Determine Cross Section Area

[bfleck]

I have geometry that can be simplified as being swept from one section to another with a guide string and now I would like to output the cross sectional area at regular intervals on planes normal to the guide string.
I have tried to use the Section Inertia Feature, but it only outputs a number with the hollow tyoe selected. This 'area' number does not match the actual area of the geometry.
Any suggestions
NX5.0.6.3

 

[JohnRBaker]

I tested this with NX 5.0.6.3 using the Solid 'Section Type' and I got a complete set of data for each section with what appears to be the correct results.

Could you provide a copy of your file, at least something similar so that we can verify that it's working as it should with your files?

John R. Baker, P.E.
Product 'Evangelist'
Product Design Solutions
Siemens PLM Software Inc.
Cypress, CA

http://www.siemens.com/plm

http://www.plmworld.org/museum/

To an Engineer, the glass is twice as big as it needs to be.

 

[bfleck]

John,
Attached is a file that we can discuss.
Dont take too much notice of creation as I wanted to replicate my real geometry which is much more complex.
For analysis, select outer surfaces and curve select central bridge curve. For distance specify 0.254in. It appears that the software uses the hollow type if any cross section becomes open.

 

[morans]

bfleck, when we need to do any kind x-sectional areas we usually just cut the solid with a plane (or extract the curves and extrude them), and then go to "Analysis > Measure Faces..."

hope that helps, I've never tried the method that you described so this may be a little too simplistic for what you're looking to do.

good luck either way

 

[bfleck]

Morans,
You can create a boundary plane with the sectional curves, however, this is cumbersome when you need to check cross sectional area at a lot of locations. Section Inertia is like an automated way of doing it.

 

[cowski]

I tried it with the options you listed and it gave me the error: "Not all sections could be analyzed; due to open sections, hollow analysis is used."

Is your goal to keep the cross sectional area constant? If so, have you tried a sweep feature with the 'area law' option? It may not work for you depending on the complexity of your actual geometry. You have to give up a little control so the software can adjust your sections to the inputs, but the few times I have used it - it has worked pretty well.

 

[mmauldin]

If you are using "Area Using Curves" make sure the X-Y plane of the WCS is positioned parallel to the plane of the curve set for each section you wish to measure.

 

[bfleck]

I am not trying to keep the cross sectional area constant. My actual geometry is much more complicated than the demo and is created with a lot of sections/guides and thru curve mesh. At the end, I want to check the cross sectional area.

Cowski,
that is the error I get and i think it is because the last section is open due to the fact that the end face is not perpendicular to the guide string. Selecting other intervals can work ok.

 

[JohnRBaker]

To start with, why did you create this sample part as individual sheet bodies and then sew them all together at the end? You had all that you needed to create the complete solid in one sweep operation, as I've done in the attached sample which is your model only I created a single solid result, Swept(13), using your original construction curves.

As for using the 'Section Inertia' function, if you're going to use the Solid 'Section Type' it appears that in order to get the accurate results, you will need to select ALL of the faces of the model, including the 'end caps'. I'm not sure this is the correct behavior, but that's how it appears to be working (I intend to take this up with the group responsible for this function).

Anyway, give it try and let me know if you think the results are closer to what you had expected.

John R. Baker, P.E.
Product 'Evangelist'
Product Design Solutions
Siemens PLM Software Inc.
Cypress, CA

http://www.siemens.com/plm

http://www.plmworld.org/museum/

To an Engineer, the glass is twice as big as it needs to be.

 

[bfleck]

I know doing a swept feature was much easier, however, I wanted to replicate the construction of my actual geometry which is much more elaborate with changing shape and area.
Selecting the end caps would work so that I can get the feature to work correctly, then omit the bad results when doing an analysis in excel.
Out of curiousity, how does the hollow type work. What does hollow refer to.

 

[JohnRBaker]

This function was originally developed as part of a set of tools designed for use by people doing automotive 'Body in White' work. This refers to those parts of a car's body which is made from formed and/or stamped sheet metal. Traditionally this work is done using sheet bodies with an assumed thickness, so if you model say a fender as a sheet body and you needed to get a series of sectional data, perhaps for some crash studies, you can use this function even if all you have is one-side-of-metal since when using the 'Hollow' 'Section Type', you can assign it a thickness and not have to actually model the fender as a 3D solid. Of course if you have actually created a solid model, then you can set the 'Section Type' to 'Solid' and as long as the sections are closed it will then assume that we are analyzing a 'solid', even if it were still nothing but sheet bodies.

Anyway, what I learned is that the software, when it encounters an 'open' section, it will automatically assume that you are analyzing a 'Hollow' part since if the section is 'open', there is no practical way of closing it so that 'Solid' results could be found. I would prefer it if it would just flag any of the 'open' section results as such when the 'Section Type' is set to 'Solid' yet still perform the solid analysis on those sections which are actually 'closed'. I intend on making that recommendation, along with a few other issues I think should be addressed as enhancements in a future release.

John R. Baker, P.E.
Product 'Evangelist'
Product Design Solutions
Siemens PLM Software Inc.
Cypress, CA

http://www.siemens.com/plm

http://www.plmworld.org/museum/

To an Engineer, the glass is twice as big as it needs to be.

 

[bfleck]

John,
I agree. That would make sense just to continue with the area calculations and flag open sections.
Thanks