Mass estimation of FEA structure

[Mike Novikov]

Hello, I would like to bring to your attention a strange mass estimation (bug) in FEA structures of OpenVSP 3.45.2.
It looks like the mass of the shell elements is underestimated by 7-8 orders of magnitude, while the beam elements are overestimated by 10 to 100 times.
I attached vsp file and mass report after "Export FEA Mesh".

[Rob McDonald]

Thanks for sharing this test case.

When I run approximate numbers, I get the shell element calculation looking correct and the beams being off by a factor of 100...

Skin surface area (from CompGeom) (this ignores the cutouts at the root, but should be close)

32.4 m^2

Skin thickness

.008 m

Skin material density

.00158 kg/m^3

Skin mass hand calc

0.000409536

OpenVSP shell mass

0.0003863

Which is 6% high -- easily explained by the cutouts in the skin at the root of the wing.

Please explain why you think this is wrong by 7-8 orders of magnitude given the input properties.



I also ran some checks yesterday for a unit sphere and got essentially exact agreement.  However, that sphere was shell only, no beams...

Wing span (cap top and bottom, so 2x the volume)

13.77 m

Main spar cap dimensions (rectangle)

0.1x0.01 m

Main spar density

2.71e3 kg/m^3
Spar cap hand calc

74.63 kg
OpenVSP beam mass

7532

The OpenVSP beams also include  a vertical portion at the wing tips that I'm ignoring in the hand calcs.

So, it looks like the OpenVSP calc is 100x off.

I'll have to spend some time and try to chase this down....

Rob

[Rob McDonald]

OK, I understand the problem.  This has been broken all along -- thanks for finding it.

OpenVSP isn't calculating the cross sectional area for the various beam cross section types.  At all.  It just uses the default value of 0.1 (model units).

The area is only used for OpenVSP's mass calculation -- the FEM file output by VSP doesn't use the area, it takes the length and width.

Until I fix this, the only beam type that allows input of the area is the General beam type.

So, until this is fixed, you can change all the beam types to General and calculate the cross sectional area (and inertias) outside of OpenVSP.

Rob

[Rob McDonald]

Actually, there is a better temporary workaround.

Change each Beam type to General.

Manually calculate the cross sectional area.

Change the Beam type back to Rectangle (or whatever is desired).

OpenVSP will remember that cross sectional area and will use it for later mass calculations.

Rob

[Rob McDonald]

Correcting the errors in the beam mass calculation on my end, I get this mass calculation for your file...

FeaPart_Name         Mass_Shells   Mass_Beams
Skin                 0.0003863   0          
MS                   2.466e-05   75.32      
AS                   1.497e-05   36.64      
Rib1                 3.262e-06   4.089      
Rib3                 3.081e-06   4.079      
Rib4                 2.81e-06    4.065      
Rib5                 2.81e-06    4.065      
Rib6                 2.586e-06   4.054      
Rib7                 2.295e-06   4.042      
FwdTrim              0           0          
AftTrim              0           0          

FeaStruct_Name       Total_Mass
W                    136.4  

How does this look?

Rob 

[Mike Novikov]

Rob, thank you for your time. To be honest, I just use common sense to estimate the weight of the wing's skin. It can't be less than a couple of kilograms for sure, considering the wing has a 14 m wingspan and a skin thickness of 8 mm.

Now I realized that the density of Carbon Epoxy AS4 3501-6 must be around 1.58 g/cm³ (thanks to pointing this out) = 1580 kg/m3.
So, based on your volume estimation of the skin, mass must be 32.4m2*0.008m*1580kg/m3 = 409.536 kg! It's 6 orders of magnitude compared to OpenVSP estimations.

I'm a newbie in FEA, but I tried to make a very stiff wing to prevent large deformations and material failure!
There is a high probability that I did something wrong with the material, but I used a default one!

"Main spar density 2.71e3 kg/m^3." I used an aluminum alloy, and the density looks right!

Probably, there is a typo with the density of composite materials.
However, you can find in the attached picture that the Input density is fine, but the To FEM density is too small

I agreed that the beam mass is way too big.

вторник, 19 августа 2025 г. в 19:46:51 UTC+4, Rob McDonald:

[Rob McDonald]

Thanks for digging through those values.

You're right, the unit conversion for g/cm^3 is way off -- instead of 3 orders smaller than kg/m^3, it was 3 orders larger -- so 6 orders shifted.

So, this will affect any of the built-in materials that are specified in terms of g/cm^3 -- which is just the composites.  The metals are specified in lb/in^3 and the wood in kg/m^3.

If you specify your own material, just don't use g/cm^3 for now.

Rob

--
You received this message because you are subscribed to the Google Groups "OpenVSP" group.
To unsubscribe from this group and stop receiving emails from it, send an email to openvsp+u...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/openvsp/80f511c6-ea11-44c2-9ea8-e1db751247aen%40googlegroups.com.

[Mike Novikov]

Rob, based on your suggestion, I got the new weight of the wing.

...FEA Mesh...
Mass_Unit: kg
Num_Els: 49854
Num_Tris: 0
Num_Quads: 49854
Num_Beams: 1536

FeaPart_Name         Mass_Shells   Mass_Beams
Skin                 386.9       0          
MS                   24.7        75.32      
AS                   15          36.04      
Rib1                 3.267       4.089      
Rib3                 3.086       4.079      
Rib4                 2.815       4.065      
Rib5                 2.815       4.065      
Rib6                 2.59        4.054      
Rib7                 2.299       4.042      

FwdTrim              0           0          
AftTrim              0           0          

FeaStruct_Name       Total_Mass

W                    579.2      

Assembly_Name       Total_Mass
WA                   579.2       kg

вторник, 19 августа 2025 г. в 12:41:03 UTC+4, Mike Novikov: