Thread Subject: determening frequency or peaks

I haven't used Simulink with DAQ cards before, so please accept my apologies if this is a very rudimentary question.

I am collecting the analog response of a paddle wheel flow meter. The device indicates the flow speed through both frequency and magnitude (higher freq and higher mag sinusoidal signal corresponding to flow of greater magnitude). I can collect the signal easily enough. However, I am having some problems converting this signal to a real world flow value. I am able to see the sinusoidal response, but I can't record the peaks vs time or determine the instantaneous frequency.

Eventually I want to be able to integrate the flow values in order to send a signal to my system after a certain amoung of fluid has passed through the flow meter. However, after looking at both the peak finder and the input counter blocks, I can't seem to get either to work to obtain some sort of flow magnitude vs time response. Any assistance (in example form if possible) would be greatly appreciated!

"Alex Pedchenko" <alexander.v.pedchenko@vanderbilt.edu> wrote in message <gua2u1$gj3$1@fred.mathworks.com>...
> I haven't used Simulink with DAQ cards before, so please accept my apologies if this is a very rudimentary question.
>
> I am collecting the analog response of a paddle wheel flow meter. The device indicates the flow speed through both frequency and magnitude (higher freq and higher mag sinusoidal signal corresponding to flow of greater magnitude). I can collect the signal easily enough. However, I am having some problems converting this signal to a real world flow value. I am able to see the sinusoidal response, but I can't record the peaks vs time or determine the instantaneous frequency.
>
> Eventually I want to be able to integrate the flow values in order to send a signal to my system after a certain amoung of fluid has passed through the flow meter. However, after looking at both the peak finder and the input counter blocks, I can't seem to get either to work to obtain some sort of flow magnitude vs time response. Any assistance (in example form if possible) would be greatly appreciated!

It sounds like you need to detect the envelope of your signal. The following demo shows two ways of doing this (it requires the Signal Processing Blockset):
http://www.mathworks.com/products/demos/shipping/dspblks/dspenvdet.html

Alternatively, you could just chop the signal into frames (using the Buffer block), calculate the RMS voltage for each frame, and rescale into flow units.

Gautam