DAQPad-6016 - Noise on all AI channels

[PowerBook User]

I have a brand-new DAQPad-6016, and I'm using MAX and VI Logger (NI-DAQmx vers 7.4.0) to measure dc voltages.

The 6016 records a large amount of noise on all analog input channels at all gain/range settings. For example, I've attached a JPG image of a VI Logger measurement made on a 9-V alkaline battery. The rms noise in this signal is about 20 mV, approximately 20 times larger in magnitude than the spec for the 6016. I have found this noise level to be consistent at all range settings; it's always about 20 times the spec value.

Has anyone else experienced a similar problem?

[AlanArmstead]

Hello, I've got a couple questions.  First, what input terminal configuration are you using?  If you are not using differential, could you try that.  Also, have you grounded any of your input channels?  If so, do you see ~20mV of noise around 0V?  Additionally, could you run the DAQ Diagnostic Utility?  It is located at the bottom of this page under Utility.  It will probably tell you if you have any bad analog input channels. 

-Alan A.

[PowerBook User]

I am using the differential input terminal configuration.

I shorted A4 and A12 together and grounded them to AI GND. With +/-50mV range, I measure 0.00863V with a std dev or 0.0002V (1000 samples). With +/-10V range, I measure 0.0742V with a std dev of 0.026V.

I ran the DAQ Diagnostic Utility. It did not report any error conditions.

Thank you for your interest.

[AlanArmstead]

Could you try to acquire from the internal ground on the device in a test panel to see if you see the same noise?  Could you re-post w/ a screen shot?  See my example here:

-Alan A.

Message Edited by Alan A. on 11-02-2005 06:38 PM

[PowerBook User]

Okay. Here's a screen shot of the Analog Input Test Panel in MAX. I selected Dev1/_aignd_vs_aignd as the input channel, as in your example.

The amplitude of the noise on my device appears to be about 10 times that on your device.

[AlanArmstead]

Hello, So there are a couple things to point out here.  First your device, the DAQPad-6016, doesn't have a -1 to 1 volt range, so it is using the -5 to 5 for you.  Also, from the specs, the absolute accuracy at that range is 2.003mV.  The test panel shows about -0.3mV to 0.5mV at its peaks.  So that appears to be well within spec.  In the picture I attached, I was using an M Series device (better accuracy than the DAQPads) with an available range setting of -1 to 1V, so the difference between your pic and mine is understandable. 

Looking back at the specs for the DAQPad-6016, it says the absolute accuracy at +-10V range is 8.984mV.  So a 20mV noise is certainly not acceptable, but it isn't 20x the spec.

Is there a chance that you were injecting some noise on your signal that you were measuring (in your initial post)?  Could you look at the signal with a DMM or something else to see if it is the source (or connecting wires) that is injecting the noise, and not the DAQPad?

-Alan A.