SDR - discone antenna setup

Question

Here's a picture of my SDR receiver setup...

This entire setup lives indoors as I'm not able to put anything up on the roof right now.

My understanding of a discone antenna is that the downward diagonal elements are supposed to act as an arbitrary, floating ground plane against which the vertical elements resonate.

This setup works as it is but I'm getting problems with pickup on the metal case of the SDR. If I put my hand on the metal case (anodised Aluminium enclosure well connected to the ground plane of the receiver PCB) then the noise floor on my HDSDR software jumps up massively. What was a relatively quiet waterfall becomes a bright red mess as soon as I lay my hand on the case. That said, I cannot hear any difference in the demodulated signal (2 Meter band, AM for instance).

It seems in my mind like this pickup should be prevented because of the connection to a real earth ground (via the PC's mains power supply). There is good connductivity between the diagonal elements of the Discone and earth ground (less than 2 Ohms at DC, according to my multimeter).

The USB is about 2 metres long, the Coax is about 3 metres. All connectors are soundly fastened and have good conductivity.

Edit: The SDR device is entirely USB powered.

Questions

It is clear that the PC's USB connection is pulling the metal case down to "Earth ground". Is this detrimental to the use of a discone antenna? Is it bad to have a metal case on an SDR? Is there anything I can do to improve this setup?

Answer 1

The symptoms described are consistent with common mode currents. These are undesired currents flowing on the outside of the shield of the coax (desired currents flow on the inside of the shield and the center conductor). The result is that the exterior shield of the coax acts as part of the antenna. By touching the coax or the electrically connected SDR enclosure, your body becomes part of the antenna system altering its performance for better or worse.

Common mode currents can be suppressed by installing a choking balun on the coax as it exits the discone. Ferrite type baluns are usually the best but you can experiment with simply winding 6 to 8 turns of coax in a 6 to 8 inch diameter bundle. This is a compromise balun that works best over the range of a few ham bands. Experiment with the number of turns and the diameter to optimize your results.

These types of baluns can be effective up to 30 MHz or so. With higher frequencies, other methods must be deployed. Unfortunately, the higher frequency solutions are usually effective for a single band and its odd multiples. An example of this is to use a coax connection that is electrically an odd 1/2 wavelength long for the frequency that is being received. This tends to present a high impedance to the common mode currents. But because the length of the coax must be adjusted according to the frequency, it is only practical for a small range of frequencies. Another example of this type of frequency dependent solution is a sleeve balun.

One side effect of reducing common mode currents is that your antenna may not perform as well as it had. Sometimes the coax shield acting as part of the antenna is helpful to the overall antenna performance, despite the negative effects. Experimentation will determine what is best for your situation.