CFD Online URL
[Sponsors]
Home > Forums > Main CFD Forum

Numerical Scheme Correlation

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

Like Tree1Likes
  • 1 Post By cfdnewbie

Reply
 
LinkBack Thread Tools Display Modes
Old   March 15, 2014, 11:35
Default Numerical Scheme Correlation
  #1
New Member
 
Matt Holford
Join Date: Oct 2013
Posts: 4
Rep Power: 3
holformr is on a distinguished road
Hi all,

I have quick question.
I am correlating the flow performance of a water valve with CFD simulations. The target flow rate I am trying to achieve is 52L/min.

1st order numerical scheme, CFD tells me I get 82L/min.

2nd order I get 53L/min

3rd order is 62L/min

The software I'm using is Autodesk Simulation CFD 2014 which is FEM based.
The mesh is at an independent state with a low y+ on average, thus the mesh isn't a problem.

In this case, why would the 1st and 3rd order schemes be yielding me odd results?
holformr is offline   Reply With Quote

Old   March 15, 2014, 12:46
Default
  #2
Senior Member
 
cfdnewbie
Join Date: Mar 2010
Posts: 551
Rep Power: 10
cfdnewbie is on a distinguished road
on the same mesh, different numerical schemes will give you (in under resolved scenarios usually) different results. Are you perfectly resolved? Is your flow laminar or turbulent? What is your Reynolds number? What time discretization do you use? Do you have cavitation going on?

You need to study the source of the discrepancies you are seeing, and it is a very common effect to see a difference when changing the spatial discretization, as you are doing.
cfdnewbie is offline   Reply With Quote

Old   March 15, 2014, 12:59
Default
  #3
New Member
 
Matt Holford
Join Date: Oct 2013
Posts: 4
Rep Power: 3
holformr is on a distinguished road
By resolved do you mean low magnitude converged residuals?
Reynolds number is on average 10^5.
It's a steady simulation.
I may have cavitation, there's a very small gap, 0.4mm.
holformr is offline   Reply With Quote

Old   March 15, 2014, 13:40
Default
  #4
Senior Member
 
cfdnewbie
Join Date: Mar 2010
Posts: 551
Rep Power: 10
cfdnewbie is on a distinguished road
By resolved I mean DNS, not LES or RANS...do you resolve all the flow features?
I doubt that Re=10^5 is steady, if you are solving it with a RANS approach, it will strongly depend on the interaction of the numerical scheme and the turbulence model.
What equation of state model do you have for water?
FMDenaro likes this.
cfdnewbie is offline   Reply With Quote

Old   March 15, 2014, 13:46
Default
  #5
New Member
 
Matt Holford
Join Date: Oct 2013
Posts: 4
Rep Power: 3
holformr is on a distinguished road
The turbulence model is k-e, petrov galerkin numerical scheme.
By steady, I meant it's not a transient solution, no time dependence.
holformr is offline   Reply With Quote

Old   March 15, 2014, 13:47
Default
  #6
Senior Member
 
Filippo Maria Denaro
Join Date: Jul 2010
Posts: 1,287
Rep Power: 17
FMDenaro will become famous soon enough
first and third order accurate schemes have different action of the local truncation error compared to second order one
FMDenaro is offline   Reply With Quote

Reply

Tags
correlation, discretization, higher order, numerical scheme, separation

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are On


Similar Threads
Thread Thread Starter Forum Replies Last Post
How L-norms are used to study stability of a numerical scheme? Ravindra Shende Main CFD Forum 7 April 22, 2012 18:15
numerical scheme without artificial diffusion artifact mpeti OpenFOAM Running, Solving & CFD 2 October 26, 2011 12:36
Numerical viscosity due to the MUSCL and HLL coulpled scheme sonsiest Main CFD Forum 0 May 23, 2011 16:37
the numerical scheme for LES. John S Main CFD Forum 2 March 14, 2004 08:52
numerical scheme ado Main CFD Forum 3 October 12, 2000 09:20


All times are GMT -4. The time now is 23:09.