Small internal aerial to large external aerial.

I’m a radio ham, licenced for over thirty years with a fair bit of experience with aerials, building, using, comparing etc but tracking aircraft is new to me.

Last weekend I built a spider and hung it in my shack window. It’s about 12-13ft AGL, the window faces roughly SE-SSE. I’m on the east coast of England, not far from Clacton-on-Sea and the CLN VOR-DME.

It works really well considering the location. I get aircraft almost as far south as Paris, regular spots from 200-250nm out.

I just swapped it with a commercial aerial claiming 6.5dBd (not even that silly dBi that some manufacturers claim) which I’ve got outside on a 20ft pole which is about 7-8ft above the ground so in theory a much better aerial in a much higher location. I’m feeding it with around 10m of RG213 so just under 3dB loss at 1090MHz.

On the spider, most of my traffic was in the direction of the window, very little to the north, west or down towards London. Now I’m getting a much wider view from all directions and down to around 5000ft over London. Great.

But my distance has dropped massively. I’m seeing very little over 100nm away, even from the same directions where I was getting lots of traffic before. The aerial is higher, allegedly better and has a clear view in the same direction.

Now I know collinears have a lower angle of radiation than the quarterwave and traffic that far away will naturally be higher so is it likely to be something as simple as that? It would explain why I’m seeing aircraft closer to the ground. But I’m not sure I want that. Perhaps I’d be better off with something like a single 5/8 wave at 27ft AGL than what I suspect is a multi-element collinear.

What’s the general feeling about this among those more experienced in ADS-B reception than myself?

You can clearly see here where I swapped the aerial over today. The two blips with the max range going back up are where I plugged the indoors aerial back in for a few seconds.

Unfortunately there was no mention of what the rest of your equipment is, but to me it looks like the ~2.83db line loss is getting the best of your signal. That’s pretty much half your signal as you know. I would suggest an LNA at the aerial to give you the needed boost, even if you are running an FA dongle or another LNA downstream for a test. I think it would wake your system up right away.

I’ve got the uputronics combined preamp/filter at the shack end of the coax. My receiver is an old NooElec R820T dongle but I’ve got the latest rtl-sdr dongle on the way although I don’t know how much difference that’ll make.

You’re right of course, 3dB being around 50% loss but in real terms that’s half an s-point which isn’t a huge amount although if the signals are weak enough to start then I guess it could be dropping them below the noise.

I can look into mounting the preamp at the top of the mast. That in itself isn’t too tricky, I can knock up an SMA to n-type tail so it’s literally right at the bottom of the aerial itself and then have the feeder going straight into it. The difficult part is making sure it’s weatherproof and getting power to it :slight_smile:

Shame the weather is so foul at the moment. I got soaked this afternoon putting that aerial up.

It looks like they now sell mast head mount weather resistant boxes for the amps.
I ordered a couple for my setup. … uct_id=101

Distant signals really are right on the edge of the receive sensitivity so that extra 3dB hurts a lot. IIRC it’s not unusual for signals to be -90dBm or less at the receiving antenna. A bit of amplification helps a lot, then you usually end up limited by your radio horizon.

The radiation pattern isn’t a big deal, the aircraft where the antenna gain matters are the ones that are within a few degrees of the horizon anyway (they are at higher altitudes, sure, but the curvature of the earth means they’re still close to the horizon)

In bias t mode, You can power the habamp unit at the aerial or very close to it (loft) by squirting 5-15 vdc up the coaxial cable.

There is a jumper switch on the pcb & voltage regulator to enable this option on some models.
You have to open up the enclosure to find it.
If yours doesnt have this facility the maufacturer can retro fit this at a small cost I think.

Heres a photo of my habamp pcb, note the bias t switch at top right.

I didn’t even consider that - The 820T2 RTL-SDR dongle I have on the way can chuck out 4.5v by bias tee so that’s that sorted, thanks for the heads up. The enclosure looks interesting, I’ll grab one of those as well.

Looking at this, I’ve probably got nearer 15m of cable so around 4.2dB loss which is more significant, that’s quite a high percentage of signal lost in the coax so I’m inclined to agree that this is the reason.

I’m seeing a lot of nearly stations, the density is much higher than before, I’ve been hovering at over 100 aircraft on the map for most of the afternoon, the highest I saw was 120 earlier.

Thanks for the help, folks.


I have an RTL dongle powering(via bias-t) a HAB/UPUTRONICS 978Mhz amp for UAT 978.
I just had to solder one connection on the Dongle.

I also have the ceramic and non-ceramic 1090Mhz amps, however, one is power by bias-t from an airspt and the other is power by USB.

They all work well. Hoping the get the 978Mhz setup outside(currently in the attic) and higher in the next few weeks.

My antennas are less than 100 ft AMSL. I occasionally (50-200 times a day) get aircraft out to 200NM. I probably get a lot of splatter from the local radars and tourist helos.

Looks great!
I whish I could also install an outdoor aerial, high and clear of obstructions, but the managemant of the building I live in dont allow any resident to install an aerial on roof top or outer wall. I have therefore to be content with strictly indoor installation.

I however have advantage that me and my arials live in an apartment which is 60 feet above road level, and large glass windows, which give me a good range, as shown in the VRS range plot below.

