Learning Forum

[Jorge Silva]

Hi everyone,

 

I am planning to use ANSYS for MEMS Modal and mechanical simulation.

I want to simulate a MEMS device that will have an applied AC current that will deflect the MEMS structure due to the Lorentz force caused by the interaction between a Magnetic field and the appllied AC current. Is it possible to simulate the mechanical deflection of a MEMS structure due to a Lorentz force in ANSYS ?

If this is possible what ANSYS modules are needed for this simulation

Thank you for your time and knowledge,

 

Jorge Silva

[Bill Bulat]

Hi Jorge,

Yes this is possible however it is not natively exposed in Mechanical so you'll need to use command objects to convert element types so that they use electromagnetostructural DOFs (UX, UY, UZ, AZ, and VOLT):

I created a small test case to flesh out a procedure (sadly I'm not allowed to send it to you). It's a fixed-fixed electrically conductive beam. A current is sent along the length of the beam - a small voltage drop is associated with the impressed current:

A lateral (z direction) external B field (1 Tesla) is applied:

I used a Coupled Field Static analysis system at 2023R1. The beam and surrounding "air" must comprise a single part (conformal mesh, no contact).

The command objects under the bodies were:

Both bodies were declared to use the following engineering disciplines:

 

 

If possible in your production model, extend modeled representation of the electric conductor traces to the external boundaries of the model in such a way that they meet the boundary perpendicularly:

 

In my test case I assumed that the device is operating in an externally applied magnetic field (e.g., a relatively large and distant PM is creating the field, which I've seen in MEMS mirrors that are actuated by Lorentz forces). I used a DFLX command to define that field on the external surfaces of the model:

 

 

Create a named selection (with no spaces in the name) comprising the external surfaces of the mesh. Enter the name of that named selection in the NODE field of the DFLX command.

The net reaction force for this 0.001 m long beam carrying 0.1 A and immersed in a transverse 1 Tesla field should be the product of these three numbers (1e-4 N), which is what was calculated by the model:

Kind regards,

Bill

[J G]

 

HI Bill,

I am new to Ansys and was trying to replicate the simulation above.

The simulation setup is almost identical to what you have done above.

I assigned both the bodies with same material (structural steel, the Command object should overwrite the properties for the air) and had one Physics region assigned for all the bodies.

Apart from that, everything else is identical to your setup above, but I kept getting the errror below (i have copied the whole solution information at the end).

 *** ERROR ***                           CP =       2.531   TIME= 22:06:58
 For element type = 2 (SOLID226), KEYOPT(4) = 4 is invalid.

 

 

Which I feel shouln’t be case as I can see the option in valid in the table from the element reference document.

Is there something else I am missing?

 

 Verification–Coupled Field Static (A5)                                      



          ***** MAPDL ANALYSIS DEFINITION (PREP7) *****
 *********** Nodes for the whole assembly ***********
 *********** Elements for Body 1 “beam” ***********
 *********** Elements for Body 2 “air” ***********
 *********** Send User Defined Coordinate System(s) ***********
 *********** Set Reference Temperature ***********
 *********** Send Materials ***********

 PRINTOUT RESUMED BY /GOP

 C*** Beam attributes


 *GET  DMNSN     FROM  PARM  typeids   ITEM=DIM  1     VALUE=  1.00000000   

 *IF  dmnsn                             ( =   1.00000     )  EQ 
      1                                 ( =   1.00000     )  THEN   

 ELEMENT TYPE       1 IS SOLID226     3-D 20-NODE STRUC-EMAG SOLID
  KEYOPT( 1- 6)=    10101      0      0        0      0      0
  KEYOPT( 7-12)=        0      0      0        0      0      0
  KEYOPT(13-18)=        0      0      0        0      0      0

 CURRENT NODAL DOF SET IS  UX    UY    UZ    AZ    TEMP  VOLT
  THREE-DIMENSIONAL MODEL

 ELEMENT TYPE       1 IS SOLID226     3-D 20-NODE STRUC-EMAG SOLID
  KEYOPT( 1- 6)=    10101      0      0        0      0      0
  KEYOPT( 7-12)=        0      1      0        0      0      0
  KEYOPT(13-18)=        0      0      0        0      0      0

 CURRENT NODAL DOF SET IS  UX    UY    UZ    AZ    TEMP  VOLT
  THREE-DIMENSIONAL MODEL

 *ELSEIF  dmnsn                             ( =   1.00000     )  EQ 
          2                                 ( =   2.00000     )  THEN   

 *ENDIF

 MATERIAL          1     MURX =   1.000000     

 ARRAY PARAMETER MATIDS    DELETED.

 ARRAY PARAMETER TYPEIDS   DELETED.

 SET PARAMETER DIMENSIONS ON  MATIDS TYPE=ARRA  DIMENSIONS=     1     1     1

 PARAMETER MATIDS(1) =     2.000000000   

 PARAMETER MATID =     2.000000000   

 SET PARAMETER DIMENSIONS ON  TYPEIDS TYPE=ARRA  DIMENSIONS=     1     1     1

 PARAMETER TYPEIDS(1) =     2.000000000   

 PARAMETER ARG1 =     1.000000000   

 PRINTOUT RESUMED BY /GOP

 C*** Air attributes


 *GET  DMNSN     FROM  PARM  typeids   ITEM=DIM  1     VALUE=  1.00000000   

 *IF  dmnsn                             ( =   1.00000     )  EQ 
      1                                 ( =   1.00000     )  THEN   

 ELEMENT TYPE       2 IS SOLID226     3-D 20-NODE COUPL-FLD SOLID
  KEYOPT( 1- 6)=      111      0      0        0      0      0
  KEYOPT( 7-12)=        0      0      0        0      0      0
  KEYOPT(13-18)=        0      0      0        0      0      0

 CURRENT NODAL DOF SET IS  UX    UY    UZ    AZ    TEMP  VOLT
  THREE-DIMENSIONAL MODEL

 *** ERROR ***                           CP =       2.531   TIME= 22:06:58
 For element type = 2 (SOLID226), KEYOPT(4) = 4 is invalid.             

 CURRENT NODAL DOF SET IS  UX    UY    UZ    AZ    TEMP  VOLT
  THREE-DIMENSIONAL MODEL


 NUMBER OF WARNING MESSAGES ENCOUNTERED=          0
 NUMBER OF ERROR   MESSAGES ENCOUNTERED=          1



 ***** PROBLEM TERMINATED BY INDICATED ERROR(S) OR BY END OF INPUT DATA *****

Regards,

Jack