Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations KootK on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Element & Nodal Contours: Corner Data Confusion

Status
Not open for further replies.

RealSaladsamurai

Mechanical
Apr 11, 2012
72
Hi All - I have been researching this one for a while. I have read the Femap help and read many posts on the subject. I am almost there, but need a little more help.

There are many options under the 'Select Contour Options' menu that seem contradictory to me. My experiments lead me to the same conclusion. Below are 2 screenshots: the Left has the NODAL radio button activated and the Right has the Elemental radio button activated. I have requested that corner data be computed by Nastran. Here are my questions/comments. Any insight is greatly appreciated.

1) Plotting Nodal Contour; Corner Data = OFF; Data Conversion (DC) = Average:
I believe that I am getting stress values at the nodes that are calculated by averaging the centroidal stresses of the elements connected at that node (the 'average' takes into account the size of the element).

2) Plotting Nodal Contour; Corner Data = ON; Data Conversion (DC) = Average:
I believe I am getting stress values that are computed directly at the corners by Nastran. No centroidal data is used and the corner values are averaged together.

3) Plotting Elemental Contour; Corner Data = OFF; Data Conversion (DC) = Average; Element Contour Discontinuities (ECD) = Property:
I get the exact same thing as in (1) above.

4) Plotting Elemental Contour; Corner Data = ON; Data Conversion (DC) = Average; Element Contour Discontinuities (ECD) = Property:
I get the exact same thing as (2) above.

Does this all sound correct? So case 1 is really an elemental contour and case 4 is really a nodal contour. So it would seem that the "Element Contour Discontinuities" should not be available for case 4. But it is and so it gives me the impression that we are not averaging across properties. But since it is really a nodal plot, I think that we are averaging across properties. And so the settings are somewhat misleading. Or at the very least very confusing.

Any thoughts on this? I have been spinning my wheels for a while now. Thanks for reading!


contouroptions2_atwwsr.jpg
contouroptions_lziunr.jpg


________________________
FEMAP v11.1.0
MSC Nastran v2013
 
Replies continue below

Recommended for you

it sounds not unreasonable.

the raw output from NASTRAN is each element having a centroidal and nodal stresses; and the centroidal stress is calc'd from the same data as the corner stresses (ie the integration points of the element). So you can use the centroidal (element) stress for your contours. If you use the nodal stresses, FeMap is asking "what should I do with the multiple values I have at the nodes ?" ... do you want to see the maximum or the average ?

At the end of the day, the plots shouldn't be too different. Unless you're in an area of high shear, when the difference is telling you to refine your mesh.

another day in paradise, or is paradise one day closer ?
 
I am facing the same situation....and i think that when you keep corner data on(which is the default setting) it shows the maximum stress at the junction or discontinuities and when you keep corner data off it shows you the membrane stresses, ignoring high stresses at junctions and discontinuities.
And can any one tell me what does the suboptions of DATA CONVERSION,AVERAGE, MAX VALUE AND MIN VALUE.... are for?
 
rb1957 said:
It sounds not unreasonable.

the raw output from NASTRAN is each element having a centroidal and nodal stresses; and the centroidal stress is calc'd from the same data as the corner stresses (ie the integration points of the element). So you can use the centroidal (element) stress for your contours. If you use the nodal stresses, FeMap is asking "what should I do with the multiple values I have at the nodes ?" ... do you want to see the maximum or the average ?

At the end of the day, the plots shouldn't be too different. Unless you're in an area of high shear, when the difference is telling you to refine your mesh.

Hi rb1957. I understand that the Data Conversion field handler the "what should I do with the multiple values I have at the nodes ?" portion of things. I guess my main point is this: When you have a nodal plot, the Element Contour Discontinuities field is unavailable. But I have shown in my original post that Case 4 is in fact a Nodal Contour even though you have the elemental contour radio button activated. This has the effect of "tricking" Femap into thinking you are using an elemental contour and so the Element Contour Discontinuities is available (but I don't think it should be).

Sarikahirpara said:
And can any one tell me what does the suboptions of DATA CONVERSION,AVERAGE, MAX VALUE AND MIN VALUE.... are for?

Hi Sarikahirpara! If you press F1 to open Femap help and then search "contour" and open the page titled "Selecting Data for a Deformed or Contour Style" and scroll down to "Contour Options" it has very good explanation of this. The short answer is that Nastran calculates stresses by element. A node (corner) is almost always shared by multiple elements. So at each node, there are multiple values of stress that come from the individual elements. The "Data Conversion" options let you choose which values to be displayed: the average, the max, or the min.


________________________
FEMAP v11.1.0
MSC Nastran v2013
 
yes, because you tell FeMap what to do with the four nodal stresses ... take the average, the maximum, ...
so FeMap has one value at each point (like it does if you use element centroid stress).

If you use element stresses you have two choices ...
1) do you want the plot to be a nice smooth stress profile, which will show a varying stress over an element; or
2) do you want to see each element with a constant stress ?

another day in paradise, or is paradise one day closer ?
 
Hi rb1957. Unfortunately, the internet is making this difficult to explain. It seems that I am not effectively communicating what my issue is. I am not talking about the Data Conversion options that offer Max, Min, and Average.

I am talking about Element Contour Discontinuity (ECD) options: Property, Material, Angle.

I believe that ECD is supposed to be available for Element Contours only and not available Nodal Contours.



________________________
FEMAP v11.1.0
MSC Nastran v2013
 
I support your belief ... element options are irrelevant for nodal plots.

how about calling the help desk ? they can be surprisingly helpful.

another day in paradise, or is paradise one day closer ?
 
Hi rb1957. We agree that element options are irrelevant for nodal plots. That is what I am getting at.

If you select Elemental Contour but you turn the Corner Data ON, you really get a Nodal Plot. So it follows that the ECD options should not be available to you ... but they are.

I do not believe that they actually do anything when you use them in this way, but it is a little confusing. It seems like they should just be made unavailable when this combination is selected.

I have never actually called the help desk. I might need to try that!

________________________
FEMAP v11.1.0
MSC Nastran v2013
 
Hey,
Thanks for your comments and have you come to any conclusion about your problem? Pls, whenevr u get the solution of it, share with me.......it will be very helpfull for me too.

Thanks
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor