Effect of sampling rate on the same signal: Fundamental question

[sm81]

I want to know why having higher sampling rate, I see a  completely different signal.

I have attached the front panel graphs of the signals acquired by the same sensor at two different sampling rates which are:
 
250KS/s Continuous acquisition; 5000 Samples per loop


 
500KS/s Continuous acquisition; 5000 Samples per loop

Of course one with the lower sampling rate is how it should be, as can be confirmed by looking through an Oscilloscope.

Please explain.

[GregFreeman]

Can you post the images as pngs, not RTF files?

[sm81]

Attached

[johnsold]

What is the device you are using for the data acquisition?

What should the signal look like?

Lynn

[sm81]

NI PXI-6251

16-Bit, 1 MS/s (Multichannel), 1.25 MS/s (1-Channel), 16 Analog Inputs

The signal should look like how it looks for 250KS/s image. I would like to understand why would it look so different if I use 500KS/s.I am using two channels so I think I am ok to use upto 0.5MS/s with this device. 

[johnsold]

Do you have all the unused channel inputs tied to ground? What is the source impedance of the signal source? I am considering the possibility of ghosting.

Can you try some other sampling rates which are just slightly different, such as 255 kHz or 497 kHz?

Lynn