🎉Community Raffle - Win $25

An exclusive raffle opportunity for active members like you! Complete your profile, answer questions and get your first accepted badge to enter the raffle.

Join and Win

Progressive Failure

mvass

Following my (deleted) post on RADIOSS, I performed an implicit solution of the problem using optistruct and it went surprisingly well. Right now I am not in a position to tell if the results agree with those obtained from a different solver (NASTRAN) however they look rational. What I would like to know is if it is possible to have results related to composite damage using optistruct. The problem was a simple progressive failure analysis and within the MAT8 card I entered the values for Xt, Xc, Yt, Yc, S (P12C, P12T I couldn't locate). However in hyperview I couldn't see any option related to 'damage' although in the relevant card (global_output_request) 'damage' option was selected.

Should you need any additional information, I can provide you the .fem input file.

PS: I made a run of the simulation by including 'Energy' and 'Failure' options in the output request. In hyperview there were options for failure, but no results. My failure options are related to Puck failure criterion and not Tsai-Wu. Is that why I didn't see anything in HyperView? Is there a way to include this criterion (Puck) within the MAT8 card?

Find more posts tagged with

English

OptiStruct

Accepted answers

All comments

Altair Forum User

Hi @mvass

DAMAGE card if valid for Fatigue loadcases. So, no wonder if you don't see any results for damage in HyperView.

How you selected the failure theory for your implicit run?

Can you share the model file? Use this File Transfer LINK

mvass

Hello.

thank you for your reply. Yes, in my effort to check for any results related to composite failure I accidentally selected options that are irrelevant to my problem. As said before, the 'energy' and 'failure' requests provided options related to failure in Hyperview, however there were no results there. In my humble opinion this has to be related to the fact that the MATX option within MAT8 was not used. But, (correct me if I am wrong) MATX accepts only TSAI-WU failure criterion. Is it possible to have additional failure criteria (user defined) or not?

I'll be glad to share the model with you (.fem file). I'll do that in a few hours.

Best regards.

Altair Forum User

Hi,

CFAILURE is the card which is used to request failure index, etc... But I dont think it is a valid card for implicit analysis.

When I have your file, I will check and update my views with you.

mvass

Hello. I have just sent you the .fem file. Please have a look and we can discuss over it. Remember: This is my first attempt with optistruct as an implicit solver. The example made is based on a model solution found here.

Thank you.

Altair Forum User

Hi,

thanks for sharing the file. I shall take a look at it and will reply soon.

Altair Forum User

Hi @mvass

CFAILURE output is only available for static analysis.

I will check if there is any other option available to output for implicit run.

mvass

Hi Prakash.

thanks for your reply and the overall effort. I am thinking that it is more appropriate to perform an implicit analysis within RADIOSS block instead of working within OPTISTRUCT bulk. As I have seen in the limited time that I had (unfortunately I do not have access to other version of RADIOSS than version 10.0), within RADIOSS block 13 or more you can use /FAIL/ property with the Puck failure criterion together with /MAT25 for the composite shells layup. Thus if it is possible for you, could you please give an example of a MAT25 card with a failure criterion associated with it (use imaginary values, I just want to see which flags in the respective cards are needed to be defined) and I'll set up the analysis. In the meantime, have a look on the implicit analysis options.

Best regards.

Altair Forum User

Hi Mvass,

Yes, PUCK failure is available in RADIOSS but not in OptiStruct for now. I will see if there is any model with MAT25 which I can share with you.

mvass

Hi, I received your email. I'll have a look. Thank you.

mvass

Hi again,

I am sorry for this 'questions bombing' session, but how do you access the /IMPL cards in RADIOSS? Are they manually entered in the _0002.rad file, or there is a specific menu in the program? If I recall well in the control card you can find only an /IMPLICIT option without any flags...

Thank you.

Altair Forum User

Hi,

IMPL should be available in Control Cards. Check for ENG_IMPL_xxxx

<?xml version="1.0" encoding="UTF-8"?>

mvass

Thank you Prakash. I think I'll be in a position to setup the problem in RADIOSS (implicit) and sent you the .rad files. I just need some time to get to a workstation having installed a newer version than the one we are currently using.

mvass

Following this lengthy discussion (btw thank you once more for your effort) there is one question that I would like to address which might interest other users as well, irrespective of the problem in discussion; I noticed that for example material law 25 has numerous flag entries that in most cases confuse users (especially those who migrate from other FEA packages). To be more specific: If we would like to define our material properties for a composite layup like the one I am working together with a failure criterion like PUCK, in the material definition we only need the values for E11, E22, v12, G12, G23, G13 & rho (optional) to perform our analysis and nothing more (supposing that one does not need to include temperature effects, structural damping coefficients etc). Is this assumption correct or not? Why does RADIOSS might need a plethora of addition information if everything needed is given in the FAIL/PUCK property card? Do one needs to include data for the TSAI-WU or crash options in the material card when they are not needed?

Thank you.

Altair Forum User

Hi Mvass,

It actual depends on complexity of the problem. If you want to analyze thermal, moisture, etc.. how this effects your design, YES then you have to include in your model.Few examples you can consider where temperature, moisture plays a role is designing of turbine blades. Here the moisture decreases the life of the blade through moisture diffusion. Similarly temperature has its own effect.

Rahul Rajan_21763

One can carryout Design of experiment considering such Parameter (Temperature or Moisture).Hyperstudy is tool under Hyperworks which is used for DOE,Optimization,fitting & stochastic studies. Refer Hyperstudy tutorials & user guide for detailed information.

mvass

Thank you all for your replies. As said before, I am interested only on performing an analysis where temperature or any other factor besides an enforced displacement does not affect the problem. Thus my question if all these mat law flags are need to be filled. Hopefully this weekend I'll complete the implicit analysis within RADIOSS using MAT LAW 25 and FAIL/PUCK property, so in case I come up with any problems I'll let you know. The resulting .rad file will be made available so every interested user may learn from this effort.

Edit:

I managed to take my work where there is radioss 13.0, but unfortunately there was no joy... (as expected). Strangely it returns an error message that the quad elements I have used cannot accept MAT25 and the property 11 defined. Probably there may be other mistakes but this one is really intriguing.

@Rahul Rajan: I have uploaded a .rar file with the two .rad files included. Please have a look.

Edit 2 (15/10/16):

Made some corrections. @Rahul Rajan: Please ignore the last file (.rar) I've sent. I'll contact you again with a revised version of my study. Sorry about that.

mvass

Hello,

after making a few tries in my limited time, I can say I have reached to a point of understanding a few things which I'll be happy to share them with other fellow users in the near future. For example I am answering a previous question of mine for the minimum flags that have to be entered in the MAT25 card. These are: E11, E22, rho, v12, G12, G23, G13, sigma1_yield_t, sigma2_yield_t, sigma1_yield_c, sigma2_yield_c (where t: tensile, c: compressive).

The good thing is that my implicit simulation within RADIOSS starts. But it does not converge. Unfortunately I didn't have the time to check in depth why. If possible, please try yourself and let me know. The 'final version' of my efforts is uploaded to Rahul Rajan's Filedrop. Once we'll have a working solution I'll post a simple guide on how to perform this analysis based on the experiences that I have gained.

Thank you all for your kind assistance. I am looking forward in reading your expert opinion.

Altair Forum User

us my question if all these mat law flags are need to be filled. Hopefully this weekend I'll complete the implicit analysis within RADIOSS using MAT LAW 25 and FAIL/PUCK property, so in case I come up with any problems I'll let you know. The resulting .rad file will be made available so every interested user may learn from this effort.

Hi Mvass,

You don't have to enter the temperature, moisture related params if you don't want to include their effects.

With the file which you had shared, we will check and update to you soon...

mvass

Hi, thank you.

Yes, as mentioned in a previous post of mine, I am aware of what I have to enter in the MAT25 card. No problem with that. Just make sure you check the file named 'PFA_final_version' that I have uploaded. The analysis runs, but it does not converge, therefore you have to examine my implicit solution scheme parameters. Sorry I didn't have the time to go through that...

Altair Forum User

Hi,

Can you try the following work arounds...

<?xml version="1.0" encoding="UTF-8"?>

mvass

Hi Prakash,

thank you for your suggestion. I tried to change some parameters in the implicit scheme, but as said before I do not have continuous and unrestricted access to RADIOSS 14 (or 13 can't remember). In order to avoid having a huge thread on this subject, please have a look on the file I' ve sent; the solution is almost complete and if you make the necessary modifications (based on your experience) you can come up with a fully working example which will benefit everybody. If you have the time, have a look, make any modifications and share with me your thoughts. In the meantime I'll work on the problem and check if I can solve it without adding any more comments on the present thread.