Floating point exception error on a pipe flow

svantevid · July 19, 2022, 09:40

I'm trying to simulate a compressible subsonic air flow through a pipe with a pressure inlet and mass flow outlet. Pipe flows into a container which I removed and added a partial sphere. Viscous model is realizable k-e with enhanced wall fn). Mesh has skewness bellow 0,97, min orth quality is 0,02.

When I run the simulation, it gives a floating point exception error after about 200 iterations. When I did the simulation without the partial sphere everything went smoothly.

I have no idea what is the problem.

Model screenshot:

Image link:
https://imgur.com/tVGrdzG

Jason.Chen · July 26, 2022, 03:28

when with partial sphere, this case may possibly not a steady flow. You may try the use the transient to simulation it

svantevid · July 26, 2022, 05:37

If I understand correctly, the problem is the shape of the sphere itself? Should I change it to a cuboid then?

AlexanderZ · July 27, 2022, 00:19

I think the problem is not the shape but the flow, which could be complicated because of shpere shape. It could oscillate thats why transient simulation was suggested

Jason.Chen · July 27, 2022, 01:43

When the flow pass through the barrier, it may cause the unsteady flow. such as the karman vortex.

svantevid · July 27, 2022, 03:13

What I meant was, how does the shape of the outlet (in my case sphere) impact the divergence or as you described it, oscilation?

Update:
I ran another simulation with a cuboid outlet shape and the case converged without any issues. My problem is now solved but I would like to understand how the sphere impacts the calculation.

July 19, 2022, 09:40	Floating point exception error on a pipe flow	#1
svantevid New Member Join Date: May 2019 Posts: 18 Rep Power: 7	I'm trying to simulate a compressible subsonic air flow through a pipe with a pressure inlet and mass flow outlet. Pipe flows into a container which I removed and added a partial sphere. Viscous model is realizable k-e with enhanced wall fn). Mesh has skewness bellow 0,97, min orth quality is 0,02. When I run the simulation, it gives a floating point exception error after about 200 iterations. When I did the simulation without the partial sphere everything went smoothly. I have no idea what is the problem. Model screenshot: Image link: https://imgur.com/tVGrdzG

July 27, 2022, 00:19		#4
AlexanderZ Senior Member Alexander Join Date: Apr 2013 Posts: 2,363 Rep Power: 34	I think the problem is not the shape but the flow, which could be complicated because of shpere shape. It could oscillate thats why transient simulation was suggested __________________ best regards ****************************** press LIKE if this message was helpful

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Floating point exception error when rotate the geometry	zyfsoton	OpenFOAM Running, Solving & CFD	1	June 20, 2022 05:49
icoFoam floating point exception (8)	leizhao512	OpenFOAM Running, Solving & CFD	7	November 1, 2018 11:43
Cylindrical pipe - nonNewtonianIcofoam - Floating point exception (core dumped)	harshchandak97	OpenFOAM	0	July 8, 2017 01:40
[snappyHexMesh] snappyHexMesh and cyclic boundaries	Ruli	OpenFOAM Meshing & Mesh Conversion	2	December 9, 2013 06:51
pimpleDyMFoam constantly giving floating point exception and solution divergence	fedvasu	OpenFOAM Running, Solving & CFD	0	November 28, 2013 00:53

July 26, 2022, 03:28		#2
Jason.Chen New Member Join Date: Jun 2022 Posts: 4 Rep Power: 3	when with partial sphere, this case may possibly not a steady flow. You may try the use the transient to simulation it

July 26, 2022, 05:37		#3
svantevid New Member Join Date: May 2019 Posts: 18 Rep Power: 7	If I understand correctly, the problem is the shape of the sphere itself? Should I change it to a cuboid then?

July 27, 2022, 01:43		#5
Jason.Chen New Member Join Date: Jun 2022 Posts: 4 Rep Power: 3	When the flow pass through the barrier, it may cause the unsteady flow. such as the karman vortex.

July 27, 2022, 03:13		#6
svantevid New Member Join Date: May 2019 Posts: 18 Rep Power: 7	What I meant was, how does the shape of the outlet (in my case sphere) impact the divergence or as you described it, oscilation? Update: I ran another simulation with a cuboid outlet shape and the case converged without any issues. My problem is now solved but I would like to understand how the sphere impacts the calculation.