Von Mises Vs Max Principle 2

[TrapperJohn]

Given the legal design requiremnt: The Stress level, under load condition, at any point in the structure shall be limited to a level that provides a safety factor of 3 against permanent deformation.

Whould you use Von Mises or Max Principle in your FEA modle?

 

[Cmfg]

If you have a near triaxial state of stresses s1~s2~s3, it is not safe to use the Von Mises stress.

 

[YYS]

Hi, All,

My question may be a little away from the original discussion and I appologize. I also appreciate if you experts can cast some lights on my question.

I am working on Aluminum casting process of automotive suspension parts such as control arms and knuckes. It seems to me that the part design is mainly focusing on stress requirement (von mises)instead of cyclic fatique.

My question is:
If the casting alloy's fatique strength is increased while the other properties (tensile, elongation, etc) remain the same, will it be helpful to reduce the part weight in the part design?


Thanks again.

YY

 

[GregLocock]

Probbly not. Suspension arms and spindles are designed to meet stiffness criteria first and foremost, then you use static strength for things like kerb strike and other maximum load cases, then you migh have to add a little metal in some places for fatigue life.

Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[051174]

In general i use vonmises for 3d modelling and for 2d max/min 2D principal stress...even if performing plastic analysis it is required to read in vonmises stress i prefer the second one also in linear analyses because of is more conservative for 2D fem.In compression what to read?vonmises is +?

 

[corus]

In general I use what ever the design standard tells me to use.

corus

 

[eduardotano]

The statement “against permanent deformation” implies plastic deformation, i.e., yielding. Yielding, in ductile materials, results from shear stresses. Only von Mises criterion (shear deformation energy) and Max Shear criterion (maximum difference between principal stresses) are useful to evaluate the tendency to yielding. Principal stresses are not useful to predict permanent deformation. Consider a “hydrostatic stress state” (S1=S2=S3). No matter the value of the stresses, the material never yields. I think you have to use von Mises (or Max Shear) criterion.

 

[rb1957]

i would have thought that a principal stress exceeding yield was an indication of plasticity

 

[johnhors]

rb1957

In a uniaxial stress state (where P1 = Von Mises and P2 = 0) , yes.

In a fully biaxial stress condition where P1 = -P2, the Von Mises stress value is higher than P1, and is thus a safer value to use. After all Von Mises is an energy based yield criterion. What does puzzle me though is why some people (are they really people?) insist on using a "signed" Von Mises in fatigue analysis!!

 

[rb1957]

i knew i'd get a multiple axis stress response ! in the case of bi-axial tension/compression the vM stress is higher indicating that failure will happen at a lower applied stress. my only comment would be the change "safer" to "more appropriate".

as for signed vM stresses ... actually that makes sense to me, to distinguish between tenisle conditions and compressive conditions for fatigue analysis (but then i'd use principal stress anyways).

 

[jagad5]

"Consider a 'hydrostatic stress state' (S1=S2=S3). No matter the value of the stresses, the material never yields. I think you have to use von Mises (or Max Shear) criterion."

If S1=S2=S3 then the von Mises stress is zero. I don't think you can say that it would not yield is not quite right. If you could produce a stress state where Si are all equal, and tensile, the material would yield. I'm not sure what a real material would do if you applied Si compressive to extremes. You could try this by dropping a bowling ball sized sphere of chewing gum into the Mariannas Trench (35,000 ft deep water) and observe what happens.

A general state of stress can be divded into the hydrostatic and deviatoric stresses. Hydrostatic stresses change the volume of the solid element while deviatoric stresses are changing its shape. Von Mises theory assumes that damage is caused by this deviatoric stress. In the case of a uniaxial tension test, the von Mises stress and the maximum principal stress are equal and both can be used to predict the onset of yield in a ductile metal.


Doug

 

[eduardotano]

It is generally assumed that if a high tensile hydrostatic stress state is applied to a metallic material (steel), it will not yield but it will undergo a brittle failure. This is the reason to avoid “triaxial tensile stress states” in steel structures (joints, notches).
A steel ball submerged in oceanic trench is the same example cited on Kachanov’s “Plasticity” text. According him, it will not yield, it just will be elastically compressed.
Yielding (change of shape) is produced just by the deviatoric component of the stress tensor. Von Mises stress is a measure of its magnitude; it coincides with the second invariant of the deviatoric component of the stress tensor.
Regards.

 

[joekm]

"Principle Stress" is stress along the "priciple axis". By definition, the principle axes is the orientation along which the transverse stress goes to zero. When an isotropic material is so orientated, you will find a maximum value in one direction and a minimum value normal to it.

This can be visualized via Mohr's Circle.

--
Great Spirits have always encountered violent opposition from mediocre minds
-- Albert Einstein

 

[johnhors]

Joekm,

To re-iterate my previous response:-

there is no such thing as "Principle Stress".

In the English language Principle and Principal have very different meanings and cannot be interchanged. Sorry if I appear over bearing on this point, but I believe this to be very important. There has already been more than enough deterioration in the quality of work produced by engineers during my career.

 

[prost]

In addition, the meaning of the phrase "transverse stress" is ambiguous. Transverse usually means 'perpendicular' to a main direction (normally I would say 'principal direction' but since we normally use principal direction to indicate the direction in which the shear stress goes to zero, I will use 'main' instead). For instance, in a rolled material such as aluminum, the 'longitudinal' direction (L in some parlance) is the direction the material has been rolled, and would be considered the main direction. The longitudinal-transverse direction (LT) is perpendicular to the L direction, and the short-transverse (ST) is perpendicular to both L and LT. Perhaps "joekm" you meant "shear stress" and not "transverse stress"?

 

[cbrn]

Hi,
Johnhors, you're right, but English is not the only language in the universe, and I know some extremely good engineers who know it only approximately, so please let's concentrate on the concept and not so much on terminology when all the context is clear... This said, of course this forum is anglo-saxon so everybody should make an effort to "speak" correctly...

The given definition is 100% correct, though perhaps it doesn't add so much to the topic. I feel this thread is now beginning to turn round and round in circles when, historically, different failure theories have been formulated for the simple reason that some are more appropriated for some materials, some for others.
And, from the given definition of "principal stress", IMHO it is obvious that "transverse stress" was written for "shear stress"...

 

[prost]

cbrn, if you think the thread is going round and round in circles, then by all means, you are free to stop reading and adding to it. It appears to carry some interest with others.

My intention was not to criticize anyone's English, I know how I struggle with comprehension of the two other languages I sometimes use as well as English; nevertheless, there are many ways to use 'transverse stress' and just because it's obvious some doesn't mean it's obvious to all. I sometimes find it very difficult to establish up front what particular terms mean, depending on the client; sometimes defining 'obvious' terms is one of the most important steps you can take on a project.

 

[cbrn]

Hi,
prost,
<if you think...stop [adding to it]> : yes, I agree. This is, by all means, the last post of mine in this thread unless it takes a well-defined direction.

<there are many ways... doesn't mean it's obvious to all> : once again, I agree: my previous was not a critic to what you wrote. Saying "transverse" is vague almost every time. Nevertheless, in THIS case, the poster gave a definition which didn't leave any space open to interpretation... Simply my opinion, of course.

Regards