[Solved] FEM and multiple forces
Forum rules
and Helpful information
and Helpful information
IMPORTANT: Please click here and read this first, before asking for help
Also, be nice to others! Read the FreeCAD code of conduct!
Also, be nice to others! Read the FreeCAD code of conduct!
[Solved] FEM and multiple forces
Hi
I have used freecad v0.15 (which came with my ubuntu distribution) for some time now, and found it a very useful tool for visualising the bits of stuff I make in my workshop. I have tried, (but failed ...) to use the FEM side of this and as it was not important to me I never bothered to worry about it. Anyway today I've been playing about with the Appimage for linux (FreeCAD-0.17.13541.9948ee4.glibc2.17-x86_64.AppImage) and as I had a rather more critical item to machine thought I'd play with the FEM options - just to get some simple ideas about displacements under forces. I've seen the video tutorials, and it all looked fairly simple.
I can activate the FE modelling, and model some simple cases but I'm having trouble when trying to use multiple forces. Let me try and explain.
I'm modelling a C clamp. the two ends are secured by a pinch bolt which applies a compressive force between the ends. I can use the inside surface of the clamp as a 'Constraint fixed' member as the pipe that the clamp is attached to is much stronger than the clamp itself.
I have tried to have two 'Constraint force' members in my analysis, but when I run the analysis the system seems to pick one of the force members and ignores the other one. I can tell this by looking at the displacement results.
I've also tried having a single 'Constraint force' member with two separate references (using the two different end faces on the clamp), but when I do this I can only ever assign a single direction for both references, obviously I need the forces to be in opposite directions.
I know that in the grand scheme of things I can model one force on the clamp, then assume that in the real world the displacements will be that amount on each leg, but I'm treating this as an academic exercise .
Am I doing something stupid, or is this a limitation of the FEM so far ?
-- Chris
PS BTW really great work guys, this is quite something !
I have used freecad v0.15 (which came with my ubuntu distribution) for some time now, and found it a very useful tool for visualising the bits of stuff I make in my workshop. I have tried, (but failed ...) to use the FEM side of this and as it was not important to me I never bothered to worry about it. Anyway today I've been playing about with the Appimage for linux (FreeCAD-0.17.13541.9948ee4.glibc2.17-x86_64.AppImage) and as I had a rather more critical item to machine thought I'd play with the FEM options - just to get some simple ideas about displacements under forces. I've seen the video tutorials, and it all looked fairly simple.
I can activate the FE modelling, and model some simple cases but I'm having trouble when trying to use multiple forces. Let me try and explain.
I'm modelling a C clamp. the two ends are secured by a pinch bolt which applies a compressive force between the ends. I can use the inside surface of the clamp as a 'Constraint fixed' member as the pipe that the clamp is attached to is much stronger than the clamp itself.
I have tried to have two 'Constraint force' members in my analysis, but when I run the analysis the system seems to pick one of the force members and ignores the other one. I can tell this by looking at the displacement results.
I've also tried having a single 'Constraint force' member with two separate references (using the two different end faces on the clamp), but when I do this I can only ever assign a single direction for both references, obviously I need the forces to be in opposite directions.
I know that in the grand scheme of things I can model one force on the clamp, then assume that in the real world the displacements will be that amount on each leg, but I'm treating this as an academic exercise .
Am I doing something stupid, or is this a limitation of the FEM so far ?
-- Chris
PS BTW really great work guys, this is quite something !
Last edited by lidgaca on Sun Mar 03, 2019 1:41 pm, edited 1 time in total.
Re: FEM and multiple forces
Hi Chris,
Maybe you should update to 0.18 pre-release which is very stable.
https://github.com/FreeCAD/FreeCAD/rele ... g/0.18_pre
Tons of improvements have been performed about Fem workbench since 0.17.
Also you could send your FC file so that the community could have a look on it.
Alex
Maybe you should update to 0.18 pre-release which is very stable.
https://github.com/FreeCAD/FreeCAD/rele ... g/0.18_pre
Tons of improvements have been performed about Fem workbench since 0.17.
Also you could send your FC file so that the community could have a look on it.
Alex
Re: FEM and multiple forces
Alex
Thanks for your reply. I downloaded the AppImage of 0.18, but it seems to behave the same.
I do have a (very) simple model that exhibits the problem, but I can't see how to attach a file to a forum post ... So here is a link to the file on Dropbox ... if I have sinned against some great forum rule a) I apologize, and b) can you tell me how I should include the file
https://www.dropbox.com/s/xrln62nkcgvbw ... FCStd?dl=0
Describing the problem with multiple force constraint members ... open the analysis and add a force constraint to Cut003.Face3, and set the direction to be inwards towards the split. A force of 30 N is sensible. Click OK, then add a second force constraint this time to Cut003.Face11. Again set the direction and force values.
