You can check the updates installed in your machine as per
Forum » Modeling (Preprocessing) » HyperMesh » To identify the software update/patch that has been installed in your machine
http://www.altairuniversity.com/support-forum/?mingleforumaction=viewtopic&t=252
you need update 201 or above in version 11.0 to be able to use the MAT9ORT card.
↧
Re: MAT9ORT
↧
Re: MAT9ORT
Mat9ORT is listed in the material card image list when you are creating a new materialthat is strange. for me the type is all and the card image is only MAT1 to 10
↧
↧
Re: MAT9ORT
ok thanks
↧
Re: Get mass and stiffness matrix of a FE-model
Dear Uwe Fechter,
I just talked to our support team. The information you are looking for is available inside the HyperWorks suite (e.g. coupling of flexible bodies in MBD). However, the information can't be exported.
Is our understandig correct, that you successfully completed the first part (fluid structure interaction using a reduced FE model). How did you create the reduced FE model?
In the second step you want to check whether the reduced model ws using "reasonable" values?
Regards
Matthias Goelke
↧
Re: MAT9ORT
I saw in an other post that there is a conversion methos to obtain the orthotropic MAT9 entries:
the conversion from MAT9 to MAT9ORT or the otherway is different. below picture is from the card of MAT9ORT. the G11 in the picture is the C11 of the stiffness matrix of orthotropic material (assuming C as stiffness and S as the compliance matrix notations). Imagecould you tell me where to find this? didn't find it myself in the reference guide
↧
↧
Re: Element type of Honeycomb core in Composite Optimization
Hi ynishio,
In general, sandwich structures are best modelled with a solid layer in between two shell layers (even better would be all solid layers)
All shell is a reasonable approximation for concept design.
Shell-solid-shell is widely regarded as the most accurate approximation for detailed sizing.
All Solid (layer by layer) is clearly the best approximation, but difficult to mesh/build correctly.
In the case that you do not have the modulus in a particular direction unavailable , just use mat1 as in the tutorial for the core, In case your core is also anisotropic, you can use psolid and MAT9ORT, etc so there are several ways to model this as per requirements.
I see that you will be using isotropic core with anisotropic layers above and below it, Please try with a simple model like a plate and we can discuss further.
↧
Re: Get mass and stiffness matrix of a FE-model
Hi Uwe,
I am afraid it is not possible to output the stiffness matrices of a full model and then use in an external program.
You can only output the matrices to a punch file for instance, of a reduced model, as you have done already using the PARAM extout control card.
↧
Re: Get mass and stiffness matrix of a FE-model
Uwe,
Can you also please explain why you would need to compare the full model with the reduced model? this defeats the purpose of the reduction, you can verify if the reduction technique is correct using a simple model as in the tutorial below
http://www.altairhyperworks.com/hwhelp/Altair/hw12.0/help/hwsolvers/hwsolvers.htm?optistruct_tutorials.htm
Also please explain why, to calculate structural behaviour of the reduced and full models you would use an external CFD solver and not HyperWorks structural solvers. you can solve the reduced and full models to compare the results directly here.
↧
Re: Get mass and stiffness matrix of a FE-model
Hello everybody,
thanks for the quick replies to my problem. To answer your questions:
Is our understandig correct, that you successfully completed the first part (fluid structure interaction using a reduced FE model). How did you create the reduced FE model?It is correct that I managed to export the reduced matrices with Hyperworks. The coupling of fluid and structure is working in some cases, but still needs improvements. I used the PARAM EXTOUT control card for the export (like Rahul Ponginan said).
you can verify if the reduction technique is correct using a simple modelI also used a simple model (a simple beam bending case and a beam oscillation).
