ADS-B DIY Antenna

[ab cd]

.....snip.....
Finally I took it down and re-imaged the SD card and put it back in service and it's working fine now.
....snip....

The easiest way to get out of software problem is: Format drive ->Fresh Install.
I have learned this in days of Win 95, Win98 & Win XP, when Anti Virus Protection was not very strong and Hackers were. I will simply copy my data on Floppies, format the Hard Drive, and reinstall Windows & all. Took few hour's time, but was better than breaking my head to find out & fix the problem.

 

[ab cd]

I too have had crashes numerous from both FA version of dump1090 and mut version.
Back now to Malcom Robb fork but a question ,
If dump1090 is run using ssh from command line :
cd dump1090
./dump1090 --interactive
There is a column "Signal" which is very handy for testing antenna installations.
I know that some wise users have customised their dump1090 :8080 web page display with the addition of range rings and other items.
Is it possible to omit an existing column such as "seen" and replace with "signal". Or just add the signal column to the display grid.
Just a thought and a request/wishlist for you clever chaps out there.

See the RSSI column in the table right side of dump1090 :8080 web page (experimental version of xforce30164)

By the way, I used the guide from kind member ab cd to install PF Client on my rpi which went okay .Thanks ab cd...but
1. Surely should there not be an official document/guide from PF regarding installation.

This is the official guide Planefinder sent to me (original in pdf):

Installation of Plane Finder Client (PFClient)
Tested on Windows, Mac and Linux - others may, and should!, work too.
Any questions or feedback please do let us know and we will try our best to help.
If using RTL please refer to the RTL driver/software notes at the end of this document
Installation of PFClient
• Install NodeJS from http://nodejs.org for your version of operating system. (Please note
that Linux users may need to install from distribution repositories specific the OS - “npm
config” can help with latest version and sudo / root required
• Run the following command:
npm install -g http://clientfiles.planefinder.net/pfclient-latest.tgz
(For windows users do this from a command window)
• Run “pfclient” and follow the steps for configuring (the options are all listed below)
If you get the settings wrong just launch “pfclient -config” to reset the configuration and start
over.
Run pfclient and watch for planes on the browser window that pops up. For users with no GUI
browse to :30053" target="_blank" class="externalLink ProxyLink" data-proxy-href="http://:30053" rel="nofollow">http://<ip-address-of-machine>:30053
Note - If you prefer not to load the browser view just add the command line option
"--showBrowser=false"
Options for PFClient installation
Receiver Type
• puck - For connecting to SBS3/Puck receiver by Kinetic. Can either be raw 30006 output
over the network from Basestation or direct to the SBS-3. Also supported are serial
connections to these units.
• avr - Any AVR format data.
• beastbinary - Handles a beast set to binary mode.
• 30003 - Handles a decoded 30003 output. If you have this available this will be the least
CPU intensive.
Connection Type
• network - Supports any of the receiver types over a network connection. You will be
prompted later for IP & port details.
• serial - Supports Kinetic SBS-3/Puck & Beast receivers over a serial connection. You will
be prompted later for serial port and baud rate of these devices.
Home Location
• Enter the receiver’s location as decimal latitude & longitude separated by a comma. i.e.
50.11340,-1.22220
Username & Password
• These will have been provided by the Pinkfroot team and allow uploads of data to the
Plane Finder servers
IP Address & Port
• IP address of the receiver - this can be 127.0.0.1 if a local piece of software is running
the receiver and you are connecting to this i.e. Dump1090, BaseStation or Virtual Radar
Server.
• TCP Port - Port address to connect to for the type of feed you have configured.
Serial Details
• COM Port - on Linux these are typically /dev/ttyUSB* or something similar!
• Baudrate - Most receivers default to 3Mbps (3000000) so the default is this but it can be
changed.
Once complete your data should be showing in Plane Finder and in the local browser.
RTL Dongle – Windows Driver/Software Notes
• Install drivers for RTL: http://forum.planefinder.net/threads/installing-rtl-drivers-forwindows-
for-aircraft-tracking.94/
• Download and install ADSB# (http://sdrsharp.com/downloads/adsb-install.zip) or use
your preferred software tool – such as RTL1090
• Run ADSB# with all defaults
• Configure the “pfclient” to use the following settings (for ADSB#)
• Receiver Type: “avr”
• TCP Port: 47806
RTL Dongle - Linux Driver/Software Notes
• Download and install Dump1090 as per http://www.satsignal.eu/raspberrypi/
dump1090.html
• Configure the “pfclient” to use the following settings
o Receiver Type: “30003”
o TCP Port: 30003
Any questions or feedback please do let us know and we will try our best to help.
Don’t forget to ask for Android or Apple apps if you can use them!
Thank you sharing your data and for your help and support.
[email protected]

