Volume Measure

[ChaseWichert]

NX 8.5.1.3

I am having a hard time getting accurate volume measurements using measure bodies. I am modeling a metal bellow in it's expanded and contracted state. What I did to make sure what I was measuring was incorrect was to model a cylinder at a fixed volume (the same height as the bellows in the contracted state). I took the volume of the cylinder, calculated it by hand, and measured the volume of the bellow itself, the volume outside of the bellow but inside the cylinder, and the volume inside the bellow. When I added all three of those measurements together it was approx 2.3 cubic inches off. It should have been around 75 cubic inches based on the cylinder, but it is measure our 72.7 cubic inches. How do I increase the tolerance of the measure, or how do I even get an accurate measure.

Thanks,
Chase

 

[JohnRBaker]

How did you isolate the internal and external volumes of the 'bellows' with respect to the cylinder? Could you provide a copy of your model?

John R. Baker, P.E.
Product 'Evangelist'
Product Engineering Software
Siemens PLM Software Inc.
Industry Sector
Cypress, CA
Siemens PLM:
UG/NX Museum:

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

 

[ChaseWichert]

So now it is only showing .2589. I did this last Friday as well, so maybe it was only off that much last Friday I forget. However this is only a small section of a large assembly.

NX 8.5.1.3

 

[JohnRBaker]

OK, the problem is that you're mixing 'Apples & Oranges'.

The solid of rotation for the 'Bellows' was created using a profile which was made of lines and arcs, while the 'inner' and 'outer' bodies were made using a profiles which were created as intersections, which means that they were Splines that were, by definition, approximations. This is why your apparent error was so high. The good news is that the measurement of the 'Bellows' was OK since that model was created using the 'Lines and Arcs'. It was the inner and outer that were wrong due to the use of the splines for the profiles.

Anyway, I went back and created a single sketch using an offset curve to get profiles are all three bodies, the 'Bellows', the 'inner' and the 'outer', so that they were all being created based on not only the same curves but also that they were all lines and arcs which means that the need to do any approximations was minimized (see the attached model). Note that I've created an expression, 'Diff_Calc_vs_Measured', which shows the difference between the measured and calculated volumes (calculated meaning taking the measured volume of the Cylinder and subtracting the measured volumes of the inner and outer). You will note that for my model that difference is 1.74494E-13 cubic inches. Or in standard form, 0.000000000000174494 cubic inches, close enough for government work ;-)

Anyway, take a look at my model and let me know is you agree now that NX is probably giving you a good result.

John R. Baker, P.E.
Product 'Evangelist'
Product Engineering Software
Siemens PLM Software Inc.
Industry Sector
Cypress, CA
Siemens PLM:
UG/NX Museum:

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

 

[ChaseWichert]

Awesome, so an intersection always creates splines, or if I uncheck the join curves option in intersection curves would it recognize that they are lines and arcs? I suppose I could test that myself, but figured you might have a quick answer.

NX 8.5.1.3

 

[ChaseWichert]

That worked, intersection curves without joining them. Thank you for the help. However I am little disappointed it isn't 0.00000000000000000000000000000000000000000 (infinite)

NX 8.5.1.3

 

[JohnRBaker]

Not all intersections result in approximations; if the system can determine an exact curve, be it either a line, an arc or a conic, it will try that first.

And yes, in your example if you had NOT used the 'Join Curve' option, you would have gotten a better result. I still think my approach is better since I don't compute any secondary profiles. However, if you go back and edit those intersection curves, changing them from splines to lines and arcs (by toggling OFF the 'Join Curves' option) your final difference between measured and calculated is only 0.000000000000288604 cubic inches still close enough for government work, and as I said, since the 'Bellows' is actually created from the original lines and arcs, the measurement of it's volume is as good as it was in my model since that's not the part that I changed.

John R. Baker, P.E.
Product 'Evangelist'
Product Engineering Software
Siemens PLM Software Inc.
Industry Sector
Cypress, CA
Siemens PLM:
UG/NX Museum:

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

 

[ChaseWichert]

Yeah, in my final volume, I trying to get the internal volume of of a lot of assembled components, so I am linking all of the bodies into an assembly and taking intersection curves of all of the components in order to get the final internal volume. I couldn't think of another way to do it, that was quick and easy as intersection curves.

NX 8.5.1.3

 

[JohnRBaker]

Intersection curves are fine as long as you're aware of what's actually happening and what it is that's having an impact on what.

John R. Baker, P.E.
Product 'Evangelist'
Product Engineering Software
Siemens PLM Software Inc.
Industry Sector
Cypress, CA
Siemens PLM:
UG/NX Museum:

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

 

[ChaseWichert]

Yeah, I think I have a pretty decent idea. Until I run into an error, then I figure it out.

NX 8.5.1.3