Which one is the best antenna?

[hellhound604]

like others have said, it is about the Noise figure.
a hypothetical example, say you have a length of coax with total attenuation of 10dB, and a LNA with gain of 20dB and noise figure of 2dB.

if you have the coax first, followed by the amp, total gain is 10dB and NF=12dB.
if you have the preamp first, followed by the coax, total gain is 10dB BUT NF is only 2.2dB!!!

so, a difference of 10dB by just placing the pre-amp first in the line, that means, you can pick up aircraft whose signal is 10 times weaker, than if you have the LNA after the coax.

the Noise figure of the first stage is the most important, that is why it is meaningless to have a very expensive attenuator (coax) in front of a LNA. Rather have a LNA followed by a cheaper coax, than the other way round. If you look at some of the better multistage-filtered-LNA’s, you will see the philosophy followed there too.

just Google “cascaded noise figure” to get the formulas and explanations.

The caveat is that in our scenario with amp 1st, followed by long coax/filter combinations, that your LNA must be able to handle large out-of-band signals too, that can be quite challening if you live near a cell-phone tower/airport, etc. in that case you need either an expensive LNA (or one of the bettter filtered LNA’s) followed by the standard chain, or plug your spectrum analyzer to the antenna, capture the signals, and start number crunching, and see what combination gives you the best compromise…. And even then, things can get compromised by RF-noise climbing into the shields of the coax, or very weird issues close to high-power transmitters, like cross-modulation caused by a number of different reasons, but that is a totally different field.

[jaymot]

Gotcha. After reading the page that butterfly2 linked to I ordered an RTL-SDR triple-filtered LNA and an NESDR SMArTee dongle to use with it as all the RTL-SDR v3 dongles on Lazada were counterfeits. I’ll use my FA orange dongle for airband VHF or HF radio or something. (And here I planned not to spend any more money this month on ADS-B or radio stuff. sigh Oh, well. Wednesday’s my birthday so I’ll consider them to be b-day presents to myself.)

[alleyoopie]

Don’t hesitate to post a picture of the box once you have it complete with Pi, receiver and power supply.

[PA1NT]

Will do, but it’s nothing special; Pi4, Airspy mini, Bias-T.

This morning was a really nice day, so I decided to get on the roof to replace some aerials.
I made a stupid mistake when connecting the SMA connector to the LNA; I was on the roof when I applied too much force and felt the connector of the LNA snapping from the PCB

Luckily, I had another LNA in the attic, so all it took was some climbing of the roof, go up to the attic, go down and climb up again.

Anyhoo, here are some pics:

Before:

After:

I replaced the no brand Discone (which was crap) for a new one and the Vinnant COL1090-5H-N for a Vinnant COL1090/7.5-P.
I added a junction box with a Sysmocom 1090 MHz cavity filter and the RTL-SDR Triple Filtered LNA

It wasn’t an easy job to rise the mast because most of the weight was on the top of the pole (mast is 6m75/~22ft, the lower guy wires are at 2m50/~8’2"). But when I took a look at Tar1090 I was very pleased to see 300 planes and a 2250 messages/sec.

The graphs:

[alleyoopie]

Very nice and looking at your hourly stats I can already see a noticeable improvement for every hour since the changes. Now you’ll have to wait a long 30 days to see where you will end up in the stats eventually.

[PA1NT]

There is a noticable increase in performance, gradually climbing the steps of the FA Stats, currently at #17 of the Dutch feeders…

[eepete]

I tried a few of the cheap antennas, then tried a slightly more expensive one, the XRDS-RF from Amazon (US). Return loss sweeps were good. Just tested indoors, but it was a significant improvement over a quarter wave.
Feed line was 10 feet of “LMR240 equivalent” (not really, but close enough) with N at one and and SMA at the other. Return loss was done with a short piece of real coax (RG214?) about 18" long. So that might have made things look a little better, but still it tested and performed well.
Have seen a few sweeps in this thread, though I toss out another. Need to change to 5dB/div on these measurements next time, the instrument is reading OK but it would be nice as a human to see a little better view.
I also swept the Flight Aware 1090 BPF, it also was very good.

[Dutchyb]

Wow, you are having an equipment to run tests. Impressive

[eepete]

Thanks Dutchyb. Yes, it’s nice having the right stuff to measure all the pieces…
Here’s the sweep on the FlightAware 1090 BPF. It’s clearly a discrete part device which is a win because that’s more rugged than the SAW filter in the FlightAware blue dongle.
Markers 2 and 3 are the bounds of the 902 to 928 MHz ISM band in the US, i’ve got other stuff going on there.
There are not a lot of SAW filters out there, here’s a URL to the only one Mouser sells:
https://www.mouser.com/datasheet/2/1100/sf2321d-1917641.pdf
Note the max DC voltage of 3V and max power of 20 dBm (.1 watts). On the plus side, note how tight the bandpass is, typical of SAW filters. And there’s your trade-off. A SAW filter directly after your antenna probably isn’t rugged enough, but it’s great to have as part of the receiver.
This is also why the RF signal chain is so important. The exact order of the antenna, lightning (gas tube) protection, BPF, LNA and then the receiver (which can also have its own BPF and LNA) is important and installation specific with no one size fits all.
For me, the trade-off of loosing 1.5 to 2 dB in order to get rid of a lot of out of band signals is worth it, especially with SDRs typically being a direct conversion receiver.

[Dutchyb]

You are way out of my league.