2. The install is lengthy,can this not be simplified into a single .deb install file ?
And please PF /Pinkfoot please tell us how to uninstall if things go wrong in future...good luck with googling that !

If the install was simpler, and and official guide to installing and uninstalling made I am sure you would get more users sharing data. Lord only knows how members managed to set up before ab cd created the guide a few months ago.

I also wish they have a light-weight, easy to install, data feeder software.
I have communicated to them, and they say they are working on a revised & improved version.

 

[ab cd]

Well, some good news. I have figured out why the primary feeder ("magic" spider) was reporting such low numbers this week: the grounding block! When I added new coax to the secondary Pi's antenna I did two other things that day. 1) I added the decoupling coils to the coax of each antenna, and 2) I added the grounding block. Since that time, my reception has been down. I didn't know if it had to do with weather or the coils, or the ground, or even Raspbian, or the dongles - or what! This morning I decided to disconnect the primary Pi's coax from the grounding block and connect it directly to the receiver. Immediately the messages per second doubled! Yup! The number jumped from the 300's to the 600's just by cutting out the grounding block! I guess the block had a pretty large loss factor.

Another advantage of strictly indoor installations... no grounding blocks required

Advantages include, but not limited to, following:
1. Lightning Stroke proof
2. Snow proof
3. Rain proof
4. Wind proof
5. Dust proof
6. Corrosion proof
7. Roof Climbing proof
8. At arms reach without a ladder or stool.
9. Fully air-conditioned working conditions (for both the antenna & the antenna owner) 24/7/365 - 75 deg F, 50% humidity.

 

[xforce30164]

(--snip--)
Run pfclient and watch for planes on the browser window that pops up. For users with no GUI
browse to :30053" target="_blank" class="externalLink ProxyLink" data-proxy-href="http://:30053" rel="nofollow">http://<ip-address-of-machine>:30053
Note - If you prefer not to load the browser view just add the command line option[/FONT]

Wow, cool, I didn't know pfclient also had an gui, need to add that to my control room/center options/webpage

Yes thats great work xforce,
I am thinking perhaps your RSSI version may be slightly confusing for non radio users (the lower the negative value, the higher the signal strength received).
The "sig" table in dump1090 --interactive uses a positive scale 0-100 which may seem more user friendly and is easier for a user/non tech savvy to compare constantly changing values.
Nice one though xforce, im impressed.

Is there a way to sort the list of aircraft dispalyed by sig strength ,distance, altitude and so on by clicking the table header ?
In other words, by clicking on the green band header called "distance" would sort the aircraft list by the closest to the station first and the farthest away at the last position in the table.Or vise versa by clicking on the distance header once more.. Now that would be marvelous as a user could arrange their own way to show the dump1090 output aircraft list "on the fly".
Great for station testing purposes.

That's a nice idea, AFAIK that is currently not possible, but will put it on my to-do list .

Also, got a PHP-based table that is completely flexible, just define what colums you want to see in the url by separating them with ';'.
For all available information go to:

http://dump1090.ayadex.com/modular/index.php?cols=all

or you could do:

http://dump1090.ayadex.com/modular/index.php?cols=hex;flight;altitude;rssi;speed;track

 

[xforce30164]

Btw, just wondering if anyone ever tried a BiQuad antenna as directional booster for 1090mhz?

