PDA

View Full Version : Ansys shell91



Theofanis Ampatzidis
11-10-2010, 02:36 AM
I am about to model our carbon fiber monocoque in ansys using shell91 and sandwich option included...

Is there anyone with prior experience to tell me how accurate results will give the analysis and any problems encountered?

Theofanis Ampatzidis
11-10-2010, 02:36 AM
I am about to model our carbon fiber monocoque in ansys using shell91 and sandwich option included...

Is there anyone with prior experience to tell me how accurate results will give the analysis and any problems encountered?

KeWLKaT
11-18-2010, 12:30 PM
You will fail miserably IF you don't know what you are doing. Make sure you understand how everything works.

Theofanis Ampatzidis
11-19-2010, 12:08 PM
Thank you for your advice. For now I am trying to do this, to understand as much as possible...

We actually intend to use the analysis to compare several solutions(number of layers,layer orientation etc...) after having done some real life experiments on sandwich laminates...

I have understood the basic stuff. Now I am trying to go deeper, such as failure criteria, examining each layer...

If someone has any experience, or any other solution for the analysis would be very helpful to make some comments, to give me some advice.

KeWLKaT
11-23-2010, 10:08 AM
I have experience with this, however I am using Altair Hypermesh. The one problem you will run into is the definition of the ELEMENT ORIENTATIONS (not to be confused with ply orientations....). There exist programs which are able to drape onto 3D models (i.e. FiberSIM), but are inaccessible (even illegally!!!). So you will need to break down your mesh manually and apply orientations with a bunch of local coordinate systems.

Theofanis Ampatzidis
11-23-2010, 10:35 AM
I was thinking of creating local coordinate systems for each area... It will be chaos, but I see no other way...

KeWLKaT
12-06-2010, 09:47 PM
It's not chaos as you put it, when done right. For my analysis it took a bit less than 40 different coord systems. Yes it takes time, but you do it once and you're done for good. I haven't played with Ansys a lot, but in HM you can divide your mesh into different components which helps a lot.

Theofanis Ampatzidis
12-09-2010, 02:35 PM
Actually I have made a model using local coordinates system, their number being around 200....

Anyway, I hope it worth the hours, and I am confident it does.