[Elecraft] K3 Filter Ring with Noise?

[Ron D'Eau Claire]

Go here:

http://tonometric.com/adaptivepitch/

But to the point of reading CW in QRM and noise, it's not pitch
discrimination that is tested but our brain's ability to detect a coherent
tone (the signal) in random noise. 

As others noted, one way to produce a pure tone is to filter random noise
down to a single frequency. Indeed, that's how almost all of our oscillator
circuits work; take noise produced by an amplifier and filter it so only one
frequency is present. You can use an amplifier (receiver) and "band noise"
(QRN) in the same way.  

So the better we filter out all but the signal, the harder it is to tell
what's signal and what's noise other than by pure amplitude change. 

I'm not expert on human hearing, but I suspect we can discriminate tones
(and hear tones in noise) much better than we can discriminate subtle
changes in amplitude, especially when one or both of the signals are varying
in strength (QSB). 

Ron AC7AC

-----Original Message-----
From: elecraft-bounces at mailman.qth.net
[mailto:elecraft-bounces at mailman.qth.net] On Behalf Of w9cf

While this seems reasonable, when I took an online frequency
discrimination test a year or so ago (unfortunately I haven't been able
to find the link), I did not find this to be true for me.  This test
was, of course, not for separating two cw signals, but instead it sent
two tones sequentially, and you had to choose whether the second was
higher or lower pitch than the first. It repeated this with smaller and
larger intervals until it had a measurement within error bars. Obviously
different people have different results or there wouldn't be a need for
a test. My results were, for the 3 ranges that the test used:

250Hz tones 3.6Hz difference with standard deviation 2.1Hz

500Hz tones 1.9Hz difference with standard deviation 1.4Hz

1000Hz tones 3.4Hz difference with standard deviation 2.2Hz 

Taking the fractions I get 1.4 percent at 250 Hz, 0.4 percent at 500 Hz,
and 0.3 percent at 1000Hz. Although the standard deviations cloud this
a bit. I know some ops like to run at low frequency as Jim says, but I
prefer higher frequencies, and this hearing test may help indicate why.

73 Kevin w9cf