abcd - my findings were that even with my supershort cable (2 feet) an amplifier improved things very much (5-10% more aircrafts and 40% more messages) - was the same with all antennas i’ve tested.
I used amplifier when I did not have RPi, and my DVB-T was inserted in my Windows Desktop. The cable run between antenna (near large window) and Deaktop (in a room on opposite side of apartment) was 50 ft /15 meters.
I removed amplifier when last year I purchased & installed RPis just below the antenna location. The reason is that the AGC of amplifier increases gain when a poor (low gain) antenna is connected, and reduces gain when a good (high gain) antenna is connected. This masks the real difference between two antennas, and makes antenna comparison results inaccurate.
I did try amplifier with short cable to RPi and my improvement was similar, but less than as yours (attic vs large window). However since I use my set up primarily to test & improve DIY antennas, and data feeding is secondary to me, I prefer to keep my setup without amplifier.
UPDATE on FA Antenna & N-to-F adapter:
Two hours ago, I replaced direct connection of RG6 to antenna by N-to-F adapter. The performance remained same, proving that problem is NOT with N-to-F adaptor. Please see the 24 graph below.
Image 1 of 2:The 24 hour performance graph.
–First 5 hrs on left are for Cantenna.
–Middle 17 hrs are for FA Antenna with RG6 coax connected directly without any connector or adapter.
–Last 2 hrs on right are for FA Antenna with N-to-F adapter.
image 2 of 2:The RG6 feed cable now connected through N-to-F Adapter.
abcd - again - i’d prooved it first against small standard antenna. fa antenna should be double as good - if not (and cable/adapter are ok) it is defect. now you just see that cantenna is better - and possibly is a superperformer.
in addition message rates are the better first and much quicker indicator than aircrafts seen.
btw. also try some more metal at the foot of fa antenna - all my antennas are mounted on aluminum photo stands.
Makes me wonder if one of the soldered connections in the antenna broke during shipping and you’re only getting one of the elements. Has anyone ever done a teardown of the FA antenna to see how they connected the pieces of coax inside?
That is what I am suspecting: some soldered connection broke down during shipment from manufacturer to Amazon or Amazon to me. I cannot open the antenna to see, as this will void warranty & I will loose the chance of getting a replacement. Unfortunately, the FA antenna does not have an interna short (like Cocos have at their top), so there is no way to do a continuity check by multimeter to find an open circuit indicating broken connection.
btw. also try some more metal at the foot of fa antenna - all my antennas are mounted on aluminum photo stands.
Ok, I will give this a try.
May be adding a groundplane will improve FA Antenna.
By the way, I have checked continuity between metallic collar at bottom and the threaded body of N-female connector. There is no electrical connection between the metal collar & antenna.
.
my intention regarding the messages graph was: when i test my antennas (e.g. flightaware against with dongle delivered) i just have to run first one about 3 minutes → then quick change an run the second one for 3 minutes. this way within 10 minutes i see differences bigger than 5-10%. so it’s very easy to check whether the fa antenna performs about the way it should - minimum double as good as the tiny dongle one. and there is no need to measure over hours to see these big deltas.
of course especially in-house - placing antennas at the exact same place is mandatory. and the problem that remains is that you don’t have a second standard 50ohm n-plug to sma cable with sma to mcx pigtail to exclude errors in cabling.
Since the DVB-T dongle is intended for TV, it has an input impedance of 75 Ohms. However, the mismatch loss when using 50 Ohm cabling and/or antenna, will be very minimal at around 0.177 dB. So dont worry, be happy, smile and mix the 75 ohm with 50 ohm system
How I arrived at figure of 0.177 dB?
When 50 ohm mixed with 75 ohm,
Reflection Coefficient Γ = (75-50)/(75+50) = 0.2
SWR = (1+Γ)/(1-Γ) = (1+0.2)/(1-0.2) = 1.2 / 0.8 = 1.5
Mismatch Loss in dB = -10 log (1 - Γ²) = -10 log (1 - 0.2²) = -10 log 0.96 = 0.177 dB
Thanks for the suggestions. This is how we try to help each other and exchange ideas. Since my opinion differred from yours, I thought it appropriate to let you know. That does not mean I dont value your suggestions & ideas, which are always welcomed.
This is great. It has advantage of side-by-side comparison and reliable results can be obtained by short duration trial runs. Was it easy to bring performance of two systems within 1% of each other? I will try to setup my two systems like you have done, when I find time.
Currently I am using a different method which requires prolonged trial runs, often 24 hrs. I use only one system, and simply change the antenna, keeping everything else same.
Your coax connector looks like the twist on type. Have you checked the continuity from one end of the cable to the other with the adapter installed? (Should be close to 0). Just wondering if the RG6 inner conductor is making a good connection into the adapter or if it is high resistance.
Out of curiosity - what gain you are using and how your signal level graph looks like? Judging by the message rate graph your are getting more over 3db messages - any possibility you overloading the dongle with FA antenna?
Just a random comment and may not apply here, when I started using the FA antenna, I was using automatic gain control and then tried Max gain. It was only until I manually lowered the gain to 43.4 out of 49.6 (the max my USB said it would support) that my average jumped about 30%.
I used the formula for the data in /var/log/dump1090-mutability.log:
“messages with signal power above -3dBFS * 100 / total usable messages”
Keeping that calculated value between 3 and 4 seems to work better for my site.
Thanks for the tip. At this hour of night, the air traffic in my area starts falling rapidly to near zero, so this not a suitable time for experimenting. I will play with the gain tomorrow and see what happens.
Amazon.ca did a lousy job of packaging it. It just sat diagonally in a box touching the box corners at both ends with the corners of the box ripped, either to fit the antenna or damaged during shipment. I hope the antenna is ok. It looks ok and dosn’t rattle when I shake it. LOL.
I can’t test the thing until my connectors come from China via a wind powered raft. Based on abcd’s experience, I’m kinda concerned.
