How to deal with the large measuring noise when measuring conductors with USB-2408?

[Benny]

My MCC measurement is USB-2408 whose P/N and S/N is 151480B-01L and 1BFC892 respectively. The measurement was carried out on my IPC (laptop PC was also tested). A K-typed thermal couple was used. However, I found that the signal (℃) waves intensively and the peak-valley value excess 4 ℃ when measuring conductors such as my hands. It shows a very fine performance in measuring non-conductor with a signal waving within the range of 1℃. I've already tried to change PCs or softwares such as DAQami and DASYLab, to calibrate the hardwares, but it didn't help. My PC versions are all win10 and I am located in China. 

I also uploaded some images. The one named "2channels" shows a comparison between measuring a conductor and a non-conductor. The one named "2408-1" shows the hardware settings. The one named "compar" shows the changes when suddenly to measure my hands. And the one named "DAQsetting" shows the setting in the software DAQmi.

I hope someone can help me to deal with it or contact me by email qd007@cqut.edu.cn. Thanks a lot. 

[JRys]

Hello,

I am able to duplicate your results with the Data Rate property at 100 Hz. Here in the USA, our power grid runs at 60 Hz. When I switched to 60 Hz it greatly improved the data. I believe China runs on 50 Hz so switch to 50 Hz for both channels. 

Also, if you want to touch the thermocouples please use a grounding strap. This is because your hands have a great deal of voltage and too much will damage the input. A ground strap should keep it to a minimum.

Best regards,
John

[Benny]

Hi, John,

Thanks for your help. I tried to turn the sampling rate to 50Hz, and it did work. Can the device be used only at the sampling rate of 50Hz, if it is strongly affected by the voltage frequency?

Thanks for your help, again.

Best regards,

Benny

[JRys]

Hi Benny,

The data rate setting sets the level of oversampling, which is similar to averaging but not the same. When set to 3750 the maximum sample rate is 1102 samples per second. However, at this rate the data is less accurate. Everyone seems to get data rate and sample rate mixed up. To determine your maximum sample rate use the following formula from page 13 in the user manual. www.mccdaq.com/pdfs/manuals/USB-2408.pdf

If using two thermocouple channels, both set to 50 Hz, the maximum sample rate is 1 / (2 * 1 / (0.02 + 640uS)) or 24.22 samples per second.

The thermocouple data rate must be set to 50 Hz to produce sufficient accuracy and to eliminate noise. 

Best regards,

John

[Benny]

Hi John, 

Thanks for your help, now I know how to set the sampling rate.

Best regards,

Benny