CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   FLUENT (https://www.cfd-online.com/Forums/fluent/)
-   -   MDM Error (https://www.cfd-online.com/Forums/fluent/35583-mdm-error.html)

mAx January 28, 2005 04:24

MDM Error
 
Hi, I am setting a MDM calculation on an hydraulic valve. I am using an UDF (Fluent Tutorial 2d Valve Motion) which automaticaly calculates the piston displacement by integrating the force difference between piston's force and spring force (behind the piston). The MDM strategy is simple: I create a layering band in front of my piston, and I destroy another layering zone behind the piston as the piston moves. I already tested it with 1/16 from my model (ie: 22.5 degree revolution symmetry) without modeling the layering to destroy. Results are great, but as I compare them with the experiments I get with Fluent, the dp should lightly grow with the massflow (linear profile). With my results the dp falls lightly. I tried to simply include the seals effects in the udf but it doesn't change anything. So I try now to simulate the half model (with the right inlet and outlet) meaning I have to include the destroy layering band. As I test the piston motion in fluent with a profile (piston motion describes by profile) I get perfect motion (eg: adding layers and suppressing layers). But as I switch with the udf motion, I got the following error:

Warning: storage skewness not shared by zones 30 and 17. Error: coalesce_rib_faces: can't interpolate data, zones of different type detected.

(I have no interpolation and my skewness is ok. Grid check ok just before I Iterate) Does anybody already have this message? What is the solution? Thanks mAx


Master312 May 31, 2022 10:25

Answer
 
Quote:

Originally Posted by mAx
;117492
Hi, I am setting a MDM calculation on an hydraulic valve. I am using an UDF (Fluent Tutorial 2d Valve Motion) which automaticaly calculates the piston displacement by integrating the force difference between piston's force and spring force (behind the piston). The MDM strategy is simple: I create a layering band in front of my piston, and I destroy another layering zone behind the piston as the piston moves. I already tested it with 1/16 from my model (ie: 22.5 degree revolution symmetry) without modeling the layering to destroy. Results are great, but as I compare them with the experiments I get with Fluent, the dp should lightly grow with the massflow (linear profile). With my results the dp falls lightly. I tried to simply include the seals effects in the udf but it doesn't change anything. So I try now to simulate the half model (with the right inlet and outlet) meaning I have to include the destroy layering band. As I test the piston motion in fluent with a profile (piston motion describes by profile) I get perfect motion (eg: adding layers and suppressing layers). But as I switch with the udf motion, I got the following error:

Warning: storage skewness not shared by zones 30 and 17. Error: coalesce_rib_faces: can't interpolate data, zones of different type detected.

(I have no interpolation and my skewness is ok. Grid check ok just before I Iterate) Does anybody already have this message? What is the solution? Thanks mAx


hello dear
you must change your boundary condition of "walls to interface" in "direction" of your interface. in other words you must change all boundary conditions that slide through "motion wall" to interface and then define them in manual interface. in the last you have all boundary conditions of slider with "interface" mode, "no" part of that.


All times are GMT -4. The time now is 00:44.