. Similar to the explanation of last speed and heading to be received, continuing being used after touchdown.(i can see the smoke of the tyres so i know its down). My aim is to try and fix this issue of 0 speed, 0 altitude. Seems like you have it solved so am keen to check if it changes things from my end. We do have ground vehicles here but they don't show like Brisbane. Maybe they are not required to transmit here?

dump1090 has very recently been updated to fix 'planes on the ground' issues (just look back a few posts)

Please log on to your Pi using PuTTy

do

mv dump1090 dump1090.old

then follow the 3 lines of instructions in the write up to download and make a new version of dump1090

git clone git://github.com/MalcolmRobb/dump1090.git
cd dump1090
make

after that do

cd ~
./fr24.sh

[Harrie45]

Harrie45, in the air they have been jumping around and showing behind plot a bit but that sorta was explained reading forum.
I am really close to the airport, really close, and the planes run off runway. Similar to the explanation of last speed and heading to be received, continuing being used after touchdown.(i can see the smoke of the tyres so i know its down). My aim is to try and fix this issue of 0 speed, 0 altitude. Seems like you have it solved so am keen to check if it changes things from my end. We do have ground vehicles here but they don't show like Brisbane. Maybe they are not required to transmit here?

PeterH did give the answer already. Use the latest dump1090. Planes running too far (even off the airport) is because of the extrapolation that is done by flightradar24. But eventually the plane should jump back to the right position and turn to the correct heading. The extrapolation is done because the upload of your data is done every 5 seconds or even less frequently.
The last days I also see planes moving odd when in the air. I think this is caused by extrapolation of data of radars that are just out of reach. When another radar is selected I see that the heading and speed is corrected. The radar selection algorithm is still a puzzle for me. I believe it is something like F- radars are in favor, then N- and then T- based on latency.

[peterhr]

PeterH did give the answer already. Use the latest dump1090. Planes running too far (even off the airport) is because of the extrapolation that is done by flightradar24. But eventually the plane should jump back to the right position and turn to the correct heading. The extrapolation is done because the upload of your data is done every 5 seconds or even less frequently.
The last days I also see planes moving odd when in the air. I think this is caused by extrapolation of data of radars that are just out of reach. When another radar is selected I see that the heading and speed is corrected. The radar selection algorithm is still a puzzle for me. I believe it is something like F- radars are in favor, then N- and then T- based on latency.

Apparently FR24 takes the data from all the feeders and combines it in some way to make sure they agree where the plane should be. FR24 say there is no preference given to showing their own receivers as the source - it make no difference if we believe them or not.

[Stephen P]

Just to follow up, i had followed instructions to update dump 1090 the first time but upon checking lines 373-375 as suggested by Harrie45, they hadn't been updated at that time, hence my questions. I again yesterday did update as per peterhr's post, checked 373-375 had been changed and they were. Might have taken a day or so for new update to come online. All good again. Things were quiet round my area yesterday due to bad weather but will keep an eye on things to see how new update works and if it has solved the problem. Once again thanks for your help and info.
As for FR24 using data from all feeders and combining, i am confused. On my local dump1090 feed i can see full ADSB aircraft, but when comparing with FR24, the planes aren't there. I have several F-Radars around me as well as much more powerful T-radars so i would have thought that if i was picking up signal from my radar so would the others. Therefore a plane should be visible. These are not MLAT planes either. We DO have a lot of smaller aircraft in my area that MLAT picks up, im still waiting to hear regarding FR24 box so i can help feed MLAT also, but no word as yet. Been quite a few months now so not holding my breath. Anyways food for thought.
Cheers

[peterhr]

Just to follow up, i had followed instructions to update dump 1090 the first time but upon checking lines 373-375 as suggested by Harrie45, they hadn't been updated at that time, hence my questions. I again yesterday did update as per peterhr's post, checked 373-375 had been changed and they were. Might have taken a day or so for new update to come online. All good again. Things were quiet round my area yesterday due to bad weather but will keep an eye on things to see how new update works and if it has solved the problem. Once again thanks for your help and info.
As for FR24 using data from all feeders and combining, i am confused. On my local dump1090 feed i can see full ADSB aircraft, but when comparing with FR24, the planes aren't there. I have several F-Radars around me as well as much more powerful T-radars so i would have thought that if i was picking up signal from my radar so would the others. Therefore a plane should be visible. These are not MLAT planes either. We DO have a lot of smaller aircraft in my area that MLAT picks up, im still waiting to hear regarding FR24 box so i can help feed MLAT also, but no word as yet. Been quite a few months now so not holding my breath. Anyways food for thought.
Cheers

FR24 claim to do this - who knows what really happens, it would be sad to think they are being fed aircraft info and they are not seeing it.

