Help with CalculiX *MASS card for my FEM models

About the development of the FEM module/workbench.

Moderator: bernd

musakucukali
Posts: 5
Joined: Thu May 07, 2020 2:48 pm

Help with CalculiX *MASS card for my FEM models

Postby musakucukali » Thu May 07, 2020 4:07 pm

Hello everyone,

I've been trying to create a clamped-free beam model with lumped mass at the free end in CalculiX but i got many errors. Then i tried to create spring-mass system to understand how i can create mass model at node, again i got error as:

Code: Select all

"*WARNING in calinput: no elastic constants were assigned to material MASS in a (thermo)mechanical calculation"
I couldn't get why it asks for elastic constants for the MASS which applied at a node. I need you guys to help me about creating FEM models with mass such as beam with lumped mass, spring-mass systems etc.

Basically i was gonna trying to create models as in the below figures:

Clamped-free(Fixed-free) beam with lumped(tip) mass

Image

Wall or fixed body attached with Spring+Mass+Spring+Mass model

Image
Jee-Bee
Posts: 2025
Joined: Tue Jun 16, 2015 10:32 am
Location: Netherlands

Re: Help with CalculiX *MASS card for my FEM models

Postby Jee-Bee » Thu May 07, 2020 4:36 pm

according to: http://www.feacluster.com/CalculiX/ccx_ ... de272.html
for the mass there is nothing needed. but for the beam there are elastic properties needed(basically a material card).
But fore more help you need to attach a FreeCAD file otherwise we are unable to help because we don't know what you are doiing
musakucukali
Posts: 5
Joined: Thu May 07, 2020 2:48 pm

Re: Help with CalculiX *MASS card for my FEM models

Postby musakucukali » Fri May 08, 2020 9:29 am

Jee-Bee wrote:
Thu May 07, 2020 4:36 pm
according to: http://www.feacluster.com/CalculiX/ccx_ ... de272.html
for the mass there is nothing needed. but for the beam there are elastic properties needed(basically a material card).
But fore more help you need to attach a FreeCAD file otherwise we are unable to help because we don't know what you are doiing
Hello Jee-Bee,

Thank you for your message, I attached the files that i'm working on it and get the errors i mentioned.

The file which is named as" MS(wmass).inp", is created based on this url:
https://github.com/mkraska/CalculiX-Exa ... ete/MS.inp
I closed some lines by putting "**" because i thought i don't need them. Also i replaced beam elements by *Mass at node due to instructions of the web page sent by you.

The other file which is named as "beammass(fixed).inp", is refined for its boundary. The issue with it is, CalculiX doesn't get some cards of FEM program properly. Therefore I might need to replace some element types and other cards(like *boundary) as how are they working in CalculiX. I'm not sure which ones need to be replaced so i'm going on step by step once I figured what does error mean.

Regards
Attachments
Masses.rar
(2.97 KiB) Downloaded 4 times
fandaL
Posts: 370
Joined: Thu Jul 24, 2014 8:29 am

Re: Help with CalculiX *MASS card for my FEM models

Postby fandaL » Fri May 08, 2020 7:51 pm

I suppose that you are doing frequency analysis of the model in your picture. See the attached analysis in FreeCAD where are few beam elements and manually added mass element to *.inp file above *STEP card:

Code: Select all

** manually added mass element with id 33 at node number 22
*ELEMENT, TYPE=MASS , ELSET=EM
33, 22
** definition of 0.0035 mass units for all (here one) mass elements in EM element set
*MASS,ELSET=EM
0.0035
1d_beam_frequency.inp.txt
(3.48 KiB) Downloaded 8 times
1d_beam_frequency.FCStd
(8.83 KiB) Downloaded 6 times

There were some scripts to convert inp from Abaqus to CalculiX, but I have not used them and I think they will not work on 100 %. You can compare your Abaqus file with inp file from FreeCAD to get some idea about differences.

Example from https://github.com/mkraska/CalculiX-Exa ... s/Discrete is about dynamic response which calculates a response on the load in time. Whereas frequency analysis (example in FreeCAD) is much simpler and gives only natural frequencies and shape (not real stresses or displacements). So you should decide what is needed in your application.

