Calculating hydrodynamic lifting

Numerical methods and mathematical models of Elmer
Post Reply
thomasatelmer
Posts: 47
Joined: 18 Jan 2019, 18:18
Antispam: Yes

Calculating hydrodynamic lifting

Post by thomasatelmer »

Hello dear Elmer users,
for a couple of weeks now, I am fascinated by the Elmer Suite of solvers and their possibilities. My initial use was a 3D magnetostatic calculation, that worked right away - great. Propelled by this success, I am trying to make use of Elmer for calculating hydrodynamic lift of a fluid sheet in a narrowing gap, where one boundary (top of the gap) is moving and the other one (bottom of the gap) is stationary. By shear forces, the fluid is forced up the slope in the gap, creating pressure (and in reality, the possibility to lift the top face upwards). But the results are astonishing me: There is almost no pressure buildup, although the velocity field looks completely correct. It would be nice if someone with experience in similar disciplines could check if I have screwed up my simulation setup with a hidden mistake (or maybe an obvious one - I consider myself still a newbie :) ). Also, convergence is easily achieved in this Setup with a direct NS solver Setting - that is in my eyes "suspicious", I'd expect more problems with such low viscosity. Just the k-Epsilon solver does not converge... I am attaching the sif, a py script for geometry and a pictorial representation of a summary for your perusal.

Thank you for your Support!
Thomas
Attachments
sketch.txt
(9.58 KiB) Downloaded 241 times
Results_NS_w_kE.pdf
(141.97 KiB) Downloaded 258 times
case.sif
(2.98 KiB) Downloaded 240 times
raback
Site Admin
Posts: 4812
Joined: 22 Aug 2009, 11:57
Antispam: Yes
Location: Espoo, Finland
Contact:

Re: Calculating hydrodynamic lifting

Post by raback »

Hi

In BC3 set Velocity 2 to zero.

-Peter
thomasatelmer
Posts: 47
Joined: 18 Jan 2019, 18:18
Antispam: Yes

Re: Calculating hydrodynamic lifting

Post by thomasatelmer »

Hi Peter,
thank you for the Swift reply. Indeed, this was a major change, although it was just one component of one boundary condition. Convergence on turbulence is now better, and the pressure distribution now is totally different. I'll have to review tomorrow what I got in detail. I'll come back with more info, maybe questions :-).

Enjoy your evening,
Thomas.
thomasatelmer
Posts: 47
Joined: 18 Jan 2019, 18:18
Antispam: Yes

Re: Calculating hydrodynamic lifting

Post by thomasatelmer »

Hi Peter and others,

finally, I am able to Report out on my latest experiences for modelling hydrodynamic lift. Using the given hint, I was able to get my model running properly, and when this happened, I decided to make a small go-around to make sure I am on par with theory on my results. I modeled the System from https://www.comsol.de/paper/analysis-of ... ring-15253, which is a simple radial hydrodynamic bearing. The match is very good for my purposes, I get with the infinite bearing Approach from 2D about 8.2e7 Pa as Peak pressure, where the Reynold equation would dictate 7.5e7Pa for a Long Journal bearing.

So now after calculating the hydrodynamic lift for my Special case successfully, I am progressing on towards adding the influence of non-infinite bearing surface and elasticity of the surrounding componentry.
Attachments
elmer_result.png
elmer_result.png (31.87 KiB) Viewed 2433 times
Post Reply