Analog waveform measurements

[Mathan]

Dear all,

From the input waveform i got the envelope of the signal and now i need to measure some test values according to the diagram shown below:

I got the Y-Coordinate of the value 'a', but how do i get the value of x for that? After finding this how to position the cursor on the value of (1+a)/2, so that i can now find the value of t2 and so on. Any inputs are very appreciated. Thanks a lot in advance. I have attached the vi.

Mathan 

[SanRac]

Hi Mathan,

You can use the "Transition Measurement.vi"

Please find attached snapshot for calculating x value.

You can play around with this VI and it also help you  to calcualte t2

Hope this helps

Regards

Santosh

[craigc]

To get the x value in a 1D array when you already have the Y value, you can use (With some caution if there are multiple Y values within range) the 1D threshold array vi under the array pallete.

To find the value of (1+a)/2 you need to find point 1 then divide the |(1-a)| value by 2 take that Slice of data and again use the 1D threshold array.vi to find its x value.

To find the value at 1; in the past i have used the point by point differentiaition vi to get the dy/dx or gradient at any given point along the curve.  When it goes past a value i.e drops off (Gradient Increases above say |0.05|) and goes below a threshold (y value) you can find point 1.

Craig

Message Edited by craigc on 12-23-2009 05:13 AM

[Mathan]

Dear Santosh,
I tried as you said but since the area of interest is at 'a' which is 0.16 (Y value), transition measurements.vi is popping out the -20310 error meaning the waveform does not have enough edges to perform the measurement.

Dear Craig,
I tried with threshold array vi but its displaying 0 only for my point (a = 0.16).

[SanRac]

Dear Mathan,

This error will come only when your input data has not enough cycles,

or

You can try the setting shown in attached snapshot.

In your waveform there is no rising edge that comes to 0.16 value.

You can also assign the edge number (Actually Cycle number) on you would like to perform the transition measurement.

In transition measurement vi change the edge number to 2 and see the change in timing 

Regards

Santosh

Message Edited by SanRac on 12-23-2009 05:58 AM

Message Edited by SanRac on 12-23-2009 06:02 AM

[craigc]

Hi Mathan,

I take it you set the threshold up correctly? if it never crosses the threshold 1 think you will get 0 out..  Do you know what Samples / x values point a is likely to be between?

My approach would be to threshold the array twice at a point just before a and just after.  take that data out and look for the point of inflection in the data.  Point of inflection is where the gradient changes sign or goes through 0.  In your case a gradient change would go -ve to +ve passing through zero for the top part of the envolope and vice versa for the bottom.

Craig

[Mathan]

Dear Santosh,

Thanks for looking into this. That works like a charm. Yes i did changed the edge number to 2 and noted the change in timings but we require only the first edge and it is OK.

Dear Craig,

Am very sorry. I did not understand your points. Can you please help me with a simple example? Thanks a lot.

Let me proceed into the calculation of t values and ill let you know once i finish.

Mathan 

[aeastet]

DId you ever get this complete? I was gone for Christmas break when I saw the request on the other thread.

[Mathan]

Dear aeastet,

Thanks for looking in to this.

Yes. Santosh and Craig showed me a wonderful path to achieve what i want. I managed to complete the work with their timely help. And so i have done 2/3 of my work in the sense out of 3 different waveform measurements, i completed 2 waveform processing and there is one more waveform bit tougher one needs to be manipulated. Ill capture the waveform from the scope and attach the data in this thread. I surely will expect your help in solving this.

BTW how is Christmas and New Year celebrations?

Belated Happy New Year wishes to You, Santosh and Craig. Thanks a lot everybody.

Thanks and Regards,

Mathan

[SanRac]

Hi Mathan,

Wish you the same.

I was also curious about the results.

Thanks for the update.

Regards

Santosh