inventor 2012 - ifa - monodimensional analysis

[La Jolla]

I beg to have done the good junior user and to have checked that there were no open threads on the same topic so I open one.
It's my first post, so I'll be prolished and bearded.

Good morning to all, I was wondering if someone had had like me pleasure, it is done to say, to use the new module "ifa" for the monodimensional modeling of inventor 2012. I use it as it is the only one made available to the technical office in which I carry out my internship, so the things are two: or I make it enough, or I make it enough!
to what I understand is a package born for the analysis of frames, but as a good almost mechanical engineer I am prodigal to exploit it as a generic monodimensional fem tool, in short, for the validation of the celeberrimi "simplifying models to rigid rods" that all of us mechanics love to flaunt, perhaps along with a beautiful frozen beer:biggrin:
I am curious to know what those who are using it think and what, possibly, will respond to my perplexity; but now I come to my experience.
my very first question:
if they have decided to insert a module for the analysis to beam elements, why insert it as superstructure (it must necessarily do the skeletral, then cotrure the frame with the appropriate function and then throw the ifa, so change section = exit and remake the frame) and not as simple beam analysis (we only disengage lines, we assemble materials loaded sections cannoli to the ricotta and launch the analysis)?
I find it unnecessarily complicated. Am I the only one?
a problem on which I clashed relates to the lcs (local coordinate system), which are literally put to the stern and perjunta are not editable from the inside of the ifa simulation, to put them in a way to display worthy results in terms of internal actions and efforts you have to exit and change the orientation of the solid elements of the frame!! (I then found that the lcs put them on the basis of the view in which we are at the moment when we insert the frame pieces, with the z axis always directed towards the summit closer to our "click", therefore I recommend to put them always "in front" and then when necessary turn them of k π/2).
another race lap. efforts and internal actions.
a praise deserves the fact that you can finally to see internal actions, even if the sign convention is all the contrary and there is no option to settle this.
another thing I noticed is that...in the results there is no comparison stress! what the ifa calls "smax" (main- >smax resolutions) can deceive us, but nothing else is that the maximum axial effort, volgarmente sigma, calculated taking into account the flexing moments and the normal actions, but without taking into account the efforts of cutting, which can be seen precisely from the "cutting resolutions" or "torsio analysis".
This is to mean that among the results there is no solicitation of comparison...is especially in this regard that I hope to have taken a dazzle, would be a ridiculous lack (a rag of v.mises? )
last in order of importance but first in order of annoyance: I'll give you the cursed labels:
when a system is solved and we want to display internal actions\sforzi by diagram on structure 3d, two labels appear for rod, ugly and bulky and that often overlap, not to mention that when an action\sforzo is 0 (we understand, 10^-8/10^-12) it also marks it with a very showy label .
I hope I'm wrong here, but I couldn't eliminate them in any way. That's annoying.

in conclusion:

for those who understand me, I regret lusas to such a point that I regret the knots and put the fuchsia auctions.
in my opinion sin of maneuverability and customization; I see it very "blocked" and actually more suitable for simple analysis of frames, but maybe I have had too many expectations on a product that all in all is to its first version.

as we say in my parts: "Cunfusu cà pessuasu" (more confused than convinced)

 

[La Jolla]

I forgot an addition and an incorrect courier.
the wrong is that by adjusting the lcs you can have the diagrams agree with your own positive verse (e.g. fibers stretched vs tablets).
the addition is that ifa analysis does not contemplate curved lines. I hope that the polyliea will have to do with this.

 

[pompeo79]

I'm missing or I don't think I've understood the definition of "one-dimensional modeling" module.
the ifa module of frame analysis is not new to inventor 2012 but comes with 2011.
the frame analysis module basically has the same task as the stress analysis module, only more dedicated and with a different calculation engine.
the need and usefulness of the separation comes from the time of analysis and the structural simplification (faster, without acumens of tension, possibility to analyze large structural assemblies,etc)
the simulation and analysis areas of inventor are very valid and are destined to improve even more, especially in the perspective of the digital prototyping with which the autodesk is pushing a lot.
the results of ansys are still a little far away, but undoubtedly does a great job, considering that it is not the first job of inventor.

What do you mean "inserting it as a superstructure"?

Consider that you are not obliged to use the frame generator to then switch to frame analysis, it can also be a simple set of parts generated with the content center assembled with the classic constraints.
returning instead to the speech frame generator you can quietly, once placed the skeleton (skeleton 2d/3d), and assigned the irons manage the sections, moments of inertia, etc., always standing within the analysis environment.