please explain why, to calculate structural behaviour of the reduced and full models you would use an external CFD solver and not HyperWorks structural solversI work on a project which has the goal to implement a solver for reduced FE-models in the open source software OpenFOAM (CFD-code for the simulation of the fluid flow). So that the reduction can be done once at the beginning and no extra program has to be used for the calculation. This should provide a fast simulation without the need of an interface to a FE program. I would like to compare the accuracy of my reduction process (for example how many points do I need for the simulation of a complex geometry, like an airplane). Therefore it would be nice to simulate a full FE-model, too. The second thing I wanted to compare is simulation time. So how big is the benefit of reducing a model in terms of simulation time and is the simulation time for the full FE-model significant compared to the CFD solution. Regards Uwe Fechter
↧
↧
Warning 2617 ,2625
Hello together,
after starting a frequency response analysis I got the error message "warning 2617 and 2625":
*** WARNING # 2617
CONTACT interface 203 has 21 Slave nodes that belong to some of
Master segments, indicating likely self-contact condition. This may result in
poor quality of contact elements, redundant contact conditions, and patchy
pressure distribution, especially on curved contact surfaces.
*** WARNING # 2625
CONTACT interface 203 includes areas that have self-contact condition
and apparently pre-penetrations. To avoid detecting false pre-penetrations
(which may appear across thin solids with self-contact on both sides)
declare SRCHDIS smaller than minimum thickness of respective solids.
Or, if it is known that there are no true pre-penetrations in
self-contact areas, use CONTPRM,SFPRPEN,NO.
I am using several interfaces in that model between components of different stiffnesses in all cases "freeze" option.
Is this maybe causing the problem?
↧
Err: 5
Hey,
I used Automesh to surface mesh my modell. If I use CFD Tetramesh now, i get a weird mesh at the boundarys. You can see elements coming out of the volume and "needles".
So I saw that I didn't delete the solid and the duplicates. So I fixed that and the free edges. Now I got a enclosured Volume but when I try to use CFD Tetramesh now I get the following error:
Err: 5 vol encolsures has volume= 0.0
I also can see some red dots in the model now. But it is not the "free edge" red.
Does this something have to do with "unspecified thickness" ?
Thanks for your help,
Burkey
![Image]()
↧
Re: Problem with constraint force output using MATE function
Hi
Meanwhile, I could solve the problem. For me, it looks like a bug in MotionSolve:
The MATE function is delivering only zero constraint forces if one of the bodies used by the advanced Constraint_MATE joint is GROUND (how it happened in my model).
Best regards
Matthias
↧
Re: MAT9ORT
Hi,
Please check the RHO filed in the MAT9ORT. If the RHO is or blank then the mass is shown as zero.
↧
↧
Re: Warning 2617 ,2625
Hi,
check if the master and slave components share the same node(s) in the model, if yes then detach the components and create separate master surface and slave nodes. Also check if there are any element penetrations in the model.
could you please explain what exactly are you trying to analyze?
↧
Re: Warning 2617 ,2625
Please refer to the link below, the warning ID maybe different but deals with similar issue:http://www.altairuniversity.com/support-forum/?mingleforumaction=viewtopic&t=434
↧
Re: help with non linear materials, acceleration vs time loading, and more block tutorials in general.
Hi,
Sorry for a late reply, yes, card edit the component and make it to rigid and check.
the independent rigid node shouldn't be shared by any other component. In which case translate the node and try applying load on it (in your case velocity)
yes, you can create a rigid wall and apply velocity to the rigid wall.
go to Analysis page-->Rigid walls--> create a rigid wall and after that you can add a motion in the same panel.
↧
Re: Err: 5
Hi Burkey,
This again indicates a problem with the input geometry or input mesh, when you zoom into these red areas you will again see shell elements that do not really allow creation of tets etc
Are you using the solids option in CFD tetramesh? or are you meshing the surface and then using tetramesh?
Please send your current geometry using the dropbox below
↧
↧
Re: MAT9ORT
roh wasn't blank. but it doesen't matter anymore. just converted the parameters to MAT9 by my self :)
thank you
↧
Connection 1D bar with 2D shell elements
Hello together,
I'd like to connect shell elements with the mesh of an 1D bar elements.
How do I manage that?
I meshed the 2D and 1D elements individualy so far.
I created the bar at the edge of the shell component (black mark)
![Image]()
↧
Re: Connection 1D bar with 2D shell elements
Hi,
if the node count of 1D elements and 2D elements (at the edge) are same then you can equivalence using Edges panel.
go to tool page-->Edges-->select 1D and 2D elements and click equivalence.
if the node count is different you may consider using rigids between 1D and 2D elements
↧