I had some other issue with FreeCAD precision in inp file where I had to change line for orientation of the beam section

Code: Select all

*BEAM SECTION, ELSET=M0B0RstdD0, MATERIAL=MechanicalSolidMaterial, SECTION=CIRC
1.0
-6.123233995736766e-17, 1.0, 0
to

Code: Select all

*BEAM SECTION, ELSET=M0B0RstdD0, MATERIAL=MechanicalSolidMaterial, SECTION=CIRC
1.0
0, 1.0, 0
@bernd:
Could you look at the issue with precision? The number -6.123233995736766e-17 is too long so that CalculiX 2.16 gives

Code: Select all

*ERROR in calinput: at least one fatal
error message while reading the
input deck: CalculiX stops.
E.g. 6.123233995736766e-17 without minus works, or -6.12323399573676e-17 with the last digit dropped works as well.

OS: Linux Mint 19.3 (MATE/mate)
Word size of OS: 64-bit
Word size of FreeCAD: 64-bit
Version: 0.19.20943 (Git) AppImage
Build type: Release
Branch: master
Hash: ceb23799c76df3ebfa7be4b9fe83bb62de60bc6c
Python version: 3.8.2
Qt version: 5.12.5
Coin version: 4.0.0
OCC version: 7.4.0
Locale: English/United States (en_US)
fandaL
Posts: 370
Joined: Thu Jul 24, 2014 8:29 am

Re: Help with CalculiX *MASS card for my FEM models

Postby fandaL » Fri May 08, 2020 10:02 pm

musakucukali wrote:
Thu May 07, 2020 4:07 pm
Hello everyone,

I've been trying to create a clamped-free beam model with lumped mass at the free end in CalculiX but i got many errors. Then i tried to create spring-mass system to understand how i can create mass model at node, again i got error as:

Code: Select all

"*WARNING in calinput: no elastic constants were assigned to material MASS in a (thermo)mechanical calculation"
I couldn't get why it asks for elastic constants for the MASS which applied at a node. I need you guys to help me about creating FEM models with mass such as beam with lumped mass, spring-mass systems etc.
This error is really strange in case of mass elements, since they should not need elastic constants. Surprisingly, it disappeared when I added common beam elements which have elastic material.
ms_with_beams.inp.txt
(1.15 KiB) Downloaded 6 times
musakucukali
Posts: 5
Joined: Thu May 07, 2020 2:48 pm

Re: Help with CalculiX *MASS card for my FEM models

Postby musakucukali » Sat May 09, 2020 8:07 pm

Hello @fandaL,

I do appreciate for your effort and your help,

Please let me tell you my main purpose to do these analysis and where I'm standing for now. I'm working on "mounted engine on wing" mass coupling modelling of an Aerostructural model whose link is in below.

http://mdolab.engin.umich.edu/sites/def ... rint_0.pdf

Figure 1, to let you imagine and understand the model of engine on the wing coupled mass:

Image

I've the model's data, geometry and mesh files which are created in Nastran/Patran software. I want to work this job on CalculiX. So I converted the files first from Nastran/Patran to Abaqus and then from Abaqus to CalculiX. After that, when I run the files on CalculiX, I got so many errors. When I closed *mass cards which are for modelling the engine coupling, I was able to run it but it wasn't useful for me. Therefore to make sure that CalculiX is capable to work on *MASS card, I wanted to create a more simple model as cantilever beam with lumped mass(tip mass) like a wing from front view. My goal with this simple model was to understand I can do mass work on CalculiX. Then i researched manuals, examples from other websites like github and found the same link you sent. It was a spring+mass system however I needed a more capable model for my work. I created a beam with lumped mass in Abaqus and converted it to CalculiX. The beam model that i wanted the create is in the below Figure 2. I have it's frequency analysis results and I wanted to create the model and make the analysis on CalculiX to compare to make sure that I can make mass (point mass,lumped mass, tip mass) analysis on CalculiX. Therefore I can say that in my application, I just need the natural frequency values to compare results.

Figure 2:

Image

I couldn't make it and asked for others help like you. Thank you so much to you and others' interest.

