Why are there oscillations overimposed on EKG signal?

[Mucha]

I hope my question isn't to sil1ly. I've already gone through most of the documentation about noise, signal wiring and double buffered data acqusition, but I didn't find any solution to my problem.

I use Lab-PC board connected through a long wire (aprx. 10 m) kardiomonitor to log EKG data and then to analyze it. The output signal of kardiomonitor doesn't exceed +/- 5V, it usually varies from aprox. -1 V to +3 V and as far as I'm concerned the output is isolated (I'm not 100 per cent sure about that).

I wrote an application in LabWindows/CVI based on Double Buffered example to log two channels of data (Measurment Encyclopedia URL http://zone.ni.com/devzone/explprog.nsf/6c163603265406328625682a006ed37d/9e2c3
685c3003a9386256480007fceaf/$FILE/DoubleBuffered.zip). To start acquisition I use NI-DAQ function Lab_ISCAN_Start, check and transfer the buffer I use DAQ_DB_Halfready, DAQ_DB_Transfer functions. The sampling rate is 100 Hz. There are not any errors during acquistion.

The problem is that there is sometimes a strange temporary sinusoidal signal overimposed on EKG (the second channel is floating - it's not connected - since I haven't still a scource of the second signal_. The amplitude of the noise varies and can ever exceed 0,5 V. I emphasize that the effect occurs only during short time periods (aprox. 10 s) and the vanishes.

Can anyone answer what the reason is? Is it connected with signal source, wiring, DAQ board or software? Is it possible that it's noise picked up from enviroment (it's very high).

[Ben]

I would guess it is coming from the environment.

Try moving the system some place away from other high field devices and try a small UPS.

Noise Reduction Techniques in Electronic Systems" by Ott is a wonderful text for addressing noise. He starts with Maxwell's equations and a hanful of material properties and derives rules of thumb for all types of noise problems.

Ben

[Filipe A.]

Hello;

Since you are working with such long cable, my guess is that extra signal in being induced by the environment. It probably is due to EMI.

This sort of problem usually doesn't have easy solutions. You can try some techniques to isolate the cable and terminal block a little better.

I'm attaching an App Note that talks about Noise and ways to isolate your system from it. Furthermore, you can try to insert a RC low pass filter right at the terminal block to filter out that noise.

Hope this helps.
Filipe A.
Applications Engineer
National Instruments