Topband: Tuned speaker revisited

[Ron Spencer]

Talked with a buddy about this. From this I've a couple new takes on this topic. I'll preface all this by saying I am NO expert and some, perhaps most of this is hogwash 



Dave and Adrian (W0FLS and VK2WF)  had 2 different experiences with the same design of speaker. Why? Initial conclusion was analog vs SDR for the radio. 



In rethinking this, I've a new view. The speaker is radio agnostic. It doesn't care what its plugged in to. Its just a filter. The variable we haven't considered is us. 



If I put two of us in a room and play a pile up I'm betting each would copy different calls first. Why? Why do some of us choose a low and some high cw side tone pitch? 



I propose its a mix of how we trained our brain to interpret cw and how the audio path from our ears to brain is working. What's our "audio curve"? Are there "holes" in it? Do we change our preferred side tone pitch as we age to better match that audio curve? 



It would be interesting to see how Dave and Adrian's experience changed if the pitch peak of the tuned speaker 
was changed. 



Mixed into the discussion was SNR (signal to noise ratio). My buddy said something interesting. Something that, perhaps, everyone already knows and I'm late to the party.





I'll start with this statement. With regard to signals we are interested in (i.e. radio signals) SNR is independent of the radio, the antenna or anything in between. The SNR at a dipole is the same as at a 4 element beam. Why? 



An astronomy example might help. Look at the stars with just your eyes. Now look through binoculars. You'll see more stars with the binocs. Did the stars get brighter? Of course not. The binocs just gathered more light. The SNR (or the visual equivalent) didn't change. Binocs, just like our antennas, are passive and incapable of changing the SNR. They simply gather more signal. 



To help us hear weaker signals we do what the binoculars did and use antennas that gather more RF of interest. So we use antennas with more gain. Then we process it through our radios and listen through our headphones or speaker. For the radio, we care about minimum discernable signal which is affected by things like noise figure and radio design. Those of you that play at the other end of the spectrum (i.e. 2, 432, 1296) know how important noise figure is. For HF, not so much as band noise is much higher making a low noise figure not nearly as important. 



Tree asked a good question. How can Joe Taylor's WSJT software decode signals we can't hear. Just because we can't hear it doesn't mean the signal isn't there. We blow a dog whistle. To the dog there's a signal :) Joe's software processes the signal to find the repetitive tones ft8 uses (now we're getting into fast Fourier transforms). Why can WSJT decode weaker ft8 signals vs ft4? I believe its because we have half as long for the fft to work on the ft4 signals. 


DSP speakers or tuned speakers or whatever simply tailor the audio to better match our audio path. Binaural effects are an example. That doesn't change the SNR but provides a signal that, for some of us, allow our ear/ brain interface to better process the signals. 



Yes, a long post and, perhaps for many, nothing new. But, perhaps just for me, provided an opportunity to do a deeper "think" about stuff I just took for granted. And write it down so I don't forget it....



Ron

N4XD
Sent using https://www.zoho.com/mail/