"Caedium stopped working" right after launching the moment monitor.

I am using the Professional add on to do an analysis of a three bladed ceiling fan by MRF. Right after I launch the Moment monitor caedium stops working. I cleaned the C drive, restarted the PC and tried to launch the Moment monitor several times, it doesn't work. please help me out!

No votes yet

Create mesh explicitly first

Before creating the Moment monitor work through the meshing process, first creating the surface mesh and then the volume mesh as described for the tutorial "Flow Over a Double-Sided Membrane". If you encounter problems review the FAQ "How do I fix mesh and solver failures?".

Meshing is done prior to the Moment monitor launch.

Thank you. Moment monitor was launched only after the creation of Surface and Volume meshes as explained in "Rotating Paddle" tutorial. Yet the problem persists. Is there anything else which is to be taken care of prior to Moment monitor launch?

Moment monitor shows zero moment after solving.

I was able to run the solver but at the end of iterations the moment shown in the monitor is zero. And the velocity color map has only changed its value from initial to reference velocity value. I am unable to identify the mistake I have committed. Please help me out!

Create Residual Monitor

Does the Residual Monitor resemble the one in any of the tutorials? It is an essential part of diagnosing whether the simulation is converging.

Did you connect the faces between the MRF volume and surrounds stationary volume?

Did you create a volume group and apply the Substance to that group?

In terms of the Moment Monitor creation crash are you using the latest 64-bit version of Caedium?

Details of the analysis

Residual monitor resemble the one in tutorials, but after the first couple of hundred iterations the lines remain straight till the last.
The fan assembly and its faces are given angular velocity condition and there is a cylinder surrounding the assembly. But fan assembly is not identified as a volume.
A fan group is made containing all the faces of the fan.
Substance is added to the entire cylinder surrounding the fan assembly.
Regarding the moment monitor crash, it only happened in 32 bit system, with the help of a friend who has 64 bit system that problem is solved.
Connection between fan group and volume surrounding it has not been done.
Till the end of iterations the values calculated in moment and force monitors remain zero.

Review original comment on MRF

This is not the way to run a MRF simulation. The best explanation I have, without doing it for you, is the described in the comment "Use Moving Reference Frame (MRF)".

You need to create 2 water-tight volumes. One that is a cylinder with a void within the fan shape (MRF volume) and a larger volume with a perfectly matched cylinder subtracted.

Applying Angular velocity to the fan surfaces is not correct. For an MRF you apply the Angular Velocity to the MRF volume.

I got where I went wrong but one more clarification.

I did not subtract the fan assembly volume from the cylinder and the reason is as follows. When I import the fan assembly it is identified as a set of surfaces rather than a volume so I used stitch to convert that into a volume and encountered the error "Warning: Invalid geometry, try fix". I tried to fix it still the error persists. Please give me suggestions on how to clean the geometry so that it can be stitched into a volume. Thank you.


Assuming you are using a CAD system to create water-tight flow volumes then try using STEP (.stp) to exchange your geometry with Caedium. STEP maintains the analytic shape and topology better than most other exchange formats.

Volume identification problem solved. But!

Thank you very much for the help. After making the required changes to the geometries in the CAD software and importing it in STEP format finally the geometries are identified as volumes. But I faced an other problem while subtracting and posted it as a new topic. Thanks again.