You are not logged in. Please login or register.

Forum updated

We have recently updated our forums system, and we have cleared all spam topics and users. In case we accidentally deleted your account, please register again. If you miss a topic you have posted before, please let us know.

Post new reply

Post new reply

Compose and post your new reply

You may use: BBCode Smilies

All fields with bold label must be completed before the form is submitted.

Required information for guests


Required information

Topic review (newest first)

11

Now i reduced the Line elements as u said, but the size of one LS is still 756 Mb.
The model has 106K nodes and 98K elements.

My harddrive just has 1Tb

10

Hi Pavlo,

The problem is, that i cannot write the input file. It always gives me that message.
Message in line number 200. error: MinimalReinforcementGroup is 90000. Probably you dont have any reinforcement element

But maybe this one works:

https://gigamove.rz.rwth-aachen.de/d/id/WqgtcLQSsviXSq


If not could share my desktop via remote control to you if u got time for that.

Im using GiD version 14.0.5

btw thanks for the tip with the "store data for this Interval steps" didnt noticed that tool.

9

Hi Florian,

The model is not associated with ATENA problem type. I cannot see any boundary conditions, materials, but only macroelements and mesh. Perhaps some files are missing. Try to copy your original model and just delete calculation results from the subfolder "AtenaCalculation". Results are stored in files with extension *.00XX where XX is number of calculation step.

Regarding the model size. I would say that one step of your model (if it is defined correctly) shall demand 100-200 MB. Hence, 300 steps turn into 30-60 GB but not 200. I recommend you to do the following:
1) Define rebar meshing one FE for one straight bar. Rounded corners can be meshed with 3-4 FE. ATENA perform additional meshing of bars during calculation, see details in the Theory book.
2) Save each 5/10th step instead of each (Interval data settings)

I might have more comments if full model is available to me. By the way, which program version and GiD script version do you use?

8

I have finally solved all these problems, but now i just have 1 single error and i really really can't find the issue.
Ill upload my model and you might find the reason.

I have 4 different models( different in their geometry and reinforcement), each with a centrically loaded reference case and 4 different eccentric loads. strangely, this error only occurs with the models with centric loads but they dont differ from the others except their loading case. Perhaps through error propagation due to copying the base model.

The error if the following:
Message in line number 200. error: MinimalReinforcementGroup is 90000. Probably you dont have any reinforcement element

Another problem is that my Solution files are extremely large ~up to 200Gb for about 300 LC's. Is this normal or can this be reduced somehow?


Here is the link:

https://we.tl/t-PTS6mlDLgV

Its a zip but u can trust me big_smile

Thank u  I will mention you in my acknowledgements

7

Hi Florian,
Can you share your model?

6

Okay problem!

now i get this error in processing Error:
Warning: group id: 110002824, name:   uses geometry id: 5, name: CableBondGeometry #5 as name CC1DElastIsotropic with incompatible prestress load. The prestress load conditions prevail. (Source line: 227, file: ..\..\Sources\CCExternalCable\calculate_slips(CCExternalCable_3D).cpp, last modified at: Wed May 13 12:42:05 2020)

I did not put any prestress on the cables.

Im using the CCelasticisotropic reinforcement with type Cable.

and set friction and cohesion to zero.

whats wrong with my system now?

5

It finally works thank you very much !

4

You may try to use an "external cable" to avoid defining friction and cohesion

I cannot open the picture...

High strength is not needed if relatively low elastic modulus is defined. However, it doesn't matter

If the cable is connected in start and end points only, the strain in the bar has to be uniform along the length, so you don't need any "Strain for whole reinforcement". Strain will be same in any node.

About directions: as I said, you will need some post processing for the second case. You can determine strain along bar length by simple formulas

3

Thank you very much !
This helped me a lot!

I will try this macro recording, it seems like a nice gadget i didnt know yet.

I have another question regarding the type of cable material.
The gauge length is 5 cm and it measures the strain in its axis. The diameter increases at the end of the gauge as an anchor, shown in the picture: https://rwthaachende-my.sharepoint.com/ … p;e=Mlcyn8


I would take the "Cable" type with active ancor "both" and 0 friction, 0 cohesion. I would chose a low Youngs modulus and a relatively high yield strength to ensure the cable is not breaking. By anchoring both sides of the Cable I get
Is that what you thought about.

And then as you said: line -> monitor for reinforcement. But what type would u chose here because in the experiment we mesured the whole strain between the endpoints of the gauge. And i cant find this type of monitor like in this case: "Strain for whole reinforcement".

The other way by just defining the two endpoints of the gauge seems to be more problematic, because i cant set the direction of the strain measured, especially if its not in the x,y or z axis.

Thank u very much in advance!

Greetings

Florian

2

Hi Florian,

1) I think that there are 2 ways if your 16 models are already done.
a) Use the macro recorder in GiD to create a procedure with definition of all monitors and then run this macro in all models sequentially. I have recorded a video with instructions . You can do anything and record any actions in macro recorder including geometry creation.
b) add monitors directly into *.inp file. However, you will need to repeat this procedure every time after you modify a model in GiD. You can find to create any commands in inp file in the ATENA_Input_File_Format.pdf. However, I wouldn't choose this way since it is more complicated.

2) I couldn't understand what is the difference btw (i) and (ii). However, if your gauges are quite long and strains cannot be interpreted as point strains, then it makes sense to define a line. I would define a line with material reinforcement with very low modulus of elasticity and type unbonded cable.
Then there are 2 options for monitors: define strain monitor for reinforcement bar or define displacement monitors for start and end nodes of the bar. In the second option you will need some post processing but the result has to be the same.

Hope it helps you.

1

Hello Atena Team and members,

I have got 2 questions

1.)
Im supposed to to some investigations in punching shear behaviour of footings. I have 16 different models. Now I need to put in a bunch of monitoring points into the models which costs a lot of time. Due to the relatively similar geometry of the models I want to copy the entities Im examining from one model to another. This are point and line entities which i want to copy. The best case would be if i could copy whole layers and the Interval Data within the monitoing rpoints from one to another model but i think thats not possible.

2.)
I want to simulate experiments we did further. We put several small "pipes", that mesured the strain along their axis, into the concrete.  What method is most suitable to rebuild this situation.

I thought about this 2 different ways.

i.)
Modelling just Line entities: but what would be the best monitoring point type for that because there are some pipes with their axis not into one of the principal directions
ii.)
Modelling line entities and assign a material to them weaker than the concrete to not make the concrete stiffer



Im modelling with GiD

thank u in advance!

Florian