normal effort

[stef_design]

Hey, guys.
I often see, on fem books or tutorials, that among the results of a fem analysis almost always appears the normal stress (according to a certain axis) and a few times the equivalent stress (von mises).
is it no longer useful for design purposes?
Thank you.

 

[peloritano]

having the normal effort is useful to have the normal effort if it comes to slender beams......
In this case von mises does not serve because even if the sunscreening falls into the admissible field could instead cause unacceptable flaws

 

[peloritano]

I've written slightly with my feet but I think the concept is understood... if it's a slender beam, you have to make another check because von mises is not enough.

 

[ilaria85]

I believe that the tutorials are intended to teach you how to perform the analysis, not how to analyze the results. probably that's why they only display normal effort. However, in each software you can see the results that interest you: normal effort, of von mises, I cut you compared to x,y,z, deformation, etc. obviously von mises is fine almost always.

 

[MacGG]

Hello stef_design,

In addition to ilaria85, I also agree with hairy, indeed I would say that nowadays vonmises has always less the idea of what is the state of real stress. In fact, if we are in rhetorical systems (rings), vonmises goes well along the various beams, but in correspondence of couplings, it is necessary to valuate the main tensions and also the tensions in certain directions according to how the component has been realized (welding direction, types of plastic processing, etc.).

Since the fea has always been used for complex structures, it follows that other criteria are always used for the valuation of the tension state.

Bye-bye

 

[Fulvio Romano]

Guys, watch out. We are making statements at least bizarre about von mises.
the doubt of stef_design is lawful. to know if the piece breaks, if the piece is steel, you have to listen to von mises. point.
in the hypothesis of de saint venant (not always verified, otherwise than crap use to make the fem?) normal effort is very close to the equivalent sigma of von mises, because the cut is usually negligible. Maybe that's why it is used in tutorials.

It's pretty useless, but I feel like making a statement. normal effort in the elastic line and normal effort coming out of the fem are two distinct things...clear not? in a beam supported and loaded with bending, the normal stress of the elastic line is zero, the fem will instead give a great value.

for those who did not or hurt construction science I would like to make two notes on von mises:
When we have some material loaded, we have an exhaustive tension state inside, called cauchy tensions. There are three sigma and three tau. depending on how you look at these tensions (angle) some components turn into others because they are projected on different planes (see mohr circles). this tension state is "true", in the sense that it is the only thing really present within the material. the equivalent sigma, according to any criterion are not "true".

Our good friend von mises asked one day: "How do I know when the material is in crisis? I can't put myself to look at six components that if I prone my head to frogs, I would need only one number, function only of the load and point in analysis, which gives me an idea of the proximity of the material to the crisis". Well, our friend found a valid criterion only for steels that starting from cauchy tensions throws out a number, which nothing has to do with tensions, but as it has the size of a pressure, and it is a punctual feature of the material, it was called "equivalent scrum according to the criterion of von mises" or in jargon " von mises scrum".
the criterion is based on the observation that a thin test, placed in traction, breaks to cut (for sliding) on a plan to 45° regarding the load. from this, with some accounts on mohr corollars you get the formula that everyone knows, but that obviously not everyone understand and above all not everyone knows how to use.

so much to make understand how fragile this formula is, and how dangerous it is to use it without stopping to reason on the theory, if you take a more cynical audition, that breaks "to cup and cone", then with a sliding area and a decoesion, the above formula is no longer worth! (although it remains conservative).
Moreover, the criterion of von mises applies only to steels. use cast iron? Can't you use it, plastic? less. land? Concrete? worse than going to night! reinforced concrete? that even follows two different laws, one for compression and one for traction.

good times when the accounts were done by hand, and von mises preferred at least in the first line to use the trespass criterion. It's easier and conservative than von mises...

 

[JUMPY_75]

hello to all I find very interesting the open discussion and I wanted to intervene by asking all of you a question:

but did you ever doubt that the fem you use gives correct results? ? ? ?

I design simple machines and carpentry so I did not invest in a specific fem as I can perform hand-checks, but sometimes I found myself in need to examine the behaviour (especially for deformations) of structural sheets or details of complex form using fem modules integrated with cad. in the end I often wondered (especially by reading the writings in small... you decline any responsibility...) if the results are correct. especially by making two accounts by hand sometimes there are appreciable differences because perhaps a different logic of verification tends to simplify the process and then I ask myself:

Do I have to trust me, my instincts, the numbers written on the card or the fem? ? ?

My question seems stupid but collaborating with other technical studies I see no one asks these questions. everyone is blindly trusting software even those who are more qualified than me in this matter, but unfortunately I have often seen software wrong in 15 years of work......... .

 

[ilaria85]

hello to all I find very interesting the open discussion and I wanted to intervene by asking all of you a question:

but did you ever doubt that the fem you use gives correct results? ? ? ?

