digtal period measurment, measured periode time for some samples drops 50% or near 0.

[ApaX]

Hi All,

I have been struggling with this annoying problem for a while. In my measurement setup I have a flow meter that will produce a block wave. The signal low is about 0.4 volt, and the high is about 4.6 volt. The duty cycle of this signal is about 50%.

I created a VI that will measure the period of 10 (with a counter input) samples and calculate the average. The problem is this. Every once in a while some period time pops up that is only half of the rest. And some time the period time seems to drop to 5e-7 (which is lower than the minimum the I set).

If I investigate the signal on the scope it looks stable. This is the weird think, if I replace the flow meter by a signal generator it is working flawlessly.

Currently I think there must be something wrong with the signal,  but there is no evidence to prove it. Everything seems to be in order.

Some suggestions and advice is needed…:D

Kind regards,

[SLee2004]

Have you tried checking the signal on an analog input channel without the scope attached just to see what the raw signal looks like? Most scope meters and function generators are grounded properly and have better noise filtering capability. When you measure the signal in parallel with a scope meter, it probably cleaned up the signal.  I found that NI cards pick up noise easily with my experience. You may need to add a resistor between the (-) pin and the ground pin so that there is a return path to ground, as it did solve my problem (2 ohms). See http://www.ni.com/white-paper/3344/en

It is better if you can keep the flow meter cable short, since longer cables pick up noise easily.

Make sure that your signal is not routed with a high power voltage, as the signal will pickup the noise from the power source.