Fluctuations in the Analog Output of NI USB 6211

[sid98]

Dear Friends,

My problem is not fatal but it sure is giving me lots of headaches and it is becoming a bit difficult to detect the source.

I am basically studying the frequency response of materials when excited by a large sinusoidal current and based on this frequency response deducting the material properties. 

For this system I am using the analog output of the USB 6211 to generate the current (I am acheiving large currents via power amplification) and using the analog inputs to detect the input voltage signals. For this system to work well I need the output to be stable.However for some strange reasons there is a sudden jump in voltage readings (few mV) but this happens only at certain random frequencies.

What could be the reason for this error? Could it be the way I am generating the output? I am using in-built LabVIEW function to generate continuous output samples. Or is it a limitation of the DC power supply used to amplify the current?

I hope I was clear in my explanation and cross my fingers hoping somebody else has faced a similar problem 🙂

Regards

Sid

[pfehlner]

Heey,

i have a few questions concerning the behaviour you are describing.

1) Does it always happen at the same frequencies? or at different frequencies?

2) Can you reproduce the behaviour or is it completly random?

3) Which LabVIEW functions are you using? Do you use the DAQmx continuous voltag output example? or some modfied code?

4) What is your measurement setup relating this article?

Field Wiring and Noise Considerations for Analog Signals - National Instruments
http://www.ni.com/white-paper/3344/en/

The measurement for the other frequences etc is fine without strange behaviour?

Greetings!

[sid98]

Hi Pfehlner,

Thanks for the reply.

Here are the answers to your question:

1) Does it always happen at the same frequencies? or at different frequencies?

It happens at random frequencies. Sometimes the reading at given frequency is ok and sometimes it's noisy because of these peaks.

The jump in the voltage is ~0.2-0.3 mV and it seems to happen only  for a small part of the cycle. It is small but enough to wreck my measurement.

2) Can you reproduce the behaviour or is it completly random?

It is random. Sometimes I get lucky and it does not seems to affect my readings much.

3) Which LabVIEW functions are you using? Do you use the DAQmx continuous voltag output example? or some modfied code?

I am using the DAQmx continuous voltage example. I am attaching  a screenshot along with the .vi, if you wish to have a closer look at it.

4) What is your measurement setup relating this article?

So, basically I am passing an AC current  (~0.5 A) thorugh a heater wire deposited on  material samples and  then measuring the voltage across the heater for post processing.

I have used a power amplifier to magnify the current through the heater, and voltage amplifiers to magnify the voltage that I am reading, as they are in the order of mVs.

Thanks in advance.

Regards

Sid

[pfehlner]

Hey,

for me the VI seems fine and i would guess the reason is some issue in the measurement setup.

So are you doing a differential measurement or a measurement with relation to the ground? According to what you wrote, it seems that your generated AC Current and your measurement setup after/during the heating has the "same ground" and therefore there may be ground loops which cause the random noise in your measurement.

Greetings!

[sid98]

The measurements are taken with relation to ground (RSE).

Do you think I should change it to differential measurement?

[pfehlner]

If you dont need all the analog input channels of your device/module then the differential measurement is in general more accurate, because there is for example no common voltage influence and also there are no ground loops. For detailled information concerning that topic, you can look up the following link.

Field Wiring and Noise Considerations for Analog Signals - National Instruments
http://www.ni.com/white-paper/3344/en/

I guess that should help!

Good luck in your project!

[sid98]

Hi,
Thanks again for your suggestions.
Unfortunately the DAQ is actually linked to a PCB, so I am limited in the number of improvisations I can do to the setup.
However, I had a closer look at the frequency response results and ran them a number of times, I noticed that the noise seems to pop up at certain frequencies i.e it is not completely random and the pattern is in fact slightly repetitive, although not identical each time.
I also analysed some voltage inputs at various channels and am almost sure that the problem is not in the input. It seems like the voltage output i.e the source is the source of noise. But why is it happening at only certain frequencies?
Any thoughts on it?
Thanks in advance
Regards
Sid

[johnsold]

Sid,

How do you detect the mv jumps? Do the jumps occur at the AO or only after the power amplifier? What is the voltage output of the power amplifier? How much does the resistance of the heater change with temperature? What is the material structure of the heater and the materials being tested? Unless you have a thermoelectric effect, how does the heater generate a voltage? What other equipment is connected to the system? What is the range of frequencies you are using?

Have you taken precautions to avoid thermocouple effects in all the relevant connections? Could something mechanical be happnening - such as a sudden, but microscopic  motion due to thermal expansion?

Can you post any data showing the behavior? Often something appears in the data that no one thought to ask until we see it.

Lynn

[sid98]

Hi Lynn,

Thanks for the message.

Answer to some your questions:

The voltage is being applied to a wheatstone bridge...the jumps show up if you take the input at the entry of the bridge i.e. point A (picture attached for this case) or across the bridge.

The current is about 0.3-0.6 A (after power amplification) and the output voltage is about 0.6-1.2 V.

The noise is reptitve in the range of 0.1 to 0.01 Hz.

Inside the box along with the PCB and the USB 6211, there is also a fan to cool the circuitry and a DC motor for automatic bridge balancing.

Regards

Sid

Picture of complete cycle at A i.e. entry point to bridge (you can see the slight distortion on the upper right side)

zoomed in figure:

[johnsold]

Sid,

Your pictures did not show up. In the toolbar at the top of the text box where you type your message is an Insert/edit image button. Use that to put the images into your post.

Lynn