[Elecraft] Final Update To Keying Mod Change

Thu Apr 15 00:56:01 2004

Guy;
It works best with random noise, but is a reasonable approximation with any
broadband signal that doesn't contain narrow peaks in intensity.  A flat
frequency response is also desirable, but rarely true in practice.

I understand  your concern about slope.  You have the exact same problem
making phase noise measurements, which are random noise, but have a slope to
them.  The finite width of most filters will introduce some error into the
measurement.

The 25ms Hanning Window has a bandwidth of about 100 Hz and doesn't have a
very steep skirt selectivity compared to most radio filters.  But its
response does fall off much faster than a rectangular time domain filter, so
I believe it is a reasonable window in this situation.

Since the intensity of the click does have a definite slope, the answer
presented will somewhat over estimate the intensity, especially at 500 Hz.
Even if the absolute answers are off by a few dB, the relative comparison
between the them should have less error since they are all subject to a
power level that is decreasing with frequency.
-John
 KI6WX

----- Original Message ----- 
From: "Guy Olinger, K2AV" <[email protected]>

> Does that apply properly to a noise power that is not distributed
> equally across bandpass, and not effectively random?
>
> Part of the nastiness of computing click power is that it is from a
> synchronous event, and diminishes with frequency so much so that just
> about all but the narrowest of bandwidths will carry unequal click
> power in the upper and lower half of the bandwidth.
>
> Don't think the usual white-ish noise methods with dBc/Hz apply.
>
> 73, Guy.
>
> ----- Original Message ----- 
> From: "John, KI6WX" <[email protected]>
> To: "Ron D'Eau Claire" <[email protected]>;
> <[email protected]>
> Sent: Wednesday, April 14, 2004 10:13 PM
> Subject: Re: [Elecraft] Final Update To Keying Mod Change
>
>
> > Ron;
> > For this measurement, I used units of dBc/Hz.  You have to multiply
> by the
> > bandwidth of your receiver to get the received power.  This is the
> correct
> > units of measurement for a signal that covers a wide bandwidth.  So
> if you
> > are using a 100 Hz bandwidth, add 20 dB to get the power that your
> receiver
> > actually sees.  If it is a 500 Hz bandwidth, add 27 dB.
> >
> > Does that make it clearer?
> > -John
> >  KI6WX
> >
> > ----- Original Message ----- 
> > From: "Ron D'Eau Claire" <[email protected]>
> > To: <[email protected]>
> > Sent: Tuesday, April 13, 2004 5:21 PM
> > Subject: RE: [Elecraft] Final Update To Keying Mod Change
> >
> >
> > That really interesting John. Thank you for the good effort.
> >
> > Let's see if I understand the levels correctly. A "stock" K2 has a
> peak
> > energy in the sidebands at 500 Hz from the carrier that is about 62
> db below
> > the carrier. If I'm scribbling correctly here, that means that if
> the
> > carrier is at 100 watts output, then the energy at the peak of the
> "click"
> > is about 0.000064 watts or 64 microwatts.
> >
> > If we use the common scale of 6 db per S-unit, if signal from a
> "stock" K2
> > at someone's receiver is S9+10 db then the clicks 500 Hz away are
> just at
> > S-zero. At 1 kHz from the carrier, the signal would have to be about
> S9+20
> > db for the clicks to be as loud as S-zero. Using the conventional
> definition
> > of 50 uV for an S-9 signal, an S9+20 db signal is putting a full 1/2
> VOLT on
> > the antenna terminals.
> >
> > That's certainly audible if the transmitter is very close to a
> receiver is
> > capable of handling that sort of RF voltage at the antenna without
> > overloading. In installations where transmitters and receivers are
> very
> > close to each other, the mods can make a significant improvement,
> obviously.
> >
> >
> > However, in normal on-the air use, I'm not sure the bands are quiet
> enough
> > for me to hear a noise pulses at S0 or even S1. And they drop in
> level very
> > quickly as we tune more than 500 Hz from the carrier.
> >
> > Ron AC7AC
> >
> >
> > -----Original Message-----
> > I've posted a final update to the 180K resistor value change for the
> > Elecraft Keying Modification.  This update adds a new way to measure
> key
> > clicks from a transmitter.  The new method can distinguish the peak
> > intensity of a key click rather than the average intensity
> measurements that
> > are made with a spectrum analyzer.  This better reflects the true
> effect of
> > a key click on a receiver.
> >
> > The new method also measures the peak intensity of a key click for
> both the
> > rising and falling edges of the waveform.  The measurements on the
> K2 show
> > that in some cases one edge of the keying waveform can generate 10
> dB
> > greater intensity than the other edge.
> >
> > The latest version of this document can be accessed at:
> >
> > http://home.pacbell.net/johngreb/improving_elecraft_keyingmod.pdf
> >
> > -John
> >  KI6WX
> >
> >
> > _______________________________________________
> > Elecraft mailing list: [email protected]
> > You must be subscribed to post to the list.
> > To subscribe or unsubscribe see:
> > http://mailman.qth.net/mailman/listinfo/elecraft
> > Elecraft Web Page: http://www.elecraft.com
> > Also see: http://www.elecraft.com/elecraft_list_guidelines.htm
> >
> >
> > _______________________________________________
> > Elecraft mailing list: [email protected]
> > You must be subscribed to post to the list.
> > To subscribe or unsubscribe see:
> http://mailman.qth.net/mailman/listinfo/elecraft
> > Elecraft Web Page: http://www.elecraft.com
> > Also see: http://www.elecraft.com/elecraft_list_guidelines.htm
> >
>