|
[Sponsors] |
February 18, 2019, 15:30 |
Error in CFX run
|
#1 |
Member
Thu Win
Join Date: Apr 2017
Posts: 70
Rep Power: 9 |
I am trying to simulate a flying wing in CFX and I got this cryptic error:
ERROR #004100018 I have tried increasing the time step but it still bawks at me after round 2. See attachment for log. Edit: Fixed problem by editing mesh. Previous mesh had error messages. Edit: Run 2 failed. Last edited by tyw7; February 18, 2019 at 17:45. |
|
February 18, 2019, 19:35 |
|
#2 |
Super Moderator
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 17,872
Rep Power: 144 |
__________________
Note: I do not answer CFD questions by PM. CFD questions should be posted on the forum. |
|
February 18, 2019, 19:37 |
|
#3 |
Member
Thu Win
Join Date: Apr 2017
Posts: 70
Rep Power: 9 |
I had a look at that but I am hitting a wall. No idea where the error is coming from.
I changed time step, smaller mesh, etc. I can't really move the outlet since I am hitting the solidworks limit. |
|
February 18, 2019, 19:42 |
|
#4 |
Super Moderator
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 17,872
Rep Power: 144 |
Solidworks limit? What is that?
Please attach an image of what you are modelling. If you have progressed beyond the output files you have already attached then attach a current output file as well.
__________________
Note: I do not answer CFD questions by PM. CFD questions should be posted on the forum. |
|
February 18, 2019, 19:46 |
|
#5 |
Member
Thu Win
Join Date: Apr 2017
Posts: 70
Rep Power: 9 |
https://transfernow.net/ddl/flyingwing1
Here's the model How to upload the mesh? https://forum.solidworks.com/thread/22416 |
|
February 18, 2019, 20:06 |
|
#6 |
Super Moderator
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 17,872
Rep Power: 144 |
I don't download from strange sites. Just post some images on the forum. Don't bother uploading the mesh, I won't look at that either. Just post images on the forum.
You have only said you are modelling a wing. How big is it? How fast? What fluid is it moving in? 2D or 3D? Any appendages? What results do you want to get (lift, drag, moments etc)? Why are you doing this analysis?
__________________
Note: I do not answer CFD questions by PM. CFD questions should be posted on the forum. |
|
February 18, 2019, 20:19 |
|
#7 |
Member
Thu Win
Join Date: Apr 2017
Posts: 70
Rep Power: 9 |
Well the wing is 62.6 m long as it's a flying wing. Doing analysis as part of year 3 project
Moving at Mach 0.85. |
|
February 18, 2019, 23:50 |
|
#8 |
Super Moderator
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 17,872
Rep Power: 144 |
You still have not explained much about what you are doing. I cannot give much of an opinion if you do not explain what you are doing.
However, I can see that your mesh is very coarse. It also does not have inflation/prism layers. While it is OK to do debugging on very coarse meshes it will give very inaccurate answers and occasionally strange error messages (such as what you are seeing). Once you have the basic simulation working correctly you will need to do a mesh sensitivity study to determine the mesh resolution you require for accurate results. I can assure you it will be far finer than you currently have. You will also need to do a convergence criteria and boundary proximity sensitivity study as well.
__________________
Note: I do not answer CFD questions by PM. CFD questions should be posted on the forum. |
|
February 19, 2019, 05:23 |
|
#9 |
Member
Thu Win
Join Date: Apr 2017
Posts: 70
Rep Power: 9 |
That's the problem. I cannot go finer or I'm hitting the 512k limit. Also, I have an inflation layer of 50 layers, first level thickness around 0.004 m
|
|
February 19, 2019, 05:37 |
|
#10 |
Super Moderator
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 17,872
Rep Power: 144 |
Do a careful sensitivity analysis to determine exactly what mesh, convergence and domain size you need. If your sensitivity analysis says you must go finer but you are at the limits of your license then your simulation is impossible. You should report that the simulation is not possible with available resources and spend your time doing something which is possible.
__________________
Note: I do not answer CFD questions by PM. CFD questions should be posted on the forum. |
|
February 19, 2019, 07:00 |
|
#11 |
Senior Member
urosgrivc
Join Date: Dec 2015
Location: Slovenija
Posts: 365
Rep Power: 12 |
50 inflation layers sounds a lot, personally I think it is too much.
make sure to have a nice transition from inflation to tets with not too much difference in the thickness, and an aspect ratio of the first layer below 150, in this case, it works quite nicely. Try 20 layers, if this makes the transition good enough you will save some elements. |
|
February 19, 2019, 07:18 |
|
#12 |
Member
Thu Win
Join Date: Apr 2017
Posts: 70
Rep Power: 9 |
I use first layer thickness. Is that not adequate? Or is there a better way?
|
|
February 19, 2019, 07:23 |
|
#13 |
Senior Member
urosgrivc
Join Date: Dec 2015
Location: Slovenija
Posts: 365
Rep Power: 12 |
Let us just say that there are infinite ways some are good some are bad
And the correct thickness is actually dependant on a few factors #Y+ #stage of flow developement #aspect ratio #turbulent models #near wall region #velocity near wall #heat transfer (for those who use it) #... |
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Can CFX run 2D problems | mariam.sara | CFX | 27 | October 8, 2015 07:05 |
Can workbench and cfx run correctly on vista? | Sandam | CFX | 1 | December 2, 2007 17:32 |
How to run CFX for inviscid cases? | MouseInMilk | CFX | 4 | July 29, 2006 00:56 |
How to make the solver of CFX run automate | Jane | CFX | 4 | April 10, 2006 15:42 |
How to run CFX on the background? | David | CFX | 1 | January 19, 2004 17:24 |