How to measure 0 to 80millivolt in U3 HV

The resolution of a high voltage input on the +/-10V range is about 5 mV, and the resolution of a low voltage input on the 0-2.4V range is about 600 uV.  So either of those meet your 8mV requirement without any external signal conditioning.

So the low voltage inputs (FIO & EIO) will do best, but not the following if the measurements close to 0.0 are important:

https://labjack.com/support/datasheets/u3/hardware-description/ain/analo...

If you do want to add an amp, check out the LJTick-InAmp:

https://labjack.com/support/datasheets/accessories/ljtick-inamp

... or make your own:

https://labjack.com/support/app-notes/amplifying-small-signal-voltages

Hello

I was wondering if you could direct me on how to remove internal noise on a U3-HV? I have. Velocity transducer attached to Ain0 and gnd. Im getting a variable reading from 0 to 0.113. Even with nothing connected I get a changing voltage. 

The problem being that 0.113 voltage reading translates into about 2.5 in/sec on my scale. This is skewing graphing by way too much. 

Thanks

Jim M

Remove your signal and instead use a jumper wire to connect AIN0 to GND.  Now you can use the Test panel in LJControlPanel to observe the noise level with an ideal signal at the midpoint of the +/-10V range.  Typically on the U3 you will see most readings sit on 1 particular value, with some readings different by plus or minus 1 count.  The weight of each count (voltage resolution) is span/2^12 (12-bit resolution for the U3), so each count is about 5 mV on the high-voltage inputs using the normal +/-10V range on the U3-HV.

The span of the low-voltage analog inputs (FIO/EIO) is about 2.4 volts, so each count is about 0.6 mV on those.  It is harder to do a noise test, as 0.0 (GND) is not always in the valid range for these low-voltage inputs.  You can jumper DAC0 to FIO4 (aka AIN4) to test, but DAC0 is not super quiet so you might see a little more noise than the analog input has itself.  A 1.5V battery (e.g. AA, C, or D) connected to FIO4 and GND is a good test.

Note from post #2 above that a low-voltage AIN is best for your 0-80 mV signal, but you need to do the differential reading as described in 2.6.3.9 so you can read all the way to 0.0.