Wideband active antenna

Thread Starter

sirch2

Joined Jan 21, 2013
1,037
I just bought a Software Defined Radio USB dongle and an up converter for HF frequencies. It is a nice toy but the tiny whip antenna that comes with it is obviously limited at lower frequencies so I was thinking of making an active antenna. There are lots of active antenna designs on the web but my question is am I better off making separate HF and VHF antennas or is a wideband one, such as this going to work?

I know next to nothing about active antennas so any thoughts and pointers appreciated

Here's the wideband circuit diagram to save you clicking through

 

DickCappels

Joined Aug 21, 2008
10,153
As far as antenna performance, you will be better off with separate HF and VHF antennas.

For most cases, in the real world, you would be best off if you use an antenna designed for the part of the spectrum you intend to use. The designs above don't really increase the bandwidth of the antenna, they only add gain (and noise, which might be acceptable in some cases).

A given antenna will have a frequency response, which will usually have a single peak, and that will give you the best signal-to-noise. A fractal antenna (Google is your friend) can give you a wider bandwidth but I don't think they have been proven in the frequency range typically covered by SDR's, but you might want to investigate them.
 
I like an active whip for frequencies below about 40 MHz and a discone for the higher frequencies. The big problem with active antennas is intermodulation distortion (IMD). Large signals will mix in the amplifier producing false signals all over the band. It's also a good idea to limit the active whip to 30 MHz or so to keep out FM signals unless the circuit has unusually good IMD performance. Keep the active antenna pretty short, maybe just 3 feet, and I'd use a circuit with little or no gain. I've been working on just such an active antenna for the last couple of weekends. See Fig 8 at http://www.techlib.com/electronics/antennas.html (near the bottom of the page) and read the red text. Similar circuits on the web rely on "pre-distortion" in the FET canceling distortion in the output stage to get numbers like -80 dBc (for a 1MHz product from 3 and 4 MHz test signals mixing in the amplifier). One can get quite a bit of cancellation of the non-linearity by playing with the currents but it tends to be temperature-sensitive and one needs a two-tone test jig to make the adjustments. After this weekend I have a new circuit that has an intrinsic IMD of better than -80 dBc that "tweaks" down to -100 dBc, really dropping into my measurement noise floor. The intrinsically low distortion means the values aren't as critical and the circuit doesn't vary over temperature; freeze spray has little effect. -80 dBc is plenty good so the cancellation is just gravy. I think it's sufficiently good to let it handle FM stations but I haven't had time to stick it on the roof for the "real" test. I may add a low-pass filter to block FM. The circuit I already have on the web page is fine; I use it all the time and I've not noticed any significant IMD even at the less impressive -65 dBc level. But I'll try to publish the newer circuit in a week or two. You may still get IMD from the receiver itself; I can't imagine they have IMD as low as -80 dBc but I don't really know. Someone needs to design a "software-defined filter" for the front-end of those things!
 
Top