[t.teschner]

I am trying to create a new variable that calculates the skin friction coefficient along the flat plate example.

I am following the NASA flat plate domain definition (https://turbmodels.larc.nasa.gov/flatplate.html) and after my simulation, I am trying to calculate the skin friction coefficient based on empirical findings (first equation on this page: https://www.cfd-online.com/Wiki/Skin...on_coefficient) for comparison.

The equation I am using in Data -> Alter -> Specify Equation ... is as follows:

V13 = 0.0576 * (ABS(69.4 * 1.225 * {CoordinateX} / 0.000017)) ** (-0.2)

This gives me an error saying that "Overflow or underflow in power operation". I can change the above to

V13 = 0.0576 / ((ABS(69.4 * 1.225 * {CoordinateX} / 0.000017)) ** (0.2))

which gives a similar error "Overflow or underflow in divide operation". Anyone came across something similar and knows how to solve that? I did something similar in Paraview without issues.

[Brandon_CFD]

Hello Tom-Robin,


It seems that you're getting a divide by zero, which can be resolved using the "IgnoreDivideByZero" parameter in $!AlterData. This requires that you use a macro, since this isn't available in the GUI. For example:
#!MC 1410
$!AlterData
Equation = 'V13 = 0.0576 * (ABS(69.4 * 1.225 * X / 0.000017)) ** (-0.2)'
IGNOREDIVIDEBYZERO = YES

[t.teschner]

Originally Posted by Brandon_CFD

Hello Tom-Robin,


It seems that you're getting a divide by zero, which can be resolved using the "IgnoreDivideByZero" parameter in $!AlterData. This requires that you use a macro, since this isn't available in the GUI. For example:
#!MC 1410
$!AlterData
Equation = 'V13 = 0.0576 * (ABS(69.4 * 1.225 * X / 0.000017)) ** (-0.2)'
IGNOREDIVIDEBYZERO = YES

Thanks for that, I figured that there was an issue with division by zero but was missing your MACRO knowledge. Appreciate your help!

[dsoder]

This may be a trivial solution and a hack, but in some cases you may be able to bypass a divide-by-zero error simply by adding a negligible number somewhere in the denominator such that the denominator won't equal zero. This method isn't pretty, but it allows you to just use the GUI rather than creating a macro.

Example:
{speed_of_sound}={velocity_magnitude}/({Mach}+.00000001)

Your case:
V13 = 0.0576 / ((ABS(69.4 * 1.225 * ({CoordinateX}+.00000001) / 0.000017)) ** (0.2))

Hope this helps!

[t.teschner]

I guess that would be the more CFD-like answer. Yes, that would do the job as well, will keep this in mind, thanks for pointing that out.