CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   CFX (http://www.cfd-online.com/Forums/cfx/)
-   -   unexpected range in pressure level CFD-Post (http://www.cfd-online.com/Forums/cfx/84506-unexpected-range-pressure-level-cfd-post.html)

stefank February 1, 2011 04:51

unexpected range in pressure level CFD-Post
 
2 Attachment(s)
Hi,

I am currently simulating flow between two clutch discs. (one flat stationary disc and one rotation (around x) with a complex geometry as seen in the pictures below)
When I want to evaluate the pressure in the flow, I am wondering, why there is such a wide range in the pressure level. In the global range I got values between -990 and +1100 MPa which is not reasonable. I expect maximal values around 20MPa, not more. Reference pressure is set to 1 atm. The pictures show two versions of a streamline. One with a global range for pressure and one with a user specified range.
Attachment 6298
Attachment 6297

Actually the picture with the user specified range is quite reasonable. But another point which makes me wondering, is, that the torque around x which fluid creates on the solid bodies is around 40 Nm. (inner radius is 59mm)
Does anybody of you have some suggestions why the pressure range is so wide? And how I should interpret these values?
Why is the torque value so big? I expect a value between 0 and 1 Nm.
I hope somebody could help me!
Thanks!

joey2007 February 1, 2011 17:06

Is it just the suggested range or the result?

ghorrocks February 1, 2011 18:46

Firstly, have you done the basic checks that your simulation is accurate?
http://www.cfd-online.com/Wiki/Ansys..._inaccurate.3F

After you have checked the accuracy of the model you may still be getting regions of impossibly low pressure. This is common for incompressible models as you do not have a cavitation model, and in reality a small cavitation bubble will limit the negative pressures. If you do an isosurface around the negative absolute pressures you will probably find it is a tiny region which in reality would have a small cavitation bubble in it. But whether it is worth modelling the cavitation is up to the requirements of the model - often it does not contribute significantly to the flow.

stefank February 2, 2011 04:35

2 Attachment(s)
Hi, thanks for replying!

@joey2007: Actually it is both. The range shows that there are total pressure values between -990 and +1100 MPa. I do not know how to interpret these values. The change of prefixes confuses me. My comprehension is that I can have lower pressure which are less than the atmospheric pressure (but >0) and higher pressure which are higher than 1 bar. But how should I explain the change in prefixes?
For example when I change the range in one of the pictures, you can see that there are definitly positive and negative pressures:
Attachment 6310

@ghorrocks: As far as I can measure it, I think that the accuracy of the model should be satisfing. I am concerning about one point. The RMS P-Mass is not getting under the default value of 1.0e-04. After 100 steps it finishes around 8.0e-03. Values of U, V, W do not make problems at all (end at 1.0e-05). May this indicate a reason for the pressure values? Should I use another advection scheme? Currently I am solving with High Resolution.

Ok, caviation could explain the highly negativ pressure values. But altough I still do not understand why I have the change in prefixes. As you can see in the following picture, I defined a volume which includes all regions with values <0 MPa. This volume contains half of the fluid geometry.
How can I define a isosurface which contains the whole range of negative values? Normally you can only define one value on an isosurface, right?
Attachment 6311

A answer would be highly appreciated!
Thanks again fo replying!

ghorrocks February 2, 2011 06:27

Based on your comments about convergence I doubt you have done any of the checks recommended in the link I gave. You are wasting everybody's time (including your own) by continuing with an inaccurate model. Might as well just use a random number generator.

Many of the tutorials show how to generate an iso-surface. Do some tutorials.


All times are GMT -4. The time now is 21:54.