Inconsistent results - Axisymetric Free Jet

jrsilvio_ver · December 12, 2020, 18:17

Dear colleagues, good afternoon.

I am modeling an axysymmetric circular jet with a diameter of 0.1 m and within a region of height 1 m and length 15 m. However, the results regarding the average velocity in the x-direction and the Reynolds stresses are absurdly different from references in the literature.

Would anyone have any idea what to change in the simulation?

The case files are available at: https://www.dropbox.com/s/unsk4c91uj...alho4.rar?dl=0

Thank you all for the attention.

jrsilvio_ver · December 14, 2020, 00:57

Nobody???????

Tobermory · December 29, 2020, 06:08

Well, I had a quick look at your set up files, and the most obvious thing that stood out to me was the shape of your domain. It is very narrow in the cross-stream (y axis) direction: aspect ratio of length to width is 30:1, and this places your atmospheric pressure boundary very close to the jet.

I have simulated free jets in the past with success, but I used a much wider mesh (to place the bouondaries far away from the jet), with many many more gridcells - orders of magnitude more than you have in your simple model.

Finally, the free jet case is very sensitive to your choice of boundary conditions, so you may find that you need to play with those to get good solver behaviour.

Indeed, for what appears to be a "very simple" test case, it can be quite a pig to get right! And that's without delving into the round jet anomoly, which means that k-epsilon models will over-predict the jet spreading rate by around 30%.

Good luck!

December 12, 2020, 18:17	Inconsistent results - Axisymetric Free Jet	#1
jrsilvio_ver Member JuNNioR Join Date: Feb 2014 Location: Brazil Posts: 44 Rep Power: 12	Dear colleagues, good afternoon. I am modeling an axysymmetric circular jet with a diameter of 0.1 m and within a region of height 1 m and length 15 m. However, the results regarding the average velocity in the x-direction and the Reynolds stresses are absurdly different from references in the literature. Would anyone have any idea what to change in the simulation? The case files are available at: https://www.dropbox.com/s/unsk4c91uj...alho4.rar?dl=0 Thank you all for the attention.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Strange results for jet dispersion into ambient	shardiali	Main CFD Forum	5	October 16, 2015 16:41
Impinging free surface water jet on a flat plane	NicoMa	STAR-CCM+	0	July 7, 2015 07:15
Free surface is below than expected..results are not good	mechovator	CFX	24	April 8, 2011 01:41
Radial jet with free surface using FIDAP	Dr. Ahmed Hassaneen	Main CFD Forum	0	October 28, 1999 14:20
How to write the boundary condition for round free jet	Jason	Main CFD Forum	8	July 26, 1999 17:15

December 14, 2020, 00:57		#2
jrsilvio_ver Member JuNNioR Join Date: Feb 2014 Location: Brazil Posts: 44 Rep Power: 12	Nobody???????

December 29, 2020, 06:08		#3
Tobermory Senior Member Join Date: Apr 2020 Location: UK Posts: 737 Rep Power: 14	Well, I had a quick look at your set up files, and the most obvious thing that stood out to me was the shape of your domain. It is very narrow in the cross-stream (y axis) direction: aspect ratio of length to width is 30:1, and this places your atmospheric pressure boundary very close to the jet. I have simulated free jets in the past with success, but I used a much wider mesh (to place the bouondaries far away from the jet), with many many more gridcells - orders of magnitude more than you have in your simple model. Finally, the free jet case is very sensitive to your choice of boundary conditions, so you may find that you need to play with those to get good solver behaviour. Indeed, for what appears to be a "very simple" test case, it can be quite a pig to get right! And that's without delving into the round jet anomoly, which means that k-epsilon models will over-predict the jet spreading rate by around 30%. Good luck!