You can reject 60Hz if you have control of your a/d sampling rate, 
or if you can integrate after the a/d.  For instance, suppose each
of your a/d takes 1ms.  The period of 60Hz is 1000/60ms= 16.67ms,
so if you can do each of your 1ms a/d separated by 16.67ms you
will catch 60Hz interference at the same phase each time and it will
appear as a DC component.  Another way to do this is to take
a 1ms sample every 1ms, and then average each block of 16 or 17 together,
so that the 60Hz components approximately average out.  Of course
neither of these techniques gets rid of noise that is not at 60Hz
or a multiple of it.  I use a microprocessor with an onboard
16 bit a/d that takes about 1/60sec, so it rejects 60Hz without
further fuss.  Check your a/d chip data sheet to see if it gives
guidance on minimizing 60Hz.

Hope this helps,

Dave
The Inexpensive Seismometer Project
http://www.infiltec.com/seismo/

> From:    "ERIC  GUINN" 
> Date:    Sat, 9 Feb 2002 12:50:26 -0500
> 
> Hey all.
> 
> I'm trying to get my ducks in a row for my first LP station with computer
> logging, and I'm pulling my hair out trying to find an old piece of data.
> Could someone please help?  Where have I heard of a sampling rate of 60Hz
> (or 30Hz?) which was synchronous and zero-crossing-locked with line
> frequency to allow elimination of 60Hz and aliases without multiple-pole
> filters?  I think I'd like to replicate this bit of Kung-Fu in my design.
> Did I just make this up or did Larry or someone publish an article like
> that?
> 
> Thanks in advance,
> Eric Guinn, AC4LS
> Sevierville, TN


__________________________________________________________

Public Seismic Network Mailing List (PSN-L)