I hardly understand what you show, sorry. Trying to learn, but this is something I am not familiar with

[LawrenceHill]

@eepete

Here’s the sweep on the FlightAware 1090 BPF

The narrower, Dark blue filter I presume
That looks pretty close to what I measured.

Here is a comparison between the light and dark blue filters.

[eepete]

Really like the comparisons, I’ll have to remember that “multiple trace trick” going forward. Thanks !
Here’s my scan of the dual-frequency filter:

Looks like I had a little extra “kick” at the top end, but very similar looking. Y axis scale is all messed up, This was a 1st scan, and when I saw the performance in the 915 MHz ISM band, I knew the filter wouldn’t work for me.
Nice to see Siglent and Rigol agreeing…

[eepete]

Keep trying to learn! It takes time but it’s worth it. To get you going on all this: Google & learn about Band Pass Filters, abbreviated BPF. The are designed to “pass” radio energy over a range of frequencies. Outside of that range of frequencies, they attenuate the energy. The curves you are seeing are just a way to measure what frequencies are passed and which are attenuated.
The value of a BPF is a lot of radios receivers don’t like to see lots of high energy (or, as it’s said “High Signal Strength”) signals even if they are far removed from what frequency the receiver is set to. If you are trying to receive at 1090 MHz, but there is a signal that is 1000 times stronger at 850 MHz, the receiver is “swamped” (both a people and technical term) and it can’t receive the signal you want. It’s like trying to whisper to someone when you’re standing next to a jet engine.
Adding a BPF reduces the strong signal so that the receiver can work.
A SAW filter is a type of BPF. It is a small component, just like a small chip/integrated circuit.

The other general piece of equipment you can use is an amplifier. As the name implies, this amplifies, or provides Gain, the any signal at the input. Amplifiers tend to have a range of frequencies they work at. An amplifier that can work from 1 MHz to 5 GHz would be very expensive, but one that is designed primarily to amplify 1090 MHz is not expensive. As such, sometimes you will see a frequency range (or Bandwidth) for an amplifier.
For RF there are two types of amplifiers. One that is optimized for Gain (what most people would reasonably call an amplifier) and see that are optimized for Low Noise. These are called Low Noise Amplifiers or LNAs.

A Software Defined Radio or SDR is just a radio that can be tuned by a microprocessor instead of a knob someone turns. There are often other characteristic of a SDR that can be defined by the microprocessor that optimize the radio for receiving various types of signals. Before SDRs, that would require building a different circuit out of electronic components. With a SDR, changes can be made, well, by Software !

So google around and learn more about BPFs, amplifiers and LNAs given this background. If this was too basic, I apologize. I applaud your desire to learn more!

• eepete

[abcd567]

Flightaware 1090 Mhz Antenna

CLICK ON IMAGE TO SEE LARGER SIZE

Source:
https://www.reddit.com/r/ADSB/comments/ljg4uy/insides_of_the_adsbexchange_9781090_and/

[abcd567]

AdsbExchange 978/1090 Mhz Antenna

CLICK ON IMAGE TO SEE LARGER SIZE

Source:
https://www.reddit.com/r/ADSB/comments/ljg4uy/insides_of_the_adsbexchange_9781090_and/

[abcd567]

JetVision 1090 Mhz PCB Antenna

CLICK ON IMAGE TO SEE LARGER SIZE

[abcd567]

Active Antenna designed by F5ANN (Francois)

JetVision Active Diapason Antenna uses this design

CLICK ON IMAGE TO SEE LARGER SIZE

[davidinjp]

@abcd567

JetVision Active Diapason Antenna uses this design

In my tests over the summer, this little 210mm 2.5 dBi + LNA antenna noticeably outperforms the 1448mm 9 dBi DPD Productions 1090 MHz antenna and the 660mm 5.5 dBi FA 1090 antenna with Uputronics or RTL-SDR filtered LNAs at the receiver - not at the antenna.

This is for a challenged station basically surrounded by tall pine tree forest. Aircraft count and positions and even max range improved with this little guy! The large DPD antenna came in 2nd place with better avg max range and positions than the FA antenna.

In terms of price, the Active Diapason costs about 10% more than the DPD (for US customers) and performance is about 10 to 15% improved for my weird site. But the FA antenna clearly takes the crown for price to performance winner, by far.

[abcd567]

 

(1) November 2014:
Antenna design posted by F5ANN (Francois) in Planefinder forum
Very small PCB active antenna 210 mm high

 

(2) JetVison’s “Active Diapason” Antenna (1090 MHz)
This antenna uses design by F5ANN (Francois)

 

(3) Two years later…(November 2016):
F5ANN HAS PASSED AWAY

 

 

[jaymot]

@davidinjp

This is for a challenged station basically surrounded by tall pine tree forest. Aircraft count and positions and even max range improved with this little guy!

Probably because its radiation pattern is less flat horizontally vs. the FA antenna (hence the lower gain) so it’s able to see at a higher angle over the treetops that would block a higher gain antenna with a wider and flatter radiation pattern. Every station and location is different, so some antennas work better at a particular location than others given the local conditions. You might get similar results with a “spider” (1/4 wave ground plane) with one of your LNAs near it. (I say “near” because I’ve been told in another thread that said LNAs aren’t water-tight and need to be placed inside an enclosure of some sort to protect them from rain and snow, which requires a bit of coax to be run between the antenna and the LNA in its enclosure.)

In other words, it has higher gain at the precise angle that you need for your particular location vs. a 5.5dB gain antenna that would be ideal in unforested flatlands or praries.