I design simple machines and carpentry so I did not invest in a specific fem as I can perform hand-checks, but sometimes I found myself in need to examine the behaviour (especially for deformations) of structural sheets or details of complex form using fem modules integrated with cad. in the end I often wondered (especially by reading the writings in small... you decline any responsibility...) if the results are correct. especially by making two accounts by hand sometimes there are appreciable differences because perhaps a different logic of verification tends to simplify the process and then I ask myself:

Do I have to trust me, my instincts, the numbers written on the card or the fem? ? ?

My question seems stupid but collaborating with other technical studies I see no one asks these questions. everyone is blindly trusting software even those who are more qualified than me in this matter, but unfortunately I have often seen software wrong in 15 years of work......... .

I actually agree with you. between paper and software I trust more than paper. if you need to check a simple beam do first with paper and pen, if you need to study complex structures you have an avalanche of problems: alive edges, welds, loads located at one point... let's say that the only usefulness of the fem is more than anything else to understand where you have a greater concentration of tension but on the quantity it is always better not to trust!

 

[Fulvio Romano]

but did you ever doubt that the fem you use gives correct results? ? ? ?

... let us say that the only usefulness of the fem is more than anything else to understand where you have a greater concentration of tension but on the quantity it is always better not to trust!!

the fem always gives the correct result, and never wrong. He can't be wrong.
the fem starts from a pattern, applies a theory, and provides a result. the result is correct. software is not wrong.

Of course, if I don't know the fem theory, I'm going to miss the model, I'm going to make the parameters wrong, I'm going to miss the questions to ask the fem. the fem does the calculations on the model, not on reality. If I'm wrong, I can't blame the fem if his numbers are different from reality.

Unfortunately there are so many people convinced that they can take a degree in engineering studying on wikipedia. Luckily, it's not.
the fem is for the engineer what the calculator is for the accountant. I with a calculator in my hand are not able to do the commercialist, I do not see for what reason fior fior of graduates think that cra@ando the fem module exited from the dixan fustine can dimensional the shell of a nuclear reactor! :angry:

 

[ilaria85]

the fem always gives the correct result, and never wrong. He can't be wrong.
the fem starts from a pattern, applies a theory, and provides a result. the result is correct. software is not wrong.

Of course, if I don't know the fem theory, I'm going to miss the model, I'm going to make the parameters wrong, I'm going to miss the questions to ask the fem. the fem does the calculations on the model, not on reality. If I'm wrong, I can't blame the fem if his numbers are different from reality.

Unfortunately there are so many people convinced that they can take a degree in engineering studying on wikipedia. Luckily, it's not.
the fem is for the engineer what the calculator is for the accountant. I with a calculator in my hand are not able to do the commercialist, I do not see for what reason fior fior of graduates think that cra@ando the fem module exited from the dixan fustine can dimensional the shell of a nuclear reactor! :angry:

It's true, the calculation is accurate, the problem is in the model. What I meant is that for a beam it is evident that the result is correct because the model is correct, but for the nuclear reactor I would say that it is necessary to realize a simplified model, and therefore it is obvious that the result will be corrected for the model created but not for the reality!

 

[MacGG]

Well, lightning,

that just never wrong, I would not put the amno on fire, there is no perfect software and only a few software, they are actually certified, that is tried and tested that give the right results in tulle the situations contemplated.

other thing are the constitutive equations of the fe method, which are valid under certain assumptions and therefore the results are accurate if such hypotheses are respected.

the reality, which does not always allow to respect the hypotheses and therefore it is necessary to understand well how a fea works, to make the appropriate choices of modeling, the possible simplifications to reality, the just considerations on the results.

It remains the fact that nowadays, for the performance you ask for products, work without the fea, even to make a simple bicycle, it is almost impossible.

Hi.

♪

 

[JUMPY_75]

with my intervention I did not want to offend anyone, neither software manufacturers nor engineers nor graduates etc. etc. I just wanted to ask you users of this site if in your mind the "throw of fear" was insidivaged as at the end of the speech the responsibility is always of man and not of the means he uses. still excuses anyone who feels offended by my statements.

 

[Fulvio Romano]

with my intervention I did not want to offend anyone, neither software manufacturers nor engineers nor graduates etc. etc. I just wanted to ask you users of this site if in your mind the "throw of fear" was insidivaged as at the end of the speech the responsibility is always of man and not of the means he uses. still excuses anyone who feels offended by my statements.

If you have it with me, I'm not offended. I used harsh words, but not to you...:finger:

fear is always there. the reasoning fades away. In principle when I double-click the fem software icon, I should already know more or less the result. The fem gives me the figures after the comma.

recalls the golden rule of the engineer: "measure with the micrometer, mark with the chalk and then cut with the axe"