Grounding Problem with NI USB 6281 DAQ device.

[Amartansh13]

Dear all, 

I am using USB 6281 (18 bits) and USB 6356 (16 bits) DAQ device for the noise measurements. I found weird problems with USB 6281 whereas USB 6356 works perfectly. The problem is: 

When I start measurement with USB 6281, the voltage level of the acquired signal keeps on decaying or growing. I have tried both differential and NRSE ground configuration and I have also tried supplying proper external ground source to the board. 

The same problem persists when I short the input port of the device. The NI USB 6356 with input port short gives the expected ADC noise floor whereas USB 6281 level is not at all stable (decays or grows). I am using same VI for data acquisition for both 6281 and 6356 devices. 

For an exact explanation of the problem, I have attached a video and photo describing the problem. 

Is there any defect in the board. We have two USB 6281 devices and both show the same problem. What can be the possible reason for this problem? 

[DuncanW.]

Hi Amartansh13,

I took a look at the video you posted and it appears that you are reading from the 6356. However, the readings you are getting are not valid readings I would expect from any sensor. You are getting small negative voltages that usually indicate you have an incorrectly wired input. Maybe I am confused as to what your video is demonstrating, but it seems to me that your 6356 device is not working either.

Would you mind uploading your code? It would help us identify any problems you might be experiencing that are caused by your DAQmx configuration.

Best,

Duncan W.

[RTSLVU]

Looks like noise to me.

Take any oscilloscope and lay a probe on the bench, turn the vertical scale down to 10mV/Div and look at the screen.

What does it look like if you short the input leads together?

[Amartansh13]

Sorry for the confusion. The problem is with the USB 6281 only. I am using the same vi for noise measurement for both 6281 and 6356, so it doesn't matter what board you select in the vi. You can look at the vi attached (Note: There are two cases in the case structure in vi, one for 6356 and one for 6281, but I am using same case (i.e. 6356) for measuring with both 6281 and 6356. So, kindly ignore the code corresponding to 6281. ).

The data I am getting is not from the transducer. In the experiment shown in the video, nothing is connected to the input port of the 6281. But I have tried with the transducer as well with the ground source (ground terminal of the large oscilloscope). I have also short the input port of 6281 with the device ground but in every case, I can see the decay or grow. I have not attached videos for all the measurements but all types of measurements show the similar problem, i.e. the voltage level keeps on decaying or growing. 

For 6356, similar experiments give nice voltage level centered around 0 V or approximately constant mean. Most importantly, when input port of the 6356 is short with metal cap, the resultant noise measurement is matching the ADC noise floor. Whereas similar experiment with 6281 gives decay problem. (Note: With the decaying wave, ADC noise variance is correct but mean changes (decays) continuously)

[DuncanW.]

I was able to use your code in conjunction with a MAX test panel to read a signal. While messy, your code also doesn't have any obvious errors. I think we can say pretty safely at this point that the issue is in the way you are attaching your sensor to the 6281.

What sensor are you using? How are you attaching it to the 6281? These questions will help us discover the proper configuration for your system. You are correct that floating inputs are typically caused by ungrounded signals or ground loops in your system. Would you mind explaining your grounding decisions and motivations for each piece of your system? It is pretty easy to accidentally create a ground loop so I think that describing and defending each of these decisions may help you find any errors. Also, it seems like you have taken the shotgun approach of trying each input configuration so far, but I think it would be better to analyze the problem and identify which configurations it needs, rather than just which configuration works.

Duncan W.

[Amartansh13]

Thanks a lot for your reply. 

I can provide you the transducer connection details tomorrow. 

But I wanna again tell you that the same problem exists even when I am shorting the ai0 and ground, i.e. decaying level. Ideally, if input port is short with the device ground, it should only measure the ADC quantization noise. With 6356, shorted input port gives exactly the same quantization noise levels with mean= 0V. But with 6281, the quantization noise variance is constant but its level keeps on decaying or grwoing as shown in the video. 

So, even without even bringing tranducer or any complex connection in the scene, we have the same problem. 

[Amartansh13]

Thanks a lot for your reply. 

I can provide you the transducer connection details tomorrow. But I wanna again tell you that the same problem exists even when I am shorting the ai0 and ground, i.e. decaying level. Ideally, if input port is short with the device ground, it should only measure the ADC quantization noise. With 6356, shorted input port gives exactly the same quantization noise levels with mean= 0V. But with 6281, the quantization noise variance is constant but its level keeps on decaying or growing as shown in the video. 

So, even without even bringing transducer or any complex connection in the scene, we have the same problem. 

[@wadhesh]

Hi,

Did you solve the problem? I am also having the same kind of issue with USB 6281 DAC.