Simulation Mechanical and Multiphysics

Reply
Active Member
TyBjarnason
Posts: 10
Registered: ‎01-22-2013

Shell problem

338 Views, 8 Replies
01-22-2013 01:04 PM

I am trying to run a Non-Linear Simulation using Shell Elements and Brick Elements. I have my constraints linked to a few Parts which have Shell Elements and a Surface Remote Force on another Part which is also a Shell Element. Chen I check my model I am geting an error for all of my Parts with Shell Elements. When I remove the Remote force the error goes awat but then I have no results due to no force. The error I am getting says;

 

"No Mean_delta_t defined in part 1"

 

and this is for all part numbers that are Shell elements. What I am doing wrong? Is there a setting I am missing?

Please use plain text.
*Expert Elite*
AstroJohnPE
Posts: 521
Registered: ‎08-30-2012

Re: Shell problem

01-23-2013 12:10 PM in reply to: TyBjarnason

Hi,

 

The lines "No Mean_delta_t defined in part" are not identified as a warning nor an error. They are just reporting that those parts not not have the temperature difference through the thickness defined. (I think that line is a bug. My memory is that shell elements do not have a temperature difference through the thickness capability.)

 

The error is at the very bottom of the text file:

  Terminal Error: FEM2ESH#142
    FEM2ESH_FAIL_SOLVE_FOR_EQUIVALENT_FORCE 1

 

In plain English, the software was trying to convert your remote force into a series of nodal forces but was unable to find a solution. My only thought about why that would happen is if all of the nodes on the surface are in a straight line. Here's how to think of it:

  1. make a truss element from the point of the remote force to each node on the surface.
  2. Suppress the part that has the remote force.
  3. Pin all of the ends of the truss elments.
  4. Run the analysis. Is the truss structure statically stable in all 6 directions? (3 translations, 3 rotations)

It may also help to upgrade to service pack 2 just in case that fixes a bug.

 

If this is not the problem, please post a picture that shows the surface remote force (or the surface itself) selected.

John Holtz, PE
Mechanical Engineer
Pittsburgh, PA

16 years experience with Simulation Mechanical
Please use plain text.
Active Member
TyBjarnason
Posts: 10
Registered: ‎01-22-2013

Re: Shell problem

01-24-2013 10:54 AM in reply to: TyBjarnason

Loading Service Pac 2 helped eliminate this problem.

Please use plain text.
Active Member
TyBjarnason
Posts: 10
Registered: ‎01-22-2013

Re: Shell problem

01-24-2013 01:39 PM in reply to: TyBjarnason

I was wrong. The message is occuring again. I have a shell model with a Surface Remote Force. I don't understand why I get it sometimes and not others. Very fustrating and time consuming. 

Please use plain text.
Active Member
TyBjarnason
Posts: 10
Registered: ‎01-22-2013

Re: Shell problem

01-24-2013 01:55 PM in reply to: TyBjarnason

When I delete the force and place it at a different point the Delta_T error goes away and I get the attached error. I have also attached a picture of the area the force is applied.

Please use plain text.
*Expert Elite*
AstroJohnPE
Posts: 521
Registered: ‎08-30-2012

Re: Shell problem

01-24-2013 03:32 PM in reply to: TyBjarnason

Hi,

 

The line of text about the "Mean_delta_t" is not an error, so just ignore it.

 

This is just a guess about the "Error: Can not get data in database!" error. I think that the model file that it was trying to access was still open by some other process. Did anything crash or exit abnormally before this new problem occurred? If you reboot your computer, does the problem still exist?

 

For the remote surface force, where is the remote point relative to the surface? It is hard to tell from the image whether it is located off the surface or if it is in the plane of the surface. My guess is that the software cannot handle a "remote" force that is in the plane of the elements. I think you would just use the "Force" commnd.

 

But just to be sure we understand the command that is being used, are you using "Setup > Loads > Remote Force" shown in this image? Or a different command?

remote force command.png

 

You may want to post an archive of the model to the thread so that someone can take a look at it. See this discussion for instructions.

 

 

John Holtz, PE
Mechanical Engineer
Pittsburgh, PA

16 years experience with Simulation Mechanical
Please use plain text.
Active Member
TyBjarnason
Posts: 10
Registered: ‎01-22-2013

Re: Shell problem

01-24-2013 03:57 PM in reply to: AstroJohnPE

 

The "Mean Delta_T" cannot be ignored; it prevents the program from moving forward.

 

Yes I added the force in a similar manner. I would like to add an applied force as you would in a normal scenario but that option is not available with shell elements; it only allows me to pick the Remote Force option. Currently the program is shutting down from another fault so I cannot add specifics.

Please use plain text.
*Expert Elite*
AstroJohnPE
Posts: 521
Registered: ‎08-30-2012

Re: Shell problem

01-24-2013 07:56 PM in reply to: TyBjarnason

Sorry. I did not remember that MES shell elements cannot be loaded with a surface force. The work around to applying a force is to calculate what pressure gives the same force: F = p*A. Then apply the pressure (which can e oriented in any direction if needed).

 

I think the text that you have been attaching to your posts is the "DS.LDD" file located in the folder "Lube Truck FEA 3D\original2006499.ds_data\1". It is a simple text file that can be opened with Notepad, Word, etc. If I understand you correctly, without the remote force the model will run and the DL.LDD file does not include the "mean delta T" message.

 

When you add the remote force, the model does not run, but the message

        Terminal Error: FEM2ESH#142
        FEM2ESH_FAIL_SOLVE_FOR_EQUIVALENT_FORCE 1

located at the end of the file (dozens of lines after the software got past the last "mean delta T" message) is not the reason the analysis stops? I know that anything is possible, but I am confident the problem is the attempt by fem2esh.exe to find the nodal forces that give the equivalent force and moment as the remote load.

John Holtz, PE
Mechanical Engineer
Pittsburgh, PA

16 years experience with Simulation Mechanical
Please use plain text.
Active Member
TyBjarnason
Posts: 10
Registered: ‎01-22-2013

Re: Shell problem

01-25-2013 08:09 AM in reply to: AstroJohnPE

 

Thank you for the help John.

 

When I said the program would not run I meant the program stalls and aborts. When I remove the remote force the program will run, but now there is no force so therefore no results.

 

When I add a pressure there were two ways to do that, I used the variable pressure option and the program is operating without any failure. If you want me to send any of the files it might be too late since I am deleting all the files an reloading the scenario with these new changes which are working.

Please use plain text.