One could discourse all day on what "proper" is in this context without ever a precise consensus among educated gentlemen. At some point of "ugly" we would all agree that anything that sounds like that is improper.
When the only stateful levels were strictly on and off, and the rise and fall transition time amplitude curves were set by a resistor and a capacitor, one could make some simplifications in the discussion. Today, often, as in the K3, the rise and fall times and shapes of CW are DATA which goes DIRECTLY to RF via a DAC. For these the nature of the rise and fall curve is not governed by a charge/discharge time of analog components, but rather by whatever function generated the waveshape data, which is stored, *read* as needed and never generated real time. In many cases it is far more accurate to say that the rise and fall of a baud is INTRODUCED, rather than the signal is keyed. If the rise and fall data curves used for "keying" vary their "sharpness" according to speed, then the description bandwidth increases by speed certainly does apply. But if not, the bandwidth is governed by the pulse generated by the rise and fall data, which could be the exact same from 1 wpm to 40 wpm. We are so easily diverted into our lifelong, unconscious, and utterly habitual analog thinking (note that I certainly do not give myself a free pass here). I was just looking at the P3 display of some traditional BC band stations around here. A 30 over 9 station at 1490 kHz fills up 1480 to 1500. Using the K3's AM-S mode on USB side the S9 station at 1500 kHz is clearly intelligible, but the standard demodulation for 1500 is obliterated by the 1490 station. That ain't your granddaddy's AM out there any more. That's some really complicated stuff that completely fills up +/- 10 kHz and clearly not done the same from station to station. I'm going to save googling that for a free evening when I don't have something broken to fix. There is so much stuff I was taught by my WCTT Chief Engineer Elmer that I am having to set aside. 73, Guy K2AV On Wed, Mar 30, 2016 at 8:28 AM, Joe Subich, W4TV <[email protected]> wrote: > > On 3/30/2016 12:21 AM, Jim Brown wrote: > > Unfortunately the guys who wrote the rules failed to understand the > > physics and the math. > > No Jim, you are wrong here. CW is a carrier modulated by a digital > (on/off keyed) signal of a given (baud) rate. That keying will > generate sidebands +/- the baud rate *just like FSK* or even AM which > can be observed on a spectrum analyzer. The ITU formula is quite > accurate in terms of the actual bandwidth for signals with *properly* > *shaped keying*. > > Again, the excess bandwidth from improperly shaped keying signals > (clicks) and/or amplifier distortion is something else. > > 73, > > ... Joe, W4TV > > ______________________________________________________________ Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:[email protected] This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html Message delivered to [email protected]
