Path: utzoo!utgpu!jarvis.csri.toronto.edu!rutgers!tut.cis.ohio-state.edu!purdue!bu-cs!buengc!bph
From: bph@buengc.BU.EDU (Blair P. Houghton)
Newsgroups: sci.electronics
Subject: Re: 60 Hz notch filter
Message-ID: <3197@buengc.BU.EDU>
Date: 17 Jun 89 00:52:06 GMT
References: <2286@ccnysci.UUCP>
Reply-To: bph@buengc.bu.edu (Blair P. Houghton)
Followup-To: sci.electronics
Organization: Boston Univ. Col. of Eng.
Lines: 29

In article <2286@ccnysci.UUCP> sukenick@ccnysci.UUCP (SYG) writes:
>Can anyone suggest a design for a good notch filter?
>I'm getting lots of noise at 60 Hz (due to pickup; the
>noise is down to 50 mV with shielding, etc., but the signals
>that we look at are in the mV realm...)

Tough to do with passive methods.  60Hz implies mondo capacitors and 
really gnarly coils.  Go active.  Op amps rule!

First, a question:  Why notch?  Why not just lowpass?  If something
below 60Hz is as important as something above 60Hz, then it's
entirely likely that there's something important at or near 60Hz
that you wouldn't want to clobber, in which case you shouldn't.
If it was a data-gathering-for later-analysis setup, I'd add an FFT to
the analysis setup, wipe the spike at 60Hz, and de-FFT to get back the
signal.  Much more elegant than filtering, and cheaper when you consider
that most signal processing software is built around the FFT, so you
don't even have to program it, assuming you're set up for the analysis
in the first place.

Second, advice:  there are lots of books on active filter design.
I don't have one next to me, and I couldn't tell you exactly what sort
of one to use unless I had more info on your requirements, ennyhoo.
The library is your best bet, now.  Go for something low on theory
and high on examples.  You probably don't want to become an expert at
the dark-blue magic of active filtering, just whip up a filter.

				--Blair
				  "Reading Is Fun-damental."