Newsgroups: sci.electronics
Path: cantaloupe.srv.cs.cmu.edu!crabapple.srv.cs.cmu.edu!fs7.ece.cmu.edu!europa.eng.gtefsd.com!howland.reston.ans.net!zaphod.mps.ohio-state.edu!saimiri.primate.wisc.edu!doug.cae.wisc.edu!kolstad
From: kolstad@cae.wisc.edu (Joel Kolstad)
Subject: Re: Dumb Question: Function Generator
Organization: U of Wisconsin-Madison College of Engineering
Date: 15 Apr 93 13:21:16 CDT
Message-ID: <1993Apr15.132117.26438@doug.cae.wisc.edu>
References: <C5J845.3B8@murdoch.acc.Virginia.EDU>
Lines: 35

In article <C5J845.3B8@murdoch.acc.Virginia.EDU> dgj2y@kelvin.seas.Virginia.EDU (David Glen Jacobowitz) writes:
>
>	I have a new scope and I thought I'd save a few bucks by
>buying one with a function generator built in.

Hmm... now where was that ad for the combination radio/hand cranked
generator/flashlight/siren I saw? :-)

[function generator has a 50mV offset, and the amplitude's too high]

>	Is there any way I could make myself a little box that could
>solve this little problem. The box would tkae the function generator
>input, lower the voltage and give an output impedance that is some
>low, unchanging number. I would want to lower the voltage by a factor
>of one hundred or so. I could just build a little buffer amp, but I'd
>like to have this box not be active.

Sure, you've already got the right idea.

Ignoring the 50 ohm internal resistance of the generator for a second, just
run it into, say, a voltage divider made of 990 ohms in series with 10
ohms.  This new circuit is the Thevenin equivalent of one that puts out
1/100 of the original voltage, and has an output impedence of negligibly
less than 10 ohms.  You may want to monkey with the values a little
depending on whether you care more about the _exact_ dividing ratio or
the availability of parts.

Hows that sound?

					---Joel Kolstad

P.S. -- This is why those 1000:1 high voltage probes for multimeters can be
accurate but still cheap.  They have something like 100 megs in series with
100k, which doesn't load the (often high impedence) source much, as well as
keeping the (probably 10 meg impedance) multimeter happy.
