Maximum Angular deviation permitted in directions other than spanwise

Johan M · April 4, 2022, 10:21

Dear all,

I had seen omnis gave recommended ranges for different mesh criteria to identify the degree of mesh quality, seen attached. It gives a range for the maximum spanwise angular deviation but does not recommend a range for the overall max angular deviation (in directions besides spanwise). I would like to know if even the overall angular deviation should be < 40 degrees, or if it can exceed that value as long as the spanwise angular deviation is < 40 degrees.

Kind regards,
Johan

colinda1 · April 11, 2022, 03:58

The span angular deviation is used for detecting issues in the mesh generation process. AutoGrid generates the 3D mesh by stacking the different blade-to-blade meshes from the different flow paths. In some cases (due to an issue linked to the computation of the intersection of the blade with the flow paths, a difference of convergence of the blade-to-blade mesh smoother,...), it can happen that two adjacent mesh layers are not the same leading to a discontinuity in the 3D mesh. These discontinuities can be detected by the span angular deviation criteria and generally have a bad effect on the orthogonality criteria and that's why we recommend to remove them.

I hope this explains why the angular deviation is relevant in span direction and not in the blade-to-blade directions?

Best regards,
Colinda

Johan M · April 11, 2022, 10:29

Dear Colinda,

Thank you for the response! Although I admit I am still trying to understand how the dm/r-θ plane plotting actually works as it seems abit abstract to picture for me, I do roughly grasp the idea of wrapping 2-d blade to blade meshes on each spanwise meridional flow path to create a the 3-d blade from your explanation. In that sense I can sort of picture how the importance lies in the spanwise direction regarding angular deviation.

To confirm, is adjacent mesh layers referring to meridional flow paths that lie along the span?

Kind regards
Johan

colinda1 · April 11, 2022, 10:39

Dear Johan,

Indeed, with adjacent mesh layers we mean the blade-to-blade mesh on adjacent flow paths.

Best regards,
Colinda

Johan M · April 11, 2022, 12:09

Dear Colinda,

Thank you for the confirmation and help

Kind regards
Johan

April 11, 2022, 03:58	Span angular deviation	#2
colinda1 Senior Member Colinda Join Date: May 2012 Location: Brussels Posts: 153 Rep Power: 14	The span angular deviation is used for detecting issues in the mesh generation process. AutoGrid generates the 3D mesh by stacking the different blade-to-blade meshes from the different flow paths. In some cases (due to an issue linked to the computation of the intersection of the blade with the flow paths, a difference of convergence of the blade-to-blade mesh smoother,...), it can happen that two adjacent mesh layers are not the same leading to a discontinuity in the 3D mesh. These discontinuities can be detected by the span angular deviation criteria and generally have a bad effect on the orthogonality criteria and that's why we recommend to remove them. I hope this explains why the angular deviation is relevant in span direction and not in the blade-to-blade directions? Best regards, Colinda Turboheld likes this.

April 11, 2022, 12:09		#5
Johan M Member Johan M Join Date: May 2021 Posts: 36 Rep Power: 5	Dear Colinda, Thank you for the confirmation and help Kind regards Johan colinda1 likes this.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
loading issue libraries:libsimpleFunctionObjects.so libsimpleSwakFunctionObjects.so	immortality	OpenFOAM Pre-Processing	15	April 7, 2024 13:35
Spanwise Angle deviation greater than 40	allenjohngeorge	Fidelity CFD	3	December 30, 2018 02:11
OpenFoam "Permission denied" and "command not found" problems.	iyidaniel@yahoo.co.uk	OpenFOAM Running, Solving & CFD	11	January 2, 2018 07:47
Diverging solution in transonicMRFDyMFoam	tsalter	OpenFOAM Running, Solving & CFD	30	July 7, 2014 07:20
compressible flow in turbocharger	riesotto	OpenFOAM	50	May 26, 2014 02:47

April 11, 2022, 10:29		#3
Johan M Member Johan M Join Date: May 2021 Posts: 36 Rep Power: 5	Dear Colinda, Thank you for the response! Although I admit I am still trying to understand how the dm/r-θ plane plotting actually works as it seems abit abstract to picture for me, I do roughly grasp the idea of wrapping 2-d blade to blade meshes on each spanwise meridional flow path to create a the 3-d blade from your explanation. In that sense I can sort of picture how the importance lies in the spanwise direction regarding angular deviation. To confirm, is adjacent mesh layers referring to meridional flow paths that lie along the span? Kind regards Johan

April 11, 2022, 10:39		#4
colinda1 Senior Member Colinda Join Date: May 2012 Location: Brussels Posts: 153 Rep Power: 14	Dear Johan, Indeed, with adjacent mesh layers we mean the blade-to-blade mesh on adjacent flow paths. Best regards, Colinda