DAQ question

[lebogzy]

Hey guys hope someone can help me with this, or atleast send me in the right direction....

I have an output from both channels of my oscilliscope on my screen, working fine.

However, I do not need to stare at these waveforms - what I actually need to do is monitor only 4 points on each of them over time and see how they change during the course of my test. So from two channels, I need to generate 8 seperate graphs where x is time and y is the voltage level of these points.

I have attached a screenshot that i edited in paint, to show exactly which points I care about - signified by red circles. Please take a look, and shoot me some ideas as to how to do this. If you need me to clarify further, i would be more than happy to.

Thanks for reading this, peace.

Message Edited by lebogzy on 05-07-2009 02:44 PM

[CoastalMaineBird]

What defines those points?

I would GUESS that it's the mid-point of each step (midway between transitions).

But that's a guess. 

Do you need to find them, or do you need to set a knob at 19 mSec and always watch that point?

You'll have to better describe what you need. 

[lebogzy]

Thanks for the reply. 

I do NOT need to find them, they will always be at the same time points because of the way my scope and trigger are set up.

[CoastalMaineBird · Accepted solution]

OK, I'll assume that the scale is really in time (mSec).

You want to pick out points at T = 19 mSec,  T = 56 mSec, T = 84 mSec, and T = 122 mSec (judging from your dots on the graph). 

I'll assume you have 50 mSec of pre-trigger data.

So, your points are at T = 69 mSec, T = 106 mSec, T = 134 mSec, and T = 172 mSec FROM THE BEGINNING of the record.

I'd guess you were sampling at 1000 Hz. (just eyeballing it).

So, convert your times to sample numbers:

S = .069 * 1000 = 69;  S = .106 * 1000 = 106; S = 0.134 * 1000 = 134, and S = 0.172 * 1000 = 172 FROM THE BEGINNING.

Use an INDEX ARRAY, and stretch it to four terminals.

Use your CHANNEL 1 as the input array for INDEX ARRAY. 

Wire four constants of 69, 106, 134, and 172 to the inputs, and wire the four outputs to  four charts (the CHART will remember history for you).

This picks out the signal at those four points.

Repeat for the other channel with four more charts.

Or, you could bundle the signals and have four plots on one chart. 

Message Edited by CoastalMaineBird on 05-07-2009 03:12 PM

[lebogzy]

booya. thanks!

[CoastalMaineBird]

Hit the KUDOS button while you're in a good mood 😉

For a more general solution, find the points with code.

Run the CHAN 1 signal through a differentiation and you'll see a big negative spike at T = 0 and T = 135.

YOu'll see positive (I assume) spikes at the inner transition points.

Calculate from those where the midpoints are. 

Message Edited by CoastalMaineBird on 05-07-2009 03:53 PM