transmitting clock signal

Gee, so all of us old pharts that have been sending Morse Code for a hundred years or so were wrong? Last I looked, Morse was the last bastion of A1 modulation (on-off keying). Seems this is just on-off on a fast scale.

Jim

someone here might have been

In message , dated Thu, 31 Aug 2006, "RST Engineering (jw)" writes

But you send through a severely bandwidth-restricted channel. The received signal is nothing like the 'square wave' you sent, which itself has the rise-time dictated by the carrier frequency, not the fast edges of a square wave.

But Morse code does use a 'square wave' with a fast rise time compared to the data rate. The square part is the envelope of the modulation envelope. From the point of view of the detector (your ears), the rise time is fast.

Look at it the other way around. How narrow would the system's bandwidth have to be for a Morse code to no longer be intelligible?

In message , dated Fri, 1 Sep 2006, Paul Hovnanian P.E. writes

I think that's the right way round. Intelligibility rules! Obviously, it depends on the data rate. Let's take 20 WPM, which I believe is pretty fast. The average character has about 3 symbols, two dits and a dah, and a dah is three dits, and the spacing is one dit. (Don't throw the ARRL book at me, this is all from memory.) That make eight UIs (unit intervals) per character and the average word has five characters, so 40 UIs per word. So a UI is about 1/800 minutes, 75 ms. Now, what 'eye height' (a measure of intersymbol interference) can you accept? Take it that the rise-time needs to be 5% or less of the symbol rate, i.e. 3.75 ms. Looks like you can use an SSB bandwidth of about 300 Hz. But I believe skilled operators can use around 50 Hz and still read signals. I guess 'no longer intelligible' is down in that region, for 20 WPM. But you can always slow down the sending. IIRC, the first transatlantic telegraph cable that worked for long enough to be useful had a maximum rate of 1 symbol per second.