🚀 go-pugleaf

walvdrk_r@research.ptt.nl (Kees van der Wal) writes:
>iqbal@omaha.eel.ufl.edu (M.Iqbal) writes: 
>>walvdrk_r@research.ptt.nl (Kees van der Wal) writes: 
>>|>
>>|> [about clock frequency to be fed to 486DX2]
>>|>
>>|>Though I still wonder where my DX2 get's its 66MHz
>>|>clock from if the crystal runs at 33MHz only. Does it contain some
>>|>internal clock doubler circuitry? 
>>
>>Square wave have infinte number of frequencies: a fact explained in 
>>any introductory communication book. If you do not believe me than 
>>try finding Fourier Transform of a pulse train or see a specturm
>>of a pulse train on a spectrum analyzer. 
>
>I believe the book. The book say that symmetric waveforms have no even
>harmonics, so the probably symmetric 25 or 33MHz clock has to be converted into
>something asymmetric to be useful. 
>

What does your book say about a Sawtooth (Ramp) wave?


      /|    /|    /|    /|    /
     / |   / |   / |   / |   /
    /  |  /  |  /  |  /  |  /
   /   | /   | /   | /   | /
  /    |/    |/    |/    |/

According to my reference, a Sawtooth wave has both odd and even harmonics,
whose amplitude is inversely proportional to the harmonic number itself.
I.e.  the 2nd harmonic is 1/2 as loud, 3rd harmonic is 1/3 as loud, 4th is
1/4 as loud, etc.

>
>It would require some kind of filtering to select the second harmonic. As 
>having some LC-filtering on-chip doesn't seem very likely, I'm still wondering 
>how the 486-designers made it work.
>

I've seen probably a dozen frequency doubling techniques in various magazines
and those huge books in the library that reprint circuits from everybody's
data sheets.  Most of them are analog, but there are enough purely digital
ones that it shouldn't be too hard for Intel's engineers to find one for
their cpu.


-- 
:::::::::::::::::::  Greg Andrews  gerg@netcom.com  :::::::::::::::::::
It has just been discovered that research causes cancer in rats.
:::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::