LabVIEW

cancel
Showing results for 
Search instead for 
Did you mean: 

Fitting transistor IV curves using Lev-Mar with Error Function

I have been working with the VIs from this thread:

http://forums.ni.com/t5/LabVIEW/Problem-Fitting-Lev-Mar-with-Error-Function/m-p/2716045

 

My modified VIs are attached.

 

It functions correctly for good data, but I'm confused how to modify the code so that I can fit low threshold data.  For example:

 

I've attached a fit for data taken at a 20mV charge injection & 5mV.  There are 1000 reads, & so 1000 hits means the transistor is reporting 100%.  For 20mV, below the threshold transition there's 100% hit. But for 5mV, the transistor is only marginally working below threshold, and so the fit calculates the 50% hit crossing point out of 750 hits instead of the real 1000.  

 

 

How can I modify the fit so that the range parameter doesn't change (1000 no matter what the data)?

Download All
0 Kudos
Message 1 of 3
(2,473 Views)

Hi bbentelin,

Have you tried turning off the autoscale for the y-axis? If you right click on your graph and click properties, you can go to the Scales tab. There is a dropdown list at the top of the page where you select your axis, so in your case the y-axis. Beneath it is an Autoscale checkbox, and if you unselect that and select the range you would like to appear, I think that should work.

 

Let me know if this is what you're looking for.

0 Kudos
Message 2 of 3
(2,409 Views)

No no, It's the fitting calculation I'm asking about. 

 

There are 3 parameters --- range, 50% crossing & slope.  I'd like to set the range as a constant so that the fitting calculation only varies the slope & crossing point.

0 Kudos
Message 3 of 3
(2,405 Views)