USB 6008 Voltage Measurement varying by 1mV

[DLack]

Hi, I'm trying to make a bias supply for an amplifier using the USB 6008. The ranges I'm looking at are -0.5 to 0.5 V and 0.5 to 2.5 V. To generate negative voltages I'm using the +2.5V output as a 'ground' for a differential voltage reading. The voltage measurements have an offset which varies with voltage, after accounting for this by plotting these variations and adjusting the output from the DAQ there is still a 'noise' of +/- 1 mv as can be seen from this plot of voltage against sample number:

It appears that at some point the values are being rounded to the nearest millivolt. I need to get to a resolution of 0.5 millivolts for my device, will this be possible with the USB 6008?

[johnsold · Accepted solution]

The USB-600x AO has a range of 0 to +5 V and resolution of 12 bits. 5/4096 = 1.22 mV. The absolute accuracy is 7 mV typical and 36.4 mV maximum at full scale.  The AO noise is not specified.  

If you are measuring the results with the USB-6008 AI, you have at least 0.5 mV system noise, similar resolution and absolute accuracy of 2.5 mV or more.

This is probably as good as you will get with the USB-6008.

Lynn

[DLack]

Thank for the quick reply, that clears some things up for me.