This is my first aerial (2013). I did not have a Raspberry Pi at that time, so the DVB-T was pluged into my Windows desktop in another room, and the aerial+satellite amplifier were connected to dvb-t with 45 feet of RG6 coax.

Uputronics have a conversion option to enable bias tee which is factory fit only so you will have to return your unit to them for reworks.
Very reasonable price though. … uct_id=100

Hey all Uputronics here. Just for clarity the switch is only there for boards fitted with the optional LDO kit. It lets you switch between bias tee and the PCB header for power. If you have a LNA without one a diode switches the power between bias tee / USB automatically.



Thanks for the input, Anthony. That means I don’t actually need to do anything with the preamp, it’ll just work when I use bias-tee to feed power to it.

You clearly do have a decent takeoff, that’s a good plot there. What are you using to give you the jagged lines around the plot, which I guess is the outer limits of aircraft actually heard in those directions please? I also very much like the distance marked on the rings.

I had a brainwave this morning. A good few years ago, I had a VHF amateur radio aerial on that pole with some very good coax feeding it. I had a near lightning strike five years ago and I took that particular aerial down and replaced it with something else but I never took the coax out. I checked before I went to work and the shack end was still there so I’ve just got home and have been into the loft to check. There it was, a length of Ecoflex 15 all coiled up.

So of course it would have been rude not to use it. I’ve fed it back out to the roof and connected it to the aerial.

What a massive difference! I’m back to how I was before although with the benefit of many more aircraft showing. The problem obviously was the well over 4dB of loss on the RG213 as I calculate around 1db loss with the Ecoflex 15. I am still going to put the preamp at the masthead but this is what I got straight after swapping the coax over.

That’s 176 aircraft there and it’s not even the busy time of day.

Here is some info on enabling bias-T on rtl-sdr dongles

Thanks Jon. In the latest version of the rtl-sdr dongle, that’s not even needed. It’s just a software command.

Feature 2: Software Selectable Bias Tee

V.1. and V.2. of our dongles included a bias tee which could manually be enabled by opening the case and soldering two pads on the PCB together. V.3. introduces a bias tee which can be enabled entirely in software.

WARNING: Before using the bias tee please ensure that you understand that you should not use this option when the dongle is connected to a short circuited antenna. Short circuiting the bias tee could damage it permanently. Only use it while connected to an actual powered device, like an LNA, active antenna or the Spyverter.

In Linux download the source from git, compile it the same way you do the regular RTL-SDR drivers, and then run ./rtl_biast -b 1 to turn the bias tee on and ./rtl_biast -b 0 to turn the bias tee off. The procedure is:

git clone GitHub - rtlsdrblog/rtl-sdr: RTL-SDR experimental branch
cd rtl-sdr
mkdir build
cd build
cmake …
cd src
./rtl_biast -b 1

Taken from here.

I’ve got a V.3. on the way, so that’s handy :slight_smile:

I use VRS (Virtual Radar Server) software, installed on my Windows Desktop, which is on the same LAN as the Pi.

I configured VRS to use two “receivers” as follows:
(1) IP of Pi, port 30005, format Beast (for adsb data from aerial).
(2) IP of Pi port 30105, format Beast (for FA MLAT feed back data).

I also configured in VRS a “merge feed” of the two above noted receivers for combined display.

I also very much like the distance marked on the rings.

Distances of rings are set in settings of VRS, but these are NOT displayed on the map. After taking screenshot, I wrote these distances using MS Paint.

Thank you.

I’ve been playing around with VRS now on an old XP VM I have here (I don’t own any physical Windows PCs) and I can get it to connect to port 30105 perfectly OK but it fails on port 30005 so that means all I can see is the MLAT traffic. It’s not the end of the world, it’s just something else to play around with.

Also I’ve got the bias tee working up the cable but I have to start it manually each time I reboot the Rpi. Also, dump1090 can’t be running or the bias tee enable fails with an error because the dongle is in use.

I need to work out how to start it prior to dump1090 on boot.

Set up a systemd service with Before=dump1090-fa.service?

Actually what I did was edit /etc/init.d/dump1090-mutability and added a line to run the startup command.

Right towards the top

# Added by Keith to start bias tee
/home/pi/rtl-sdr/build/src/rtl_biast -b 1

That seemed to do the trick.

I also use a v.3 from, and tested quite a few options in terms of antennas and receiver.
Try to get rid of coax cable altogether by mounting a Pi and dongle at the antenna straight away. B&Q weatherproof boxes do the job perfectly, or electrician’s supplies, the latter is cheaper with wider selection.
I’m currently testing a Pro Stick Plus against a v.3 sharing a common half-wave, and getting 2x more position reports with it. Might want to look into a Plus, gave more range and reports in every comparative test so far.
Just my .02p


Here’s a plot taken a few minutes ago, Jetvision A3 antenna at 50’, 15’ LMR400, FA filter, 15dB LNA into a MiniCircuits splitter, feeding a blue Nooelec 820T2 on a RPi B. Rings are 50nm. I just managed to catch a couple of pings from a 747 out well over 550nm. It showed up three or four times then went away.

Long range plot is direct from Dump1090, the darker shorter range is MLAT from Piaware.