Noisy Data from USB-231 Differential DAQ

[Bisan]

Hello,

When I set my voltage generator to 0V and I connect it to the analog input pins of the DAQ, I get 0.02V. I was using my code to acquire the analog data and also InstaCal and with both I get noisy data. Is there a solution to this problem?  

 

Thank you.

[DAQman]

If you want to test the input offset, you can just connect the input to the analog ground terminal. If the USB-231 is in differential mode, connect both sides to the analog ground. Assuming the measurement is very close to zero, I suspect your voltage generator has an offset, or a ground loop exists between it and the USB-231.  If you suspect a ground loop and have a laptop, run it on battery power to see if the measurement improves.

 

 

[Bisan]

Thank you for helping.

I checked the input offset and indeed, I was getting 0.

I also checked for the ground loop, but I was getting the same offset so there is no ground loop.

I was using before the USB-205 single-ended DAQ and the offset was 0.04V so I moved to a differential DAQ in order to reduce this offset and it got reduced to 0.02V. Is there anything else I can do to get rid of this noise? I thought that changing the DAQ to a differential model will reduce the noisy in a more significant way.

[DAQman]

Depending on what is connected, sampling multiple channels can add what appears to be noise. Can you explain how you measure the noise and whether it is a 20mV peak or peak-to-peak value?

[Bisan]

I was sampling just one channel using a voltage generator. I was using InstaCal and doing an analog test. When looking at the data I was noticing a 0.02V error for the differential DAQ and 0.04V for the single ended one based on the value set for the voltage generator.  

[DAQman]

The noise is not from the USB-231. Either the wires pick up the noise from the surrounding, or the voltage generator is low-quality. Invest in shielded twisted-pair wire if the wires are picking up the noise. Connect the shield to the earth's ground at one end only. 

If the noise is from the voltage generator, use averaging to reduce it. For example, if you want 100 samples per second, increase the rate to 1000 and average every ten samples.

[Bisan]

Understood, Thank you.
One more question please. Are the analog input channels synchronized? Will the data acquisition start at the same time for all channels or there is a delay between them? 

[DAQman]

The USB-231 is a multiplexed device with a single A/D that measures each channel one at a time. The channel-to-channel is the inverse of the sample rate. 

[Bisan]

That is the case for the software paced mode right? Can I use the hardware paced mode to scan two channels at the same time? If yes how can I do it?

[DAQman]

You can use the AInScan function for hardware pacing. However, the channels will not be sampled at the same time. Let's say you want to read four channels at 1000 S/s. The time between each channel will be 1/1000 or 1mS. The time from when the first channel is read to when the fourth channel is read is 4mS.