FlightAware Discussions

Which is better? Longer antenna cable or longer USB cable

Not withstanding weather, waterproofing,cost etc.

Which tends to be better? Using a longer antenna with the dongle plugged directly into the Pi/computer or using an USB extension cable?

I assume having the dongle plugged directly into the Pi/computer removes one failure point. But, it seems it might be easier to run CAT5/6 cable rather then sourcing a well made antenna cable.

  1. The USB cable has voltage drop that might affect the dongle and, not being shielded, it becomes an antenna - data can be corrupted by noise
  2. The receiver performance is influenced by the thermal noise. Keeping the receiver in a conditioned space yields better results.
  3. For coax there are available lightning protectors (gas discharge type). Those are not avail for USB or CAT5.

IMO antenna followed by filtered LNA and coax is the best way to transmit the signal at long distance. My case is 45 meters away, with cheap RG-6 cable.

If you’re Raspi will have access to wifi, then both should be short - make your power cable long and use a $1 DC-DC converter to provide a stable 5.1V into the Raspi from a 12V source. If you’re using just antenna+cable+dongle (no LNA), then you want that cable as short as possible to maximize your receiver range. Long USB cables are also not a good idea for the reasons mentioned above.

There’s a separate thread on DC-DC converters. 12V AC adapters and power extension cables are common / cheap because of webcams.


While longer wire is never great for anything, the signal lost from an analog signal (radio) vs digital (USB) would make me want to have a long USB cable (like 50ft long in my case) if I have to have a long wire somewhere. The USB is for the most part is either going to work or not work. The lost to analog will be harder to see and correct . A fair number of active USB cables make the voltage drop moot.

Instead of buying an active USB cable one might just as easily buy the rtl-sdr LNA.
Then you can use RG6 or other relatively cheap coax (i still wouldn’t want more than 10 dB attenuation for the length of the cable).

But i think in this case the author wasn’t talking 50 ft but rather 5 ft.
In which case i would just use 5 ft of coax and be done with it, that’s really not a problem.
Get some LMR240 or CFD200 or LMR195 and attenuation for runs up to 25 ft will be quite small.
After that using more expensive LMR400 cable still gets you acceptable levels of attenuation.

Anyway either you keep the coax quite short or you get really god quality coax or you use an LNA.
It’s that simple.

1 Like

Contrry to common belief, the attenuation of LMR-240 & LMR-200 in NOT better than attenuation of RG6.

Type Impedance
Max. attenuation, 750 MHz
(dB/100 ft)
RG6 75 5.65
LMR-200 50 9.0
LMR-240 50 6.9
LMR-400 50 3.5

SOURCE: https://en.wikipedia.org/wiki/Coaxial_cable#Standards


Always best to check the data sheet of the individual product you are intending to buy.

You have a bit additional attenuation due to 75 Ohm if i’m not mistaken, but that’s not per length of cable?
Also if you have F connectors with adapters, that will also add a little bit.

1 Like

From Page #2 of: https://catalog.belden.com/techdata/EN/1694A_techdata.pdf

Belden, RG6


1 Like

You should bear in mind that “RG-6” is not a very tight specification of cable. It says nothing about how it will perform in any particular application.

To quote this article:

Take RG-6, for example. The only thing one can say generally about RG-6 type coaxes is that they have an 18 AWG center conductor. An RG-6 might be a cheap generic CATV coax, designed strictly for economy, with a thin aluminum braid and a copper-coated steel center conductor; it might be a better-grade CATV coax, with an aluminum “quad shield” arrangement; or it might be a precision serial digital video coax like Belden 1694A, with a dense copper braid and double-foil shield, solid copper center conductor, nitrogen-injected PE foam dielectric, and extremely broad bandwidth and tight impedance tolerance. When people ask, “can I just use RG-6 for this application,” the only correct answer is a question: what do you mean by “RG-6”? High-bandwidth serial digital video and cable TV distribution may both use RG-6 type cable, but that doesn’t mean that the cable for one is necessarily suitable for the other.

In short, don’t buy cheap and nasty cable, buy decent quality from a reputable manufacturer. It might be cheaper to buy some kind of RG-6 than LMR-240 or whatever, but if you get something made to minimal cost, you might as well use a piece of wet string.


True, but this is true for LMR, as well as all other types of coax also.

The only way to find out quality of any type of coax is to go for a good manufacturer, and check manufacturer’s specs.

I feel the prejudice against RG6 is based on the fact that RG6 uses 75 ohms, F connector, suitable for Ghz frequency of TV / Satellite, rather than ham’s beloved 50 ohms SMA Voice communication standard. :wink:

Later Addition:
The DVB-T stick is designed for TV and uses 75 ohms system :slight_smile: . Why shall we use 50 ohm system (antenna + coax) for its RF input?


LMR-400 is a trademark of TImes Microwave, so unless you are buying a knock-off it should meet the stated spec. There are various clones, but they usually have a similar, but different name.

I’m not sure it’s really a prejudice, rather than it being a matter of using the right tool for the job. If you have a system that is 50 Ohm, then using 50 Ohm co-ax is the optimal way to do it. A 75 Ohm coax will work, with some varying penalty to performance depending on what quality coax you use. As with most things, there is a tradeoff between performance and cost, and it is up to the individual to decide where the balance lies.

Also, the input impedance of various rtl dongles depends on the manufacturer. The originals intended for use as TV receivers are likely 75 Ohm, but certainly the rtl-sdr.com variants are nominally 50 Ohm inputs, as are the airspys.

Likely only the ones with strange TV connectors are 75 Ohm.

I’d expect the ones with MCX or SMA connectors to be 50 Ohm.

Isn’t LMR cable manufactured by Times Microwave?

I guess if you are going to brand cheap coax with LMR you may as well brand it Times Microwave and make it completely counterfeit. Hence

That is a bit condescending. You try running a couple of KW over a transmission line that is just any old Rg6 with you- beaut type F connectors suitable for GHz and see what goes crackle, fizz, bang, corona followed by a very expensive flash bang.


1 Like

I use neither but put the Pi in the roofspace out of the weather and feed it with Ethernet and PoE.

The dongle plugs directly into the Pi so no usb cable at all and there is only about 5M of coax to the antenna.

The power over Ethernet compensates for the cable run and provides the right voltage to run the Pi.

Use PoE with 802.3ef power control.


This discussion is about receiving ADS-B signals, and power ovre coax is not more than miliwatts. Received signal’s power handling wise, RG6 coax, and even thin pigtails RG316 & RG174 are ok.

Is there anyone here in this forum who transmits 1090Mhz signal from ground?


All Generic DVB-T dongles (black or light blue) with MCX connector have 75 ohms input impedance.

Some dongles use PAL or SMA connector, but this does not change input impedance to 50 ohms, unless the manufacturer has modified the input circuitry also.

1 Like

Clones of LMR240


CNT240: https://www.commscope.com/SiteSearch/?q=cnt+240&ns=1



I’ve built 2.4GHz and 10GHz tranverters (down to 2m and 70cm respectively)
I’ve got an Andrews 100W amp for 2.1GHz retuned to 2.4GHz
(I think that’s a ‘yes’)

Unless the dongle has a matching pad (I’ve never seen one), the input trace is too narrow to be 75Ω (50Ω needs to wider still)

PCB Trace Impedance
Microstripline Impedance Calculator
Using this calculator and its default values, a 75 trace needs to be 1.5mm wide and a 50 would need to be more than 3mm

To get a sense of scale, the R820T chip is 4x4mm
The input pin is the bottom pin on the LHS
(imagine what that would look like connected to a 3mm track

Here is something nasty for everyone


Thanks - I read the datasheet.
What you’ve shown is the reference design that is then up the (dongle) manufacturers to implement as they see fit.
The data sheet says it “should be 75Ω”
and I’ve provided a photo that isn’t (and explained why)

If you look closely at that circuit diagram, you’ll notice the RF traces are bold-green. This indicated they are (should be) microstriplines.