Introduction

This page addresses issues and helpful hints involving the use of the varions SDR programs (e.g., Rocky, Winrad, etc.) The focus here is on the RX side of the softrock.

Problem: "Mirror" Images of the Signal (TX: Pete M0FMT)

Common Causes

In most cases, mirror images in the display on the PC are the result of the PC's SDR software not seeing two, independent audio signals from the board: I and Q, via the Left and Right stereo inputs. They MUST be independent, identical in all respects except the phase, which MUST be 90 degrees apart (the "I" signal preceding the "Q" signal by 90 degrees).

Identification of the Problem

  • Apply a signal generator output to the RX board and sweep it across RX frequency range with the display set such that the Xtal frequency is in the middle of the pamoramic display.

  • Ok

    If the RX is operating correctly, you will see a peak run from left to right or visa versa right across the display with out changing direction.

  • Not OK

    If however two peaks run to the centre (Xtal) frequency and then reverse direction this means you do not have IQ signals going to your sound card for correct operation of your SDR software irrespective of the which package you are using.

Possible Sources of the Problem

  • Software Settings

    Some Software packages like WinRad allow you to combine L and R inputs, but this is not the same effect as descibed above and will not serve to over come your issue.

  • Stereo Inputs

    A simple way to check your soundcard Stereo input is working, of course, is to do the "old wet finger" test on the two inputs to see if they are independent by listening to the "open grid hum" on the two outputs. Once that is established you should be able to identify the root cause from there.

  • Soundcard/PC Problems

    If your Soft Rock board worked OK on another PC, then the problem is either:
    • that you have either a mono mike input typical of a lot of laptops and you can do nothing about it apart from a new sound card that does have a stereo mike input and indeed a "Line In" jack socket
    • Or you may have a "Line In" which will be stereo (i.e.L/R - IQ) but for some reason some computers don't have a "Line In" socket (one of the most common cost-cutting tactics in laptops).
    • Or you have a box ticked somewhere in the sound card set up software that makes the input mono. If you open up the M$ Windows Sound mixer menu the Mic input horizontal balance slider may be "Greyed out" i.e. it can't be moved. In which case the machine does not have a stereo input for the Mic.... not very SDR friendly!

  • Construction/Connection Problems

    However if the condition exists on other PCs then there may be a construction error on the softrock board
    • This could be that the input transformer in the BPF stage is made incorrectly or it is correct but not hooked up (or soldered) correctly i.e. the phasing is wrong. More than a few builders have inadvertently gotten wires/windings crossed when mounting the BPF transformer(s) only to discover the dreaded "mirror image" symptom.
    • Or there is a fault in the soldering of any of the chips. In this case suggest you resolder the pads and clean off surplus solder with solder wick braid, an indispensible aid in getting the little blighters to connect properly.
    • Or your stereo plug has got a shorting whisker.
    • One other, but extremely rare, cause could be a defective (or ESD zapped) mixer or opamp chip.
GL 73 petefmt

Problem: "Hump" at and around Center Frequency

Common Causes

What you're seeing are audible-spectrum audio signals travelling into your sound card.

The "hump" around the center frequency is almost always caused by ground loops. The more different ground paths/points that exist for the audio I/Q signals to reach the PC, the greater is the likelihood that common currents (i.e., 60 hZ - 50 hZ for the Europeans) and their harmonics will be caused along the audio signal path(s). This bogus "signal" manifests itself as a "hump", typically about 120 hZ wide, centered on the center frequency.

As Bill, KD5TFD puts it:
" What is happening is that the hardware takes the swath of RF centered on, for example, 7.056 MHz, and downconverts it such that 7.056 MHz is essentially at DC. Also down there near DC is noise and 60 hz (and its harmonics) hum. When one tries to tune down near 7.056, the local software oscillator will be down in the 100's of hz's. The output of this will be Software LO +/- 60 Hz (and harmonics). Things will always be a bit messy near the center frequency -- quieter less hummy sound cards can help, but I don't believe there's any practical way to get rid of all the gunk in the middle."

Possible Sources of the Problem

Multiple ground points

A good example of this is the RXTX series, where there are two separate stereo cables, each with a ground line, connecting the board to the PC. It is even worse when two separate sound cards are used, one for TX and one for RX.

Eliminating reduntant ground points is a good tactic for reducing the "hump"  For example, in the RXTX, you could disconnect the ground lead on the TX input, since the ground lead for the RX will serve double duty.

Common Currents from the Power Supply

Often, builders will use a "wall wart" to power their rig.  This will work, but it introduces yet another ground point  for ground loops.

If at all possible, you should use battery power or some other dc supply with isolation from the mains ground.  Most successful builders use a small gel cell that can be kept on a trickle charger when not in use.  For the SR Lite versions, a simple 9 volt battery will yield good life and excellent operation.

You should have one ground connection between the SR-40 and the PC, using the signal cable(power should not be flowing), and the power of the two units should be totally separate. That will cut down 60 Hz hum and it's harmonics.