I'm not clear about your observation "a problem I've met about the lcs (local coordinate system), which are literally put to the smurf."
and even "in praise it deserves the fact that it can finally see internal actions, even if the sign convention is all on the contrary and there is no option to fix it." I don't understand.



with the ifa no longer speaks, by definition, of tetrahedral elements but of beam elements, therefore in this perspective a profile has 2 knots and 6 degrees of freedom.
as I said before this abstraction allows both to simplify and speed up structural analysis without disconnecting too much from a real analysis.
within the frame analys therefore a profile has a length greater in proportion to its section and the loads it is intended for. (if this condition is not satisfied (a measure at least 10 times so much the section) so much it is worth using the stress analysis environment).
then we will talk about bending deformation and maximum flexing stress that will be the equivalent stress that most interests (> > smax= [max momento flettente/momento d'inerzia]*neutral axis cost ).

to tell us if you do a fem and ifa analysis of the same profile, with the same boundary conditions you will find a stress of von mises≈smax.

labels can be moved as desired or disappeared.

 

[La Jolla]

first of all thanks for the interested and exhaustive answer.

I'm missing or I don't think I've understood the definition of "one-dimensional modeling" module.
the ifa module of frame analysis is not new to inventor 2012 but comes with 2011.

for single-dimensional modeling I meant the beam elements that are, in fact, single-dimensional elements.
I'm sorry, I was updated to 2009 and I thought it was a 2012 novelty.

the results of ansys are still a little far away, but undoubtedly does a great job, considering that it is not the first job of inventor.

I agree.

What do you mean "inserting it as a superstructure"?

Consider that you are not obliged to use the frame generator to then switch to frame analysis, it can also be a simple set of parts generated with the content center assembled with the classic constraints.

So I must have lost something on the street, because when I try to launch the frameless annalisys says to me "impossible to find supported frame members"?

returning instead to the speech frame generator you can quietly, once placed the skeleton (skeleton 2d/3d), and assigned the irons manage the sections, moments of inertia, etc., always standing within the analysis environment.

I meant that I can't change the skeleton, first I used another approach and it's the first attempts with this modeler.
What do you mean by the "ferri"?

I'm not clear about your observation "a problem I've met about the lcs (local coordinate system), which are literally put to the smurf."
and even "in praise it deserves the fact that it can finally see internal actions, even if the sign convention is all on the contrary and there is no option to fix it." I don't understand.

For the convention I have already denied myself, I was obtuse. for the lcs I mean precisely that it puts them independently from which view is at the time of the "frame" command, and that they cannot be modified within the simulation, I cannot be oriented and I cannot be put local systems (e.g. spherical, cylindrical).

with the ifa no longer speaks, by definition, of tetrahedral elements but of beam elements, therefore in this perspective a profile has 2 knots and 6 degrees of freedom.

In this regard I wonder, but the knots I see, are they real?
In the sense, if I want to analyze a beam in ten beam elements, do I have to make ten segments?

to tell us if you do a fem and ifa analysis of the same profile, with the same boundary conditions you will find a stress of von mises≈smax.

I didn't understand that. So on a slender body an important torque moment should I not consider it?

labels can be moved as desired or disappeared.

Please tell me how!!

 

[La Jolla]

a doubt I solved it from me:
the knots are real, I tried to load a "top" element away from a knot getting no feedback on the deformed, adding a custom knot at the same point went differently, according to the theory at the base of the beam.

 

[pompeo79]

- if you use the frame analysis you have to use by force or by the center contained or by the frame generator, otherwise > stress analysis

-by definition when talking about beam analysis within the frame analysis members will have 2 knots and one direction. therefore we talk about simple monoaxial tension.

the formula of von mises therefore becomes smises=axial sygma and the smax is the maximum

-The labels will always be orthogonal to the observer. to move them just click on the label and move it, to delete them if they are maximum or minimum go to the "display" group and uncheck the entries. if they are labels placed by the user select them right button deletes.

 

[La Jolla]

but in the state of monoaxial tension is also contemplated the cutting effort, which if it exists is combined with the axial sigma (mohr, to understand us) to give the effort in the main direction (the only other than 0 in monoaxial), or is something wrong?
I still can't remove the labels from the diagrams, I don't find the "display" group. Would you be so kind to attach an image to me?

 

[La Jolla]

I correct my beaver, but I was at work and I read and responded a little too fast.
if on a frame (a rigid rod system, in my abstraction) I have a twist, and then a cutting effort combined with a normal, say that the smax=smises is not correct.
my question in fact was, why not put between the results a smises?

 

[pompeo79]

__________________________

 

[La Jolla]

I had already tried with those buttons, but unfortunately it does not delete the labels of the "diagrams", apparently I am one with the real diagrams, I can not even move them. .