[Stephen P]

Update works sweet, even in pissing down rain. Stops at end of runway, return track and parking also. As for data fed?? like you said "Who knows?"
It WOULD be sad. All food for thought.

[peterhr]

Excellent

I'm trying the FlightAware feeder alongside the FR24feeder on the Pi

see https://flightaware.com/adsb/piaware/ - it seems to present little load on the pi

No stats from them yet - lets see what we get.

top - 12:44:00 up 12 min,  1 user,  load average: 1.37, 1.23, 0.76
Tasks:  62 total,   2 running,  60 sleeping,   0 stopped,   0 zombie
%Cpu(s): 41.1 us, 33.1 sy,  0.0 ni, 18.2 id,  0.0 wa,  0.0 hi,  7.6 si,  0.0 st
KiB Mem:    447996 total,    71864 used,   376132 free,    10880 buffers
KiB Swap:   102396 total,        0 used,   102396 free,    25256 cached

  PID USER      PR  NI  VIRT  RES  SHR S  %CPU %MEM    TIME+  COMMAND
 2094 pi        20   0 15780 6064  920 R  60.2  1.4   7:04.00 dump1090
 2095 pi        20   0 47300 4064 1368 S  15.7  0.9   1:52.26 fr24feed_arm-rp
[color=red] 2110 root      20   0 18948 6624 3112 S   1.6  1.5   0:10.11 piaware
 2115 root      20   0  2776 1092  768 S   1.3  0.2   0:08.78 faup1090[/color]
 2144 pi        20   0  4672 1372 1044 R   1.0  0.3   0:02.79 top

[Mutt08]

I installed the Flight Aware feeder on the pi last night & all went well & within about an hour I was showing as feeding in my FA account.

Did a reboot of the pi this morning & now the feed to FA has stopped, faup1090 is running but piaware doesnt seem to be.

I ran the start command but the pi said that piaware was already running, just wondering if their is a conflict of some type on reboot!

Any thoughts on this Peter?

Tasks: 62 total, 2 running, 60 sleeping, 0 stopped, 0 zombie
%Cpu(s): 30.2 us, 22.9 sy, 0.0 ni, 40.9 id, 0.0 wa, 0.0 hi, 6.0 si, 0.0 st
KiB Mem: 447996 total, 77108 used, 370888 free, 15072 buffers
KiB Swap: 102396 total, 0 used, 102396 free, 25796 cached

PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
2099 pi 20 0 15780 6052 908 S 46.1 1.4 17:57.21 dump1090
2100 pi 20 0 47300 4056 1368 S 9.5 0.9 3:46.98 fr24feed_arm-rp
2109 root 20 0 2776 1044 720 S 1.0 0.2 0:19.15 faup1090
2208 pi 20 0 4672 1352 1024 R 0.7 0.3 0:00.11 top
2002 ntp 20 0 5516 1716 1340 S 0.3 0.4 0:00.52 ntpd

[peterhr]

I cheated

I did

sudo update-rc.d -f piaware remove

to stop the service starting at boot.

Then in the fr24.sh added

"sudo service piaware stop" before the 'killall' statements

and

"sudo service piaware start" after restarting the fr24feed

(all on separate lines) - could use sudo /etc/init.d/piaware start / stop / restart / whatever either works.

BTW the service startup script creates a file as it starts piaware - I expect it tests for that file to see if it's already running.

(been feeding for one day, should pass 'oblivian' in the FA feeder rank soon - and winning the radarcode war counts for everything!!!!! - Oh Yes!)

[Mutt08]

Thanks for that Peter, Edited the fr24.sh & alls working great now.

I managed to get around position 35 in the 'Top Sites' ranking before I went away on holiday for 2 weeks & had to switch off the laptop that was feeding through Planeplotter.

Now that I'm feeding through the pi I've had to start from scratch with a new feed, so lets see how high it climbs this time...

BTW For the feeders on this thread that start feeding to Flightaware through their pi's, why don't we set up a 'Flight Crew' & see what combined stats we can achieve & see if we can knock the American feeders from the top

[YSWG]

What is the load like for Virtual Radar Server running on mono? Wanting to move away from a high energy PC (which the fan is stuck on a high rpm). Tried doing an emulate of a Raspberry Pi on a laptop but had no luck in getting to install rtl-sdr and Peter's image fails to mount, so have to get the real thing (~ $50 AUD). How is the B+? Better then the original B?