LabVIEW

cancel
Showing results for 
Search instead for 
Did you mean: 

Linearity - Potentiometer standard deviation

Solved!
Go to solution

 

I am testing this 2 potentiometers and I want to determine whether or not the pot is linear.
The data collected on a good graphs look like this.
 
Where the RED line is the measured value and the GREEN line is the tolerance.
If the RED line is out of the GREEN line than these POT fails.
Attached below is a curve for linearity that I think will work with what I am doing?
If somehow we can make the GREEN LINES as wide at the end would fit into the model below.

 

good graphs.jpgGoodbad graphs.jpgBadzoomed of bad graph.pngZoomed of bad

The current program is not acturate and would always fail good units. It will pass today but fails tomorrow. The consistently of passing is not there. I think I need to open up the limits or find another way to determine pass or fail.

Do you think you can help?

 

Vi.PNG

 

Best regards,
Krispiekream
0 Kudos
Message 1 of 7
(1,158 Views)
Solution
Accepted by topic author krispiekream

Hi kream,

 

I want to determine whether or not the pot is linear.

How do you define "linearity" with your pots?

I would define this as function of R_pot/R_full vs. slider position…

 

If somehow we can make the GREEN LINES as wide at the end would fit into the model below.

That's easy: either increase your LinTol value - or even set the tolerance to MAX(LinTol*value; FixedMinValue)…

 

I did change your core routine to:

check.png

Why do you need loops for polymorphic functions?

Best regards,
GerdW
CLAD expired, using LV2011SP1 + LV2017 (+LV2019 sometimes) on Win7+cRIO
Kudos are welcome Smiley Wink

Message 2 of 7
(1,127 Views)
Highlighted

I would define this as function of R_pot/R_full vs. slider position…

How would I do something like that? Do you have an example?

 

I increased the tolerance to 0.25 and its seems to be okay but I don't know if its accurate or not.

pot.PNG

 

v2.PNG

 

 

How would I implement something like this

vs what it is currently is.

 

deviation.png

 

Best regards,
Krispiekream
0 Kudos
Message 3 of 7
(1,097 Views)

Hi kream,

 

I don't know if its accurate or not.

According to your images you need to expect ±0.2% of non-linearity (type 3500), with respect to resistance range. For a 100kOhm resistor you need to expect ±200Ohm non-linearity…

 

How would I do something like that?

You need to know the slider position to know the expected resistance value. Only then you can compare with real/measured resistance and calculate the non-linearity…

Best regards,
GerdW
CLAD expired, using LV2011SP1 + LV2017 (+LV2019 sometimes) on Win7+cRIO
Kudos are welcome Smiley Wink

Message 4 of 7
(1,088 Views)

Hi

I have these data I have collected from the tests.

Pot Datasheets.pngPot Deviation.pngpot.PNG

 

I am not familiar with linear fit.vi

or its function.

The current code is what I see. But I don't understand how the code works.

It is currently in use for production now but I think its not accurate and doesn't meet the criteria.

LINEARITY of a 10K pot should be 20 ohms or 0.20% error.

Can someone help me look at my code and see where it can improve?

Thanks

 

 

 

Best regards,
Krispiekream
0 Kudos
Message 5 of 7
(1,017 Views)

Hi krispiekream,

 

is there a reason you need to put your measurement data into an array of cluster?

check.png

You use an arbitrary X axis for your fit, but I told you once you should use the current position of your pot(s) instead…

 

I'm really tempted to merge all your threads about this "linearity" topic! (And I did for this reference…)

Best regards,
GerdW
CLAD expired, using LV2011SP1 + LV2017 (+LV2019 sometimes) on Win7+cRIO
Kudos are welcome Smiley Wink

Message 6 of 7
(994 Views)

thanks for your help.

I will look into the current position of the pot instead.

I didn't get that last time.

 

Best regards,
Krispiekream
0 Kudos
Message 7 of 7
(980 Views)