[Lockheed]

[.]

 

[qusp]

did you ground the input or leave it floating? if you left it floating you have only really measured the difference of the voltage of each power supply rail. so theres that and given you dont seem to have given it any signal, what is it offsetting? the effort is appreciated i'm sure, but its not really very useful info as is, myself I tend to trust the datasheets with this sort of spec. your meter is very unlikely to provide useful measurements down that low either.

 

[Lockheed]

[.]

 

[qusp]

I didnt mean it like that, its a good exercise anyway, but yeah you havent really measured anything but the power supplies. the opamps generate a virtual ground between their supply rails, any imbalance between them will reflect in what that middle ground point is and thus what the signal is referenced to. Without being tied to anything it will just float around.
 
Even if this was set up correctly, unless its a good quality bench meter, I doubt very much it would be able to make useful measurements in the single digit mV, let alone fractions of it. if you look at your meters manual it will tell you the expected accuracy in various ranges. At the low range, most will at best be 5%, usually worse.
 
if you still have it set up, measure a couple with the inputs tied directly to input ground, then try while feeding it an AC signal, though many meters will also have trouble measuring DC floating on an AC signal accurately.

 

[Lockheed]

[.]

 

[peranders]

Never thought of having some gain? If you set the gain to 100 or 1000 then you won't need any high resolution of your voltmeter.

 

[qusp]

yeah but then he'll be measuring the accuracy of the resistor =)