Now create the mesh (max 3mm and min 2mm are reasonable values), and run calculix. The resulting displacement mesh will show that one of the clamp legs has been moved inwards and the other one has not moved. Obviously the model has ignored one of the force constraint members.
Describing the problem with a single force constraint member with two reference surfaces ... add the force constraint as above for Cut003.Face3. Now add a reference to Cut003.Face11. The reference list will show the two members. If you now try to set a direction for the force. the force on both faces will be set to the direction you choose. There does not seem to be any method of selecting one of the references and setting a direction to it without affecting the other one.
As I say this is all an academic exercise, and I do understand how to get round the problem for this particular model. I just wondered if the FEM was restricted in some way or if I was just doing something wrong.
-- Chris
Thanks for your reply. I downloaded the AppImage of 0.18, but it seems to behave the same.
I do have a (very) simple model that exhibits the problem, but I can't see how to attach a file to a forum post ... So here is a link to the file on Dropbox ... if I have sinned against some great forum rule a) I apologize, and b) can you tell me how I should include the file
https://www.dropbox.com/s/xrln62nkcgvbw ... FCStd?dl=0
Describing the problem with multiple force constraint members ... open the analysis and add a force constraint to Cut003.Face3, and set the direction to be inwards towards the split. A force of 30 N is sensible. Click OK, then add a second force constraint this time to Cut003.Face11. Again set the direction and force values.
Now create the mesh (max 3mm and min 2mm are reasonable values), and run calculix. The resulting displacement mesh will show that one of the clamp legs has been moved inwards and the other one has not moved. Obviously the model has ignored one of the force constraint members.
Describing the problem with a single force constraint member with two reference surfaces ... add the force constraint as above for Cut003.Face3. Now add a reference to Cut003.Face11. The reference list will show the two members. If you now try to set a direction for the force. the force on both faces will be set to the direction you choose. There does not seem to be any method of selecting one of the references and setting a direction to it without affecting the other one.
As I say this is all an academic exercise, and I do understand how to get round the problem for this particular model. I just wondered if the FEM was restricted in some way or if I was just doing something wrong.
-- Chris
Re: FEM and multiple forces
Here is your file attached. Just drag and drop over text edit in the post. Or "Add file" from attachment tab under. I think that there is 1MB limit per file but cant find where it was mentioned.
- Attachments
-
- forum-ex-clamp.FCStd
- (23.45 KiB) Downloaded 27 times
Re: FEM and multiple forces
Thanks !
-- Chris
-- Chris
Re: FEM and multiple forces
There is a displacement but it is very small compared to the other halve because of the way the fixed constraint is applied:
Re: FEM and multiple forces
This happens because of the way OCCT (Open Cascade Technology) kernel represents cylinders - see the seam line. So the resulting cylindrical surface is not a single face:
This is after properly fixing all faces:
You could also set Refine property of "clamp" which will make that surface as a single face to be properly fixed:
Re: FEM and multiple forces
Aha !
Thanks very much. I think I have it. As you can see I made the model by boolean addition and subtraction of shapes. When I click on the 'face' to make my constraint I'm clicking on the upper half of the inner surface. That selects only *part* of the inner surface due to the split. Hence as you say my constraint is acting on 270 deg of the inner surface, leaving the one leg unconstrained where the forces acting will cause maximum displacement.
Salutary lesson learned ...
Thanks very much for your time to explain this.
(To the forum admins ... this subject can be marked solved or closed or whatever)
-- Chris
Thanks very much. I think I have it. As you can see I made the model by boolean addition and subtraction of shapes. When I click on the 'face' to make my constraint I'm clicking on the upper half of the inner surface. That selects only *part* of the inner surface due to the split. Hence as you say my constraint is acting on 270 deg of the inner surface, leaving the one leg unconstrained where the forces acting will cause maximum displacement.
Salutary lesson learned ...
Thanks very much for your time to explain this.
(To the forum admins ... this subject can be marked solved or closed or whatever)
-- Chris
Re: FEM and multiple forces
You should edit your first post to reflect that.
Re: [Solved] FEM and multiple forces
Hello Chris,
Welcome to the forum.
As explained in the Install_on_Unix page, on Ubuntu we provide PPA repositories for the latest release as well as a development version called "daily" (it's not always built daily though). The advantage of using the PPA is that all updates go through Ubuntu's update system; and there is no duplication of libraries. If you install the freecad-stable PPA, you will automatically get the 0.18 release when it is published, which should be a matter of weeks.
Welcome to the forum.
Even when Ubuntu 16.04 was released 3 years ago, that package was outdated. The Ubuntu repositories are always behind.
As explained in the Install_on_Unix page, on Ubuntu we provide PPA repositories for the latest release as well as a development version called "daily" (it's not always built daily though). The advantage of using the PPA is that all updates go through Ubuntu's update system; and there is no duplication of libraries. If you install the freecad-stable PPA, you will automatically get the 0.18 release when it is published, which should be a matter of weeks.