Star Time Integration

doug · March 25, 2014, 16:09

I'm running a transient analysis of an airfoil going through a sinusoidal motion. I specify the velocity as a function of time through a user-defined field function and link that to the oveset mesh. Star then integrates the velocity to obtain the new position for each step of the simulation.

It appears that Star is using a rectangle rule for time integration, which is a very low-order method. Is there a way to specify a higher-order method for that time integration? Like 4th-order Runge-Kutta? With the current integration scheme, my position during the sinusoidal motion is off by 2% with 128 time steps per cycle. Even trapezoidal integration would reduce that significantly.

Any input would be appreciated.

doug · March 25, 2014, 19:05

I just found an option that allows me to specify the overset grid position rather than the overset grid velocity. This option allows me to accurately define the position. However, I would like to check the velocity of the overset grid as a function of time. It should simply be the derivative of the position, but I would like to check it for accuracy. Does anyone know how to output the grid velocity (not the velocity of the fluid passing through the grid)? I've searched the documentation but haven't found any information on it.

Thanks in advance.

March 25, 2014, 16:09	Star Time Integration	#1
doug Member Doug Hunsaker Join Date: Mar 2009 Location: Logan, UT Posts: 63 Rep Power: 17	I'm running a transient analysis of an airfoil going through a sinusoidal motion. I specify the velocity as a function of time through a user-defined field function and link that to the oveset mesh. Star then integrates the velocity to obtain the new position for each step of the simulation. It appears that Star is using a rectangle rule for time integration, which is a very low-order method. Is there a way to specify a higher-order method for that time integration? Like 4th-order Runge-Kutta? With the current integration scheme, my position during the sinusoidal motion is off by 2% with 128 time steps per cycle. Even trapezoidal integration would reduce that significantly. Any input would be appreciated.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
time step directories naming issue	Andrea_85	OpenFOAM	3	April 3, 2014 08:38
[Commercial meshers] Using starToFoam	clo	OpenFOAM Meshing & Mesh Conversion	33	September 26, 2012 04:04
plot over time	fferroni	OpenFOAM Post-Processing	7	June 8, 2012 07:56
[Other] StarToFoam error	Kart	OpenFOAM Meshing & Mesh Conversion	1	February 4, 2010 04:38
[Commercial meshers] Trimmed cell and embedded refinement mesh conversion issues	michele	OpenFOAM Meshing & Mesh Conversion	2	July 15, 2005 04:15

March 25, 2014, 19:05		#2
doug Member Doug Hunsaker Join Date: Mar 2009 Location: Logan, UT Posts: 63 Rep Power: 17	I just found an option that allows me to specify the overset grid position rather than the overset grid velocity. This option allows me to accurately define the position. However, I would like to check the velocity of the overset grid as a function of time. It should simply be the derivative of the position, but I would like to check it for accuracy. Does anyone know how to output the grid velocity (not the velocity of the fluid passing through the grid)? I've searched the documentation but haven't found any information on it. Thanks in advance.