CC not always calcualte real volume

[DA523]

A year ago, I tried to use CC to find volume of excavation/stockpile.
I realized that 2.5 D volume tool has limitation because it uses convex hull not concave hull.
So it can sometime give not precise volume depends on
1- XY plane of excavation/stockpile.
2- Z level of cutting plane ( it is some complex situation)
I find that using CC and Blender is the optimum solution.
CC is very good to prepare XYZ pointscloud and meshing prior of Blender processing.

Please check my attached file that explain this issue.

I use Z=13 as cutting plane (AVG natural ground before excavation )
2.5 D volume is OK
[00:58:23] [Mesh Volume] Mesh 'Mesh': V=5672.66 (cube units)

then I use Z=26 as cutting plane (any xy plane )
2.5D volume is not OK


Conclusions:
1- Use CC+Blender to find real volume (Bender >> knife - extrude - bisec)
2- Do not use 2.5D volume if you have 1 -bad XY convex hull and 2- Z of cutting plane not same Z of surface open face

Waiting your comments

[daniel]

It seems you don't have a lot of points... This tool is clearly not meant to be used with a sparse cloud.

One thing you can do is to actually compute the 2.D Delaunay mesh before running the 2.5D Volume Calculation tool. You have an option to set the maximum edge length that can help you get something closer to the concave hull. Then you can sample (a lot) of points on this mesh. At this point you could even use the scissors tool to refine the cloud extents.

Last, you can compute the 2.5D volume with this dense cloud, and not activate the 'interpolation' option.

[DA523]

I attached here new open mesh and sold mesh to z=13 and z=26 planes
I did sample points of the mesh
I know from Blender that volume of
Mesh.sampled to plane Z=13 is 5672 cu
and volume to plane Z=26 is 13546 cu

How I can get same volume using 2.5D tool?

[daniel]

