[1000mp] ARRL testing of Clicks

[Hare,Ed, W1RFI]

Hi, guys and gals,

Here is an interesting graph. This is draft data and may not represent the
way we want to test, but I wanted to toss it out and see what people think.

http://www.arrl.org/~ehare/temp/sidebands.pdf

Because it may be of some interest to this group, and this has been the
subject of considerable discussion here, this is a test of the Field. We set
the carrier reference level to 0 dB and the measurement bandwidth to 10 Hz.
(Using a 300 Hz bandwidth actually ended up measuring the bandwidth of the
analyzer -- the closer-in data were made higher by the fact that the
analyzer was responding to the fundamental carrier signal.)  We then
normalized the 10-Hz measurement by +10log(500/10) to show the value in a
500 Hz bandwidth. Because the carrier level would not be affected by the
bandwidth change the way that the noise-like sidebands are, we removed the
data +- 250 Hz from the carrier.  The transmitter was adjusted to rated
output power. This is the keying sideband spectral plot of a string of dits,
not representing the stronger click that would be manifest from the first
dit, by my read of the keying waveform.  At 5 kHz away, the key clicks from
this rig would be about 70 dB down in a 500 Hz bandwidth receiver.  	

I am not 100% convinced that 10log(500/10) is as absolute as it looks at
first glance.  The keying sidebands are literally that -- an OOK CW is a
full-carrier, double-sideband signal with discrete keying sidebands that
result from the shaped square wave used to modulate the carrier.  If those
sidebands are noiselike, their quasi-peak detected amplitude should vary as
10log(BW); if the sidebands are discrete, at multiples of the keying
frequency, the detected amplitude would vary as 
20log(number_of_sidebands_in_the_measurement_bandwidth).  In this case, the
difference at +- 5 kHz between 300 Hz measured data and 10 Hz measured data
was about 15 dB, indicating that the energy was noiselike.  

This graph is intended to be a "work in progress" shot, to faciliate
communication and serve as a starting point for input and discussion.

What do you think of this approach?

73, 
Ed Hare, W1RFI
ARRL Lab
225 Main St
Newington, CT 06111
Tel: 860-594-0318
Internet: w1rfi@arrl.org
Web: http://www.arrl.org/tis