Vector Projection onto Tangent of Polyline

u2berggeist · May 17, 2017, 11:28

So I am try to calculate the percentage area of flow reversal on the blades of a centrifugal compressor (negative Z axis being the axis of rotation with flow moving in positive direction) in CFD Post. The current way I am doing it is to do a countTrue of wall shear being positive in the Z direction and dividing that by the count of the surface. (While ideally I'd like to do this by area instead of by node, node works fine since I'm mostly concerned with the exisitence of reversed flow, not necessarily the exact amount of reversal.)

This works pretty well on the suction side of the blade, but not the pressure side, where flow is mostly being moved in the radial direction, with some of that flow having a negative Z component. What I'd like to do is have the directional component of wall shear where the direction is determined by a polyline (ie, the shroud contour of the compressor).

Basically, I want to have a variable equal to:
$\vec{Wall Shear Vector}\bullet \vec{Directional Vector}$
where directional vector is determined by the tangential of a polyline. The tangential of the polyline can simply be the tangential vector of the polyline at the same Z height as the point where the wall shear is being requested.

Is this possible? Seems like a simple thing to do, but I can't figure out how to do it in CFX Post. I am using ANSYS WB 17.2

Thanks,

ghorrocks · May 17, 2017, 21:58

You will need to write the dot product out as its components.

The boundary normal variable is available on polylines. Not sure if this will give you what you want, you will have to try it.

The wall shear is available as the components of the vector so that one is fine.

u2berggeist · May 17, 2017, 23:27

Does using * on two vector variables not do a dot product? Results I've done using this have worked in the past...

I saw that polylines had the normal components, but I wasn't sure if there was a way to get the tangential components of the polyline (either by mathematical trickery or a hidden function).

I guess I could transform the normal components into tangential since I know the plane in which the polyline lies. I was just hoping there was a way to directly get the tangential components as that would be more robust of a method.

Thanks!

ghorrocks · May 18, 2017, 06:54

Quote:

Does using * on two vector variables not do a dot product?

I have never tried. If it seems to work then great.

Quote:

I was just hoping there was a way to directly get the tangential components as that would be more robust of a method.

Yes, it is not ideal. But if you can work with it then your problem is resolved.

May 17, 2017, 11:28	Vector Projection onto Tangent of Polyline	#1
u2berggeist Member James Wright Join Date: Oct 2015 Posts: 40 Rep Power: 11	So I am try to calculate the percentage area of flow reversal on the blades of a centrifugal compressor (negative Z axis being the axis of rotation with flow moving in positive direction) in CFD Post. The current way I am doing it is to do a countTrue of wall shear being positive in the Z direction and dividing that by the count of the surface. (While ideally I'd like to do this by area instead of by node, node works fine since I'm mostly concerned with the exisitence of reversed flow, not necessarily the exact amount of reversal.) This works pretty well on the suction side of the blade, but not the pressure side, where flow is mostly being moved in the radial direction, with some of that flow having a negative Z component. What I'd like to do is have the directional component of wall shear where the direction is determined by a polyline (ie, the shroud contour of the compressor). Basically, I want to have a variable equal to: $\vec{Wall Shear Vector}\bullet \vec{Directional Vector}$ where directional vector is determined by the tangential of a polyline. The tangential of the polyline can simply be the tangential vector of the polyline at the same Z height as the point where the wall shear is being requested. Is this possible? Seems like a simple thing to do, but I can't figure out how to do it in CFX Post. I am using ANSYS WB 17.2 Thanks,

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
CFD-Post: Polyline Export at Integration Points	minger	ANSYS	4	February 22, 2014 10:23
Polyline at intersection of two planes	Francesco	CFX	2	July 29, 2009 09:03
Help Needed: Polyline intersection in Post	PHabib	CFX	4	July 13, 2009 05:04
velocity on surface and Polyline	Alex	CFX	2	September 26, 2006 02:31
plot variable against length of polyline	Francesco	CFX	0	April 4, 2006 12:49

May 17, 2017, 21:58		#2
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,872 Rep Power: 144	You will need to write the dot product out as its components. The boundary normal variable is available on polylines. Not sure if this will give you what you want, you will have to try it. The wall shear is available as the components of the vector so that one is fine.

May 17, 2017, 23:27		#3
u2berggeist Member James Wright Join Date: Oct 2015 Posts: 40 Rep Power: 11	Does using * on two vector variables not do a dot product? Results I've done using this have worked in the past... I saw that polylines had the normal components, but I wasn't sure if there was a way to get the tangential components of the polyline (either by mathematical trickery or a hidden function). I guess I could transform the normal components into tangential since I know the plane in which the polyline lies. I was just hoping there was a way to directly get the tangential components as that would be more robust of a method. Thanks!