At this point, I have some questions for the forum right now:

Is CalculiX capable to work point mass, lumped mass projects like i've been trying to create as I mentioned in my recent posts?(I saw many examples which were created masses as beam element on CalculiX and their creators claimed that it is the same as working point mass or lumped mass jobs, not mass element that's why i'm wondering about *MASS card and CalculiX's capability of mass works.)
musakucukali
Posts: 5
Joined: Thu May 07, 2020 2:48 pm

Re: Help with CalculiX *MASS card for my FEM models

Postby musakucukali » Sat May 09, 2020 8:23 pm

Hello,

I forgot to attach the document that i'm working on. I'm attaching the .inp file of whole Aerostructural model to let you see all the work I need to do. I opened the first MASS element for you to let you run it and see the errors that i got. I also want to thank you again for the .inp files that you downloaded because it helps a lot to me to understand that I can do mass work for at least the simple model as first step of my whole work.

For now, if i get any errors again, i might need your help again as i mentioned to create that simple beam model with tip mass to compare with the analysis results for the natural frequencies.

https://www.researchgate.net/publicatio ... ned_method

Image
Figure: simple beam model with tip mass

Attachment: Whole job's input file
Attachments
run_uCRM-135_modal_coarse_abaqus_deneme.rar
(216.06 KiB) Downloaded 3 times
fandaL
Posts: 370
Joined: Thu Jul 24, 2014 8:29 am

Re: Help with CalculiX *MASS card for my FEM models

Postby fandaL » Sun May 10, 2020 9:44 pm

My understanding is that mass element with *MASS property card is a mass concentrated at node, which I guess is what you need for the simple wing and engine model, supposing that beam or shell elements will simulate structural stiffness, and mass elements will simulate (structural) mass.

For frequency analysis structural masses of ordinary elements (including beam elements) are given by their geometry and material density as it is e.g. in original CalculiX example beamf.inp. Just for comparison I tried to remodel it with beam elements in one model; and in the second model with beam elements with 0 density and separated masses at nodes (by mass elements), which gives similar results to original reference results, but not exactly same.
beam_with_density.inp.txt
(3.33 KiB) Downloaded 11 times
beam0density_and_nodemasses.inp.txt
(3.64 KiB) Downloaded 13 times
I have not much experience with masses in CalculiX, so if you would need some deeper help in this regard, you can ask on CalculiX mailing list https://groups.yahoo.com/neo/groups/calculix/info
musakucukali
Posts: 5
Joined: Thu May 07, 2020 2:48 pm

Re: Help with CalculiX *MASS card for my FEM models

Postby musakucukali » Sat May 30, 2020 3:18 pm

Hello fandaL and everyone,

I want to thank you and the others so much for your help. As you tried to model the elements like beam element and mass element like I've been trying to do the same process, your input files that you uploaded, meant a lot for my work. I got where I'm, how can I put that in right way and now I think, I'm able to go further steps for my work. I appreciate for your time, help and kindness. Hope you and the other guys in the forum the best and health.

@fandaL sorry for my late respond, I was struggling with my online midterm exams. :roll: :)

The problem is resolved, authors can carry the subject to archive.
User avatar
bernd
Posts: 10179
Joined: Sun Sep 08, 2013 8:07 pm
Location: Zürich, Switzerland

Re: Help with CalculiX *MASS card for my FEM models

Postby bernd » Wed Jun 03, 2020 11:15 am

fandaL wrote:
Fri May 08, 2020 7:51 pm
I had some other issue with FreeCAD precision in inp file where I had to change line for orientation of the beam section

Code: Select all

*BEAM SECTION, ELSET=M0B0RstdD0, MATERIAL=MechanicalSolidMaterial, SECTION=CIRC
1.0
-6.123233995736766e-17, 1.0, 0
to

Code: Select all

*BEAM SECTION, ELSET=M0B0RstdD0, MATERIAL=MechanicalSolidMaterial, SECTION=CIRC
1.0
0, 1.0, 0
@bernd:
Could you look at the issue with precision? The number -6.123233995736766e-17 is too long so that CalculiX 2.16 gives

Code: Select all

*ERROR in calinput: at least one fatal
error message while reading the
input deck: CalculiX stops.
E.g. 6.123233995736766e-17 without minus works, or -6.12323399573676e-17 with the last digit dropped works as well.
git commit ab0965872e