Measuring millivolts from volts
Test & Measurement World now has a LinkedIn group. Yesterday, John Kirk, Research Engineer at Raymond Corporation posted the following question.
Here’s my problem: measuring voltage accurately. That sounds easy, but I want to distinguish 20.000V from 20.005 and go on up to 40.000 versus 40.005V. If we were measuring between 0 and 1V then mV accuracy would be easy, but the mV on top of Volts is my problem. Any suggestions?
As with many engineering problems, it has more than one solution. One is to use a bridge or other kind of nulling circuit to remove the high-level voltage. Another is to have an instrument (DMM) with enough resolution. John needs 6 digits of resolution to measure 5mV from 20V. There are several bench DMMs on the market that can do that. The problem may come down to money because a new 6-1/2 digit DMM costs over $1000. Used DMMs are available for less.
Martin Rowe commented:
M. Sell is correct. Good measurement techniques are essential. Errors from EMI, thermal effects ground loops, and numerous other problems can kill measurement accuracy.
M.Sell commented:
It's one thing to have a DMM with the resolution to measure 5 millivolts, but quite another to measure 5 millivolts ACCURATELY.
Don't forget - resolution is not accuracy.
