[Omran]

Hi,
I'm using remeshing for simulating my problem. I have some monitor points for pressure in my domain. The values of pressures before and after remeshing are really different and there is a jump in pressure values. I think it happens because of the position of my monitor points. In fact I think the position of a monitor is the nearest vertex and the nearest vertex before and after remeshing changes and it causes the problem.
I defined some source points in cfx pre and then I created my monitor points with the expression probe(pressure)@source point 1 .
what can I do to avoid these jumps in pressure values?
thanks.

[ghorrocks]

Here's some ideas:

Use a finer mesh.
Do not use remeshing
Put a meshing constraint on such that there is always a node at the monitor point (so snapping to the nearest neighbour is not required)

[Opaque]

Have you looked at your options when setting up monitor points ? If I recall correctly, it is your choice to use nearest vertex, or interpolate within the element the point lies in.

Here is from the recent documentation

(applies only when Option is set to Cartesian Coordinates or Cylindrical Coordinates)

Monitor Location Control settings determine where the monitor is placed with respect to the mesh, and how often the monitor position is updated. The following Monitor Location Control options are available:

Interpolation Type
Nearest Vertex

This option causes monitor point data to be taken from the mesh vertex nearest to the specified coordinates. The nearest mesh vertex is determined initially and again every time the monitor position is updated (based on the Position Update Frequency setting).

Trilinear

This option causes monitor point data to be interpolated to the specified coordinates using a weighted average of the data from the vertices of the containing mesh element. The containing mesh element, and the values of the weighting factors, are determined initially and again every time the position is updated (based on the Position Update Frequency setting).

Note: If Interpolation Type is set to Trilinear and the specified coordinates are not contained by the current mesh, CFX-Solver reverts to Nearest Vertex behavior. As a result, you may receive unexpected monitor output. However, if the monitor position updates, normal Trilinear behavior resumes whenever the coordinates are contained by the mesh. This might occur if the simulation includes a moving mesh or a stationary monitor point in a rotating domain.

What version are you using ?

[ghorrocks]

Good point, I have not looked at the monitor point options for a while

[Omran]

Thanks ghorrocks

Use a finer mesh.
I'm trying it.

Do not use remeshing
There is a big deformation and I couldn't do it without remeshing.

Put a meshing constraint on such that there is always a node at the monitor point (so snapping to the nearest neighbour is not required).
If I set my monitor point on a specified node (a node number) I think I'll have the problem again because the position of the node number is changing during remeshing and a node number before and after remeshing relevents to different nodes.

Thanks

[Omran]

Thanks Opaque

When I set interpolation type to trilinear and place my monitor point on the wall of my rigid body this error happens

Details of error:-
----------------
Error detected by routine PEEKCS
CDANAM = MONP9 /MONLOCCON/CZONE
CRESLT = NONE

Current Directory : /FLOW/CONTROL/OUTPUT/MONITOR

+--------------------------------------------------------------------+
| ERROR #001100279 has occurred in subroutine ErrAction. |
| Message: |
| Stopped in routine MEMERR

When I place my monitor point anywhere else this error doesn't happen.
How can I solve the problem?

Thanks

[ghorrocks]

I think if you turn interpolation off it handles being covered by a rigid body by not returning a value for that time step. But of course the down side is you get jumps in the values as the mesh passes over it. But these jumps are a sign of too coarse a mesh, so a finer mesh will reduce this.

[Omran]

Thanks ghorrocks

I think if you turn interpolation off
How can I turn interpolation off?

these jumps are a sign of too coarse a mesh, so a finer mesh will reduce this.
As you said I did it and I used a finer mesh and the jumps reduced but I still have them.

Thanks

[ainrahman]

hi i am new to cfx, i want to simulate an amphibious boat and monitor its heaving and pitching..somehow can anyone help me what the expression that we can put for output monitoring for heaving and pitching...for example if we want to monitor the ship drag we use force_x()@location..how about heave and pitching point monitor. thank you in advance

[ghorrocks]

I presume you are doing a rigid body simulation. Have a look in the CFX reference manual for the available variables to report on rigid bodies.

[hamed1983]

Hi all,

I am simulating a airship in wind tunnel and wish to monitor pressure data in some point over the airship in tunnel but i faced with some problem when i wanted to tilted the airship within tunnel that problem is the monitor point which i determined for pressure monitoring over airship body are not tilted with the model and stayed on 0 degree tilt. henceforth i imposed to manage the monitoring data on nearest vertex and i am in doubt that the output given data would be correct. so that i pleased really anyone to help me with some note and tips please.

Many Thanks

[-Maxim-]

Originally Posted by hamed1983

i am in doubt that the output given data would be correct.

Hi,
is the output without the tilt "correct"? Did you go through the FAQ?
https://www.cfd-online.com/Wiki/Ansy..._inaccurate.3F

[hamed1983]

Thanks for your kind reply anyhow The source of my results correctness is an experimental results approach they are almost correct except in some condition that results are very far from each other.