View Single Post
  #49   Report Post  
Old October 11th 14, 04:00 PM posted to rec.radio.amateur.homebrew
gareth gareth is offline
external usenet poster
 
First recorded activity by RadioBanter: Sep 2012
Posts: 1,382
Default Frequency accuracy in older RXs

"Michael Black" wrote in message
news:alpine.LNX.2.02.1410111058430.23447@darkstar. example.org...
On Sat, 11 Oct 2014, Ralph Mowery wrote:
"gareth" wrote in message
...
Oops! Got the two swapped round!
Square = odds, Triangle = evens, Sawtooth = raucous!

Still sort of wrong.
Square and trigangle are both odd harmonics.
Square wave harmonics are reduced by 1/N and triangle waves are sort of
1
over the square of the harmonic.

I forget the reasons for triangle waveforms in electronic music, or
testing, but one thing is, it's easy to generate, and with diodes, you can
turn it into a reasonable sinewave. There was the Intersil 8038 function
generator IC (Exar had similar products) that put out square, triangle and
sinewaves. The sinewaves were synthesized with diodes from the triangle.
The squarewave was a byproduct of generating the triangle. And there was
an adjustment so you could turn the trianble into a sawtooth, which has
its own uses.


I'll have to revise my Fourier from 40 years ago about the triangle harmonic
content

But CAVEAT EMPTOR!!!!!!

The 8038 has a hardware bug in it in the triangle plus sine shaper in that
at
the peak of the waveform, it seems to switch over to a reverse function
resulting
in a very deep spike at the top of the waveform.