Path: utzoo!utgpu!news-server.csri.toronto.edu!rutgers!cs.utexas.edu!sun-barr!newstop!exodus!bovic.Eng.Sun.COM!agn
From: agn@bovic.Eng.Sun.COM (Andreas G. Nowatzyk)
Newsgroups: sci.electronics
Subject: Re: High Voltage Probes
Summary: Not so easy
Keywords: capacitive divider
Message-ID: <3328@exodus.Eng.Sun.COM>
Date: 22 Nov 90 20:57:45 GMT
References: <1990Nov15.163940.17119@batcomputer.tn.cornell.edu> <1270002@hplred.HP.COM>
Sender: news@exodus.Eng.Sun.COM
Distribution: na
Organization: Sun Microsystems, Mt. View, Ca.
Lines: 14

Building a high voltage resistor of 100 or 1000 Mohm solves only 1/2 of
the problem if you want to measure anything besides DC. The problem is
due to the input impedance of the scope, which has a capacitive component
to it (say 1 Mohm and 5 pf). Furthermore, the high voltage resistor will
have some parasitic capacitive component too. The net result is an R/C circuit
that distorts your frequency response.

As a first order approximation, you need to parallel the 100 Mohm with
some C such that the total C (including parasitic C) is 0.05 pf. Combined with
the 1Mohm || 5pf of the scope input, you end up with an resistive and
capacitive divider of the same ratio, which doesn't distort. Unfortunately,
there are also inductive components and the effect of the cable, etc.
The cost of good scope probes has a lot to do with the compensation networks
used to minimize distortions.