Stepped Voltage Check

[Pablop]

I this particular application, I actually don't need the transition crucially, I am really just after amplitudes after all.

However, generally when I am testing signals like this (usually TTL square waves, sometimes mV level analog waves), I need to locate transitions so that I can accurately measure the width of each on/off state. This is required to determine symmetry and phase for each cycle.

Not sure if that is what you meant by "dynamic", but I need the width of each and every on/off section, I do this by finding the transitions.

[aeastet]

Try this. It out puts an array of points that match both sides of the up and down parts of your signal. I know it is a little messy and maybe you woun't even use it but it was fun seening if I could get it to do what you want. If you use the points you should be able to calculate anything you need.

[aeastet]

This one is easier to read

[aeastet]

Set the peak limit to .02 and the point apper. I should have made this the default value. Just having a little fun. Hope this helps.

[Pablop]

Ack!

aeastet,

You're on LV9 and I'm still 8.6

Where are my new DVD's! 🙂

[aeastet]

Here is 8.6

[Pablop]

Instersting approach!

Looks like the filter is defintely needed.

In my applications I vary the sample rate and number of points aquired depending on the frequency of the signal I'm analysing. Some of the signals are 500kHz, som are 400hz. This means the array coming in will have quite a variation in number of samples, from 25k to 2 million. Filters tend to bog down on large sample sizes, so that might be something I would have to toy with. Also, there are a few control settings in this VI that would probably need to change as a function of the above,fs, Peak Limit, the range settings on the Min/Max, etc.

But thanks for sharing, I'll have to keep looking at this until I get it ;).

[aeastet]

I wasn't sure what your final data would be. I just thought it was a good way to look for changes. I messed around a little with different data sets and it seem to be somewhat flexible as long as the magnitude of the data doesn't change too much. I had figured that is was an idea starter and not the final code. I was expecting that you might make a few more thing controllable. I do not think the frequency on the filter matters too much. It seemed that the lower I went toward 1 for the frequency the better the data looked. Let me know it you found a way to make it work.