Thanks Dan for copying me. I just subscribed to the list.
The issue is discussed at length by Doug Smith, who is also the Editor
of ARRL's QEX.
http://www.doug-smith.net/cwbandwidth1.htm
In the digital communication world, we have a baseband modulation
scheme called Pulse Amplitude modulation (PAM), which is one of the
simplest form of modulation. CW can be thought of as one form of PAM.
To avoid Inter Symbol Interference (ISI), the pulse shape used in the
PAM scheme should have certain properties. Key Clicks are nothing but
abrupt changes in the waveform, which interferes with the neighbouring
pulse shapes, which is ISI. The Sinc Pulse is one such waveform, but
is difficult to synthesize in real world. Another such pulseshape
which satisfies this is the Raised Cosine waveform.
http://en.wikipedia.org/wiki/Raised-cosine_filter
Another variant is the square root of raised cosine shape, where
transmitter and reciever both use a square root of the above filter
response in (frequency domain) and when you convolve them together in
time domain you get the raise cosine response.
I think I have complicated it too much. :-( The above links explain
it in a much better way.
This is the basic principle behind it, but there are many refinements.
Ramakrishnan, VU3RDD
On 9/20/06, Dan KB6NU <[EMAIL PROTECTED]> wrote:
This discussion is very interesting to me. I hate copying stations
whose keying isn't sharp enough. The dits and dahs seem to blend
together. It seems to me that with all the DSP power that modern rigs
have, there should be a way to "sharpen up" a CW signal to make it
more intelligible. I'm copying this message to my friend VU3RDD, who
is a DSP engineer, to see what he might think about this.
73!
Dan KB6NU
----------------------------------------------------------
CW Geek and MI Affiliated Club Coordinator
Read my ham radio blog at www.kb6nu.com
LET'S GET MORE KIDS INTO HAM RADIO!
On Sep 20, 2006, at 12:31 AM, Don Wilhelm wrote:
> Darrell,
>
> That has ben dropped from the more recent handbook (does not appear
> in my
> 2005 edition).
> I am going 'out on a limb' here by saying that this stems from the
> concept
> that 'some keyclicks are good' philosophy. Sidebands on a CW
> signal are the
> result of the keying shape, and there is more to it than just the
> rise and
> fall times - there is the rounding at the corners to consider too.
>
> I have heard many an operator state that 'hard keying' will get you
> through
> a pile-up better. While that may be true, it certainly is not
> 'neighborly'.
>
> I am glad to see that this statement does not appear in the more
> recent
> handbooks.
>
> 73,
> Don W3FPR
>
>
>> -----Original Message-----
>> From: [EMAIL PROTECTED]
>> [mailto:[EMAIL PROTECTED] Behalf Of Darrell
>> Bellerive
>> Sent: Tuesday, September 19, 2006 6:08 PM
>> To: Elecraft List
>> Subject: [Elecraft] Hearing CW - Fundamental Keying Waveform?
>>
>>
>> The recent thread on filter settings and hearing reminded me of a
>> question I
>> would like to get an answer to.
>>
>> In the ARRL 2001 Handbook on page 15.7 we find:
>> "The dots and dashes of a CW signal must start and stop
>> abruptly enough so
>> we can clearly distinguish the carrier's presences and absences
>> from noise,
>> especially when fading prevails. The keying sidebands, which sound
>> like
>> little more than thumps when listened to on their own, help our
>> brains be
>> sure when the carrier tone starts and stops.
>> It so happens that we always need to hear one or more harmonics
>> of the
>> fundamental keying waveform for the code to sound sufficiently
>> crisp."
>>
>> What is meant by "the fundamental keying waveform"?
>>
>> How do we take "the need to hear one or more harmonics of the
>> fundamental
>> keying waveform" into account when setting up the IF and audio
>> filters?
>>
>> 73,
>>
>> Darrell VA7TO K2 #5093
>>
>> --
>> Darrell Bellerive
>> Amateur Radio Stations VA7TO and VE7CLA
>> Grand Forks, British Columbia, Canada
--
Ramakrishnan - VU3RDD
_______________________________________________
Elecraft mailing list
Post to: Elecraft@mailman.qth.net
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com
