Uneven numeric integration

[gdrag]

Just to clarify here, are you expecting the Original (dM/dt)/(H) to look like the Hysteresis loop M(H)?

Are they integrating the same exact data?

[HrvojeM]

Hi gdrag,

Original (dM/dt)/(H) graph looks ok, but Hysteresis loop M(H) graph should look something like this (red graph)

The integration is done only to get Hysteresis loop M(H). Original (dM/dt)/(H) doesn't need integration.

[gdrag]

Hi HrvojeM, 

I'm not sure why it looks different. I looked through the VIs and didn't see anything that stood out.

One thing I noticed, it seems you are still using uneven integration. Have you tried what gregoryj last suggested?

[HrvojeM]

Hi grag,

I've tried it but result is not even close to the hysteresis loop. I'm still using uneven integration because my field values (voltage) are not evenly spaced. Here is an example:

0,000

0,002

0,004

0,005

0,007

0,009

0,011

0,012 and so on.

The closest I've got to the hysteresis loop is by using Excell and using formula for uneven numeric integration (trapezoidal rule),

but the loop is overlapping and start and end point are not connected. That's why I tried to use LabVIEW's Uneven numeric integration.vi because I suspected that my formula is wrong or excel made rounding errors. Also I would like to automate the whole procedure so that the user only has to click the Run button in order to get hysteresis loop and some data from it. 

Thank you for your time,

HrvojeM.

[gdrag]

Hello HrvojeM,

I would make sure the correct data is being inputted into the Uneven Numeric Integration VI.  What you could do is set up a VI that takes your data from excel, feed that data into the Uneven Numeric Integration, and see if you get the correct result.   That will narrow down the issue to whether the problem is coming from the Uneven Numeric Integration VI or from data flow somewhere else in your VI.

Another thing you could do is open up the Uneven Integration VI and check specifically what happens to your data in there. If the formula is different than the one you want, you can potentially copy the code from Uneven Numeric Integration.VI, paste it in a blank VI, and modify it as you want to reflect the correct formula you want to use.  From I can tell, it uses a dot product to perform the integration.

[toshas]

It is possible to remove reorder function from library VI. So Uneven Numeric Integration will be able to calculate hysteresis loops.