(something like http://trevormarshall.com/biquad.htm)

I might try and make one coming week.

Edit:
I added/placed an empty tin can under the small stock antenna of my receiver near Schiphol, and It got between 50%-100% more messages compared to before. (still roughly same amount of planes, so higher resolution data)

 

[ab cd]

Try these links for PF Client web interface:
http://<ip-address-of-RPi>:30053 (shows a table of planes)
http://<ip-address-of-RPi>:30053/gm.html (shows planes on Google Map)
http://<ip-address-of-RPi>:30053/ge.html (shows max range plot on Google Earth)

. .

 

[ab cd]

...snip....
I added/placed an empty tin can under the small stock antenna of my receiver near Schiphol, and It got between 50%-100% more messages compared to before. (still roughly same amount of planes, so higher resolution data)

The stock antenna has a small magnetic base which provides the ground plane, but insufficient. Putting a can improves it. Try a larger can or a metallic/foil food plate, will result in improvement.

Monopole without ground plane is the worst antenna. I tried a "zero-groundplane" monopole antenna by simply cutting-off the shield of 69mm tip of vertical feeder coax. It proved to be worst than stock antenna, which has some groundplane (magnetic base)

 

[xforce30164]

The stock antenna has a small magnetic base which provides the ground plane, but insufficient. Putting a can improves it. Try a larger can or a metallic/foil food plate, will result in improvement.

Monopole without ground plane is the worst antenna. I tried a "zero-groundplane" monopole antenna by simply cutting-off the shield of 69mm tip of vertical feeder coax. It proved to be worst than stock antenna, which has some groundplane (magnetic base)

View attachment 1409

thanks for the idea, btw does the groundplane have to be circular metal or can it be any large metal object?

 

[ab cd]

thanks for the idea, btw does the groundplane have to be circular metal or can it be any large metal object?

Groundplane need not be circular.
Any electricity conducting object can act as ground plane. A vehicle's body acts as the ground plane for vehicle's whip antenna. The core of feed cable is connected to whip & shield is connected to vehicle's body.

Another thing which is not groundplane: Touch DVB-T Dongle's whip antenna, and see the increase in number of planes. Have you ever noticed that if an FM station is weak, and you touch the telescopic antenna of your FM Radio, the reception improves. Human body conducts electricity, hence acts as a whip antenna and picks RF signal.

DO NOT try this with Transmitter or Transceiver Antenna, as they pump enough power to feed line to cause RF burns to any one touching it.

 

[xforce30164]

(snip)
Another thing which is not groundplane: Touch DVB-T Dongle's whip antenna, and see the increase in number of planes. Have you ever noticed that if an FM station is weak, and you touch the telescopic antenna of your FM Radio, the reception improves. Human body conducts electricity, hence acts as a whip antenna and picks RF signal.
( snip )

Reminds me of this Mr. Bean episode .

And good to know about groundplane objects. I'll experiment with different ones when I have some time to do so.

 

[ab cd]

"zero groundplane" antenna fitted with large groundplane

 

[ab cd]

Something like this?
View attachment 1383
...snip....

The RSSI sort function does not work on clicking the RSSI header of the table.
Reason: function names in gmap.html and script.js are different
sortBySignal() in gmap.html
sortByRSSI() in script.js

You wrote:
In the file gmap.html add:
<td id="signal" onclick="sortBySignal();" align="right">RSSI</td>

In the file script.js add:
function sortByRSSI() { sortBy('signal', compareNumeric, function(x) { return x.rssi; }); }

 

[radiostationx]

In dump1090 --interactive at the command line , the received signal value is 0-100, not RSSI which I feel would be more user friendly if it were placed in the 1090:8080 map & table output view as 0-100. Trying to resolve quickly changing negative values when moving antennas or experimenting makes things more difficult.
PF web output found on :30053 does not use the signal values in the aircraft list.

Planeplotter has a very handy function too, a messages received rate signal meter which is autoranging but the user can manually change ranges by a right click on the meter itself, this opens in a secondary window by clicking "view" and "message rate -m" a user can position this anywhere on top of their map display.

I wonder if some wizard out there could produce same sort of thing in the space underneath the UTC clock in dump1090:8080.
Creating a total/combined msg received meter or scale per sec at the top under utc clock and signal received from each aircraft in the table. Cool !

 

[bramj]

Feeder Coax is parallel to vertical limbs at a distance of 1/4 wavelength. Too close, the shield of coax may severely affect reception.

I changed the position off the feeder coax and the performance of the antenna has increased.
The raspberry is really working hard around 85% cpu utilization..

 

[xforce30164]

I changed the position off the feeder coax and the performance of the antenna has increased.
The raspberry is really working hard around 85% cpu utilization..


View attachment 1416 View attachment 1417 View attachment 1418

That's alot of planes bramj! very nice antenna

In dump1090 --interactive at the command line , the received signal value is 0-100, not RSSI which I feel would be more user friendly if it were placed in the 1090:8080 map & table output view as 0-100. Trying to resolve quickly changing negative values when moving antennas or experimenting makes things more difficult.
PF web output found on :30053 does not use the signal values in the aircraft list.

Planeplotter has a very handy function too, a messages received rate signal meter which is autoranging but the user can manually change ranges by a right click on the meter itself, this opens in a secondary window by clicking "view" and "message rate -m" a user can position this anywhere on top of their map display.

I wonder if some wizard out there could produce same sort of thing in the space underneath the UTC clock in dump1090:8080.
Creating a total/combined msg received meter or scale per sec at the top under utc clock and signal received from each aircraft in the table. Cool !

I'll add it to my todo list as well, but just to make sure, the "graph"-type scale on the background gives the converted (0-100) RSSI strength (averaged over last received messages?) and the text is just the messages per second?

 

[xforce30164]

The RSSI sort function does not work on clicking the RSSI header of the table.
Reason: function names in gmap.html and script.js are different
sortBySignal() in gmap.html
sortByRSSI() in script.js

You wrote:
In the file gmap.html add:
<td id="signal" onclick="sortBySignal();" align="right">RSSI</td>

In the file script.js add:
function sortByRSSI() { sortBy('signal', compareNumeric, function(x) { return x.rssi; }); }

Whoops, thanks for spotting that one!

 

[radiostationx]

That's alot of planes bramj! very nice antenna



I'll add it to my todo list as well, but just to make sure, the "graph"-type scale on the background gives the converted (0-100) RSSI strength (averaged over last received messages?) and the text is just the messages per second?

Hi xforce,
The ideal flavor of my dump1090:8080 I think would be replace the existing "RSSI" negative values for each aircraft in a converted table header column called "Sig" at scale 0-100 and some sort of graphical meter indicating combined msgs received by the station as a whole per second near the utc clock.

Edit, Bramj already has mode s messages per sec in his dump1090 display above (how is this done ?) can this be replaced with graphic like the planeplotter version ?

 

[xforce30164]

Hi xforce,
The ideal flavor of my dump1090:8080 I think would be replace the existing "RSSI" negative values for each aircraft in a converted table header column called "Sig" at scale 0-100 and some sort of graphical meter indicating combined msgs received by the station as a whole per second near the utc clock.

Edit, Bramj already has mode s messages per sec in his dump1090 display above (how is this done ?) can this be replaced with graphic like the planeplotter version ?

IIRC, the number of messages/sec is displayed by default by dump1090. I'd need to look into how to make that graphwise. I havent really done those kind of things before, but it could be a nice learning experience!

 

[giacomo1989]

New DUT! (Device Under Test)

NINE (9!) legged spider antenna with 65mm legs

I've to correct the sightly distortion on the main monopole

 

[radiostationx]

IIRC, the number of messages/sec is displayed by default by dump1090.

My version doesnt, but it is the stock malcom robb version, can i apply a custom change my display to include RSSI and msgs received column in table easy without swapping versions of dump1090 ?