best antenna

[abcd567]

Gregg,
My setup is a 24 segment coax collinear, satellite amp, 10m RG6, power inserter, dongle and Raspberry Pi. All F connectors except the fly lead from the dongle which is MCX to coax then coax to F adaptor.

I do have a few T connectors. I've got a spare Pi which I'm just about to press into service as a "test bed".
I'll wait for ab cd to come back before dismantling anything

DO NOT dismantle anything from your gold mine .

As far as I know, RPi / dump1090 can combine two DVB-T dongle inputs. Add 2nd INDEPENDENT system to your RPi:
2nd antenna (franklin or cantenna) + 2nd coax + 2nd amplifier + 2nd power inserter + 2nd DVB-T dongle.

[abcd567]

I am not very sure if dump1090 can merge the feed from 2 dvb-t dongles. This needs input from members who know this for sure.

Directly paralleling two antennas with a Tee is tricky & requires phasing harness.

Using a Satellite TV Splitter 950 -2450 Mhz is another option to combime two antennas. The attenuation to flow of signals between two ports marked TV is very high so two antennas connected to TV or out ports will not suck each other's signal, an all signal flows to in or lnb port.

However the attenuation in the flow from in to out, or lnb to tv, is 3 to 5 dB. If an amlifier is present, it can overcome this loss. This option is worth trying, as it involves cutting the feed cable to CoCo some distance from coco, and inserting the splitter, and connect 2nd antenna to splitter also. Amplifier will have to be located between splitter & power inserter, so that it amplifies the combined signal of two antennas.

If this fails, simply remove splitter, and join CoCo with feed cable directly using a barrel connector, and system will be like before.

antenna combining.png

[SpaxmoidJAm]

Get a second dongle, build a directional yargi (or use your original coco) and aim it at the airfield.

You can then run TWO instances of dump1090 and mix the two streams using modesmixer2 or similar best of both worlds then. Much simpler than matching stuff and no extra losses from combiners.

There is a command line switch in dump1090 that you can use to select which dongle you want to use, have a script for each instance and have dump1090's output on ports 30001 & 30002 (not what you are currently connecting too) then have modesmixer2 combine the two and output on 30003 (or the port your feeder connects to)

[trigger]

Get a second dongle, build a directional yargi (or use your original coco) and aim it at the airfield.

You can then run TWO instances of dump1090 and mix the two streams using modesmixer2 or similar best of both worlds then. Much simpler than matching stuff and no extra losses from combiners.

There is a command line switch in dump1090 that you can use to select which dongle you want to use, have a script for each and have dump 1090 output on a port say 30001 & 30002 (not what you are currently connecting too) then have modesmixer2 combine the two and output on 30003 (or the port your feeder connects to)

Looks like a plan ! Thanks. I've got a spare Pi, dongle etc. to give it a try before going onto my production system.

[trigger]

I am not very sure if dump1090 can merge the feed from 2 dvb-t dongles. This needs input from members who know this for sure.

Directly paralleling two antennas with a Tee is tricky & requires phasing harness.

Using a Satellite TV Splitter 950 -2450 Mhz is another option to combime two antennas. The attenuation to flow of signals between two ports marked TV is very high so two antennas connected to TV or out ports will not suck each other's signal, an all signal flows to in or lnb port.

However the attenuation in the flow from in to out, or lnb to tv, is 3 to 5 dB. If an amlifier is present, it can overcome this loss. This option is worth trying, as it involves cutting the feed cable to CoCo some distance from coco, and inserting the splitter, and connect 2nd antenna to splitter also. Amplifier will have to be located between splitter & power inserter, so that it amplifies the combined signal of two antennas.

If this fails, simply remove splitter, and join CoCo with feed cable directly using a barrel connector, and system will be like before.

[ATTACH=CONFIG]5611[/ATTACH]

Hi ab cd, I tried this but the loss across the splitter was too great. My plane count went down from 120 to about 80 as soon as I put the splitter in. One thing I did notice is that my amp was cold. I'm feeding it with 12V. Maybe I'll see if I can find a slightly higher voltage "wall wart". Do you think the higher voltage will help?

[MatthewE]

Is it possible to feed two different PI's to the same Streamer or, like said would I have to get a splitter for the antennas OR use a seperate dongle, as I have two. - Not sure of the best way to do this, was quite interested in using that dongle to pick up ATC frequencies on a different PI; can't say I've tried it but was willing to have a go and stream it.

[abcd567]

Hi ab cd, I tried this but the loss across the splitter was too great. My plane count went down from 120 to about 80 as soon as I put the splitter in. One thing I did notice is that my amp was cold. I'm feeding it with 12V. Maybe I'll see if I can find a slightly higher voltage "wall wart". Do you think the higher voltage will help?

It may help to some extent. How much, depends on the individual amplifier's design.

If you alresdy have a 15v or 18v adaptor, give it a try. These amps are rated to work from 13 to 18 volts dc, but have some tolerance as well.

You may also try 2 amplifiers, both on feed line, one near splitter, other near receiver or midway. The 2nd amplifier will more than cover for insertion loss of splitter.

[SpaxmoidJAm]

@trigger you could use both on the same pi, or (a quick Google tells me) that the modesmixer2 will take in a stream from another IP address ie different unit.

that would involve you changing your current set-up or probably a better solution you disable the fr24 feed on you current pi. Set up the second pi combine the dump1090's on that one then feed to fr24 on the second. That way if it goes horribly wrong you only have to enable the fr24 feeder again on original box.

if your going to use the same pi i would create new services for each instance ie dump1090One and dump1090Two fr24feedTwo so again easy to revert back.

[trigger]

@SpaxmoidJAm
I'm reluctant to alter my current Pi's software as it is working just great. I'll play with the second Pi and try a few different bits of software. Modesdeco2 and Modesmixer2 are on the list to try as well as the mutability version of dump1090.
@abcd
OK, I'll try a second amp.

[abcd567]

.......I'm reluctant to alter my current Pi's software as it is working just great...........

I have not tried ModeSMixer2, but did try ModeSDeco2, took screenshot of Map, then switched to dump1090-mutability and again took a screenshot of Map. The number of planes in ModeSDeco2 map are 1/2 the number of planes in dump1090-mutability map (time difference between two screenshots is about 6 minutes only).

I have stopped using ModeSDeco2 till I find out how to fix this issue. I tried to fix it by setting ModeSDeco2's argument --gain 49.6 (i.e. maximum), but no improvement.

modesdeco vs dump mut 2.PNG . modesdeco vs dump mut 1.PNG

[abcd567]

SPLITTER INSERTION LOSS

Splitter-Insertion-Loss.jpg

[abcd567]

Hey! Our friend trigger's successful 24-element coaxial collinear (Coco) is showing it's strength. His feed is #25 now, going up (↑ besides his name). Waiting for him to reach #1!

Screenshot_2015-02-06-15-32-44.png

[MatthewE]

Hey! Our friend trigger's successful 24-element coaxial collinear (Coco) is showing it's strength. His feed is #25 now, going up (↑ besides his name). Waiting for him to reach #1!

[ATTACH=CONFIG]5641[/ATTACH]

Congrats, should have a coco today so fingers crossed I'm able to further increase results

[trigger]

Congrats, should have a coco today so fingers crossed I'm able to further increase results

Thanks. Good luck with the CoCo. Take care with the build.

[Scroggie]

Hi again ABCD567

After a couple of weeks use I can see that my installation is about as good as I could expect. I may still replace my coax colinear (This design but joints pushed together rather than soldered :- http://www.rason.org/Projects/collant/collant.htm ) with one that has soldered joints and uses lower loss coax (just using cheap Chinese RG59 at the moment)
Using an on-line calculator to work out what distance an aircraft at a particular height goes behind the horizon from my location (which is 790 feet above see level) I find that at 37,000 feet it works out as 236nm and even at 41,000 feet it is only 246nm. As the maximum I'm getting regularly 248 - 252nm I can't see that making any changes is going to result in much improvement.
What brings me back here are your polar plots which appear to be on a Virtual Radar map like the one I'm using. I've gone through the settings page and have set the the range rings to various settings but I'm curious as to how you have got the 24 hour plot. Is it hidden in the settings of Virtual Radar somewhere or are you using some other software?