It's all about the Software

In the Audio Spectrum display when selecting the oscilloscope view, the audio is still unsymmetrical for AM mode (see circle in screenshot). The display is no longer centered evenly above/below zero as it used to be and is on all other modes.

Also, the icon (see arrow in screenshot) is not properly rendered until or unless the mouse icon hovers over it.

Doesn't look that bad in your image... but here's a close up mine. I never use that feature, due to occupying valuable space for Mode, Filters, and the settings below them. So I never noticed, I spend most of my listening on LSB or USB, with an occasional HF BC in AM.
It appears on both the Airspy and the RSP1A, so it's not SDR mfg specific.
As for the 'Quick Access Tool Bar' icon, one can either left click to open it, or right click to move it. Works 100% on v3.1 here.


73 Kriss KA1GJU
Home of the SDRC Super Station Servers in NH, USA

Hi Gregsa,

Take a look at this thread:

viewtopic.php?f=46&t=853

Barry
WB2ZXJ

I've seen that DC offset in all versions since 3.0.26. Did not occur in 3.0.25 or older.

The scrolling waveform appears to be about half its normal amplitude, has significant DC offset, and is asymmetrical (positive-going peaks are smaller than negative-going peaks). The DC offset is coupled all the way through to my speakers, so the woofer pops outward whenever I use AM mode.

My radio is an SDRplay RSPdx, and before that an SDRplay RSP2pro.

yep, activate the 25Hz high-pass filter in the equalizer settings.

Equalizer??
<dig dig dig ...>
Oh look, an audio equalizer.
Yes, that fixes it.
Thank you!

Perhaps the answer was in the above-mentioned thread, but "You are not authorised to read this forum".

So why is Simon so silent for so long on this? Turning on 25Hz HPF in equalizer is not a fix. Look at the white noise when tuned to an inactive frequency, it's still unsymmetrical. SAM mode doesn't have this problem. Neither does CW, BC-FM, N-FM, W-FM, LSB or USB. As much as I like this program and as good as it is with all its features, this issue makes me wonder about the faithfulness of the AM audio demodulation fidelity. Something's amiss here for that display to not be symmetrical. And, as another poster said, it goes all the way out to the speakers to pop the woofer with that DC offset.

It very probably traces back to the fact that you are not looking at a continuous spectrum as much as a spectrum of average levels over a short period of time several thousand to several million RF samples long. FM spectrums look very asymmetrical for this sort of reason. I'd expect AM spectrums to show some of the same effects. I've never paid much attention to it, especially given how Simon derives his spectrum information. (He's designing a ham/SWL radio not an instrument. So he apparently wants to emphasize features that help hams and SWLs rather than lab instrument accuracy.)

This asymmetry affects both Simon's radio, SDRSharp, and my own play with SDR applications.

{^_^}

Interesting catch. I've been playing around with this and noticed a few more things about it.

If you tune to WWV and then reduce the bandwidth as low as it will go (10 Hz), observe how it looks in SAM. Then when switch to AM (make sure both the SAM and AM filters are 10 Hz), observe how the center line immeadately moves up but the modulation artifacts are almost gone (probably not the right terminology but hopefully you'll see what I mean). Anyway, very strange. Then when you switch back to SAM, notice how the center line slowly drifts down to the scope center where it is supposed to be.

Also noticed on SAM that while the center line is generally where it should be, at seemingly random moments (at least for me) it would occasionally drift up and down a bit off the center scope line and then eventually went back to the center. If you turn the EQ on with the 25 Hz HPF on, then it immeadately brings it back to center if it was drifting. If it wasn't drifting at the moment HPF was engaged, then there would be no change.

Another thing, if you go back to AM with the 25 Hz HPF on, notice how fat the center line is. It's not like that in SAM.

RadioSDR1, yes, very interesting. I'd really like to know what's going on deep inside the decoding here. Something's amiss and enabling a 25Hz HPF isn't a fix. There shouldn't be any DC offset, where is it coming from?