# Unusually large initial Cd upon starting my simulation, is this normal?

 Register Blogs Members List Search Today's Posts Mark Forums Read

February 5, 2021, 12:49
Unusually large initial Cd upon starting my simulation, is this normal?
#1
New Member

Join Date: Oct 2020
Posts: 13
Rep Power: 3
Hi everyone,

Upon starting my simulation the Cd listed is unbelievably high, as you can see it does begin to fall but it seems as though it would take an age to be what would be expected for the cylindrical case im running.

I have run a previous simulation of a similar case where the geometry was simply 1/5 of the current size and this resulted in results that made sense.

Does this fact mean there is an issue with my simulation or is this to be expected?

I have attatched my, mesh and a screenshot of the solver for convenience.

Attached Images
 72d59bc6a76c2c7773a283d3a42ec091.jpg (122.4 KB, 13 views) 499db0231c1b5c31a140bc09d218686c.png (87.2 KB, 9 views)

February 6, 2021, 04:13
#2
Senior Member

Pay D.
Join Date: Aug 2011
Posts: 158
Blog Entries: 1
Rep Power: 12
Quote:
 Originally Posted by Creg Hi everyone, Upon starting my simulation the Cd listed is unbelievably high, as you can see it does begin to fall but it seems as though it would take an age to be what would be expected for the cylindrical case im running. I have run a previous simulation of a similar case where the geometry was simply 1/5 of the current size and this resulted in results that made sense. Does this fact mean there is an issue with my simulation or is this to be expected? I have attatched my, mesh and a screenshot of the solver for convenience. Many thanks in advance
I don't know what is your physical problem but as long as you converge to the right answer initial calculations just can be considered as initial guess.

But if you don't, you must double check everything, including your initial condition.

Initial condition includes your flow condition and boundary condition.

Then check your reference values, for your 2D case it is the reference length. For example for an airfoil reference length is the chord length that usually is scaled to 1.

If flow and boundary condition and reference values are all right, you should converge to the right answer!

 February 6, 2021, 06:28 #3 New Member   Bradley Craig Join Date: Oct 2020 Posts: 13 Rep Power: 3 Thanks for replying I appreciate it. Is it common for the cd to start so high though? I'd expect it to be around 1 but it starts at 500M and furthermore, takes an age to converge. Whereas with the smaller scale case and coarser mesh the initial cd is already almost as Id expect and thus converges in a fraction of the time. (3hours vs 22hours) Is this just an inherent property of cfd that upon refining the mesh the initial guess gets huge?

 February 6, 2021, 08:33 #4 Member   Amit Join Date: May 2013 Posts: 85 Rep Power: 10 If you can share the cfg file/case, it may help in answering better ....

February 6, 2021, 09:01
#5
New Member

Join Date: Oct 2020
Posts: 13
Rep Power: 3
Of course, apologies. Here are both my mesh and cfg file
Attached Files
 mesh.txt (1.8 KB, 1 views) cfg.txt (7.4 KB, 1 views)

 February 6, 2021, 11:41 #6 New Member   Bradley Craig Join Date: Oct 2020 Posts: 13 Rep Power: 3 Fixed!! Thanks for your time, I had just misstyped my Reynolds number

 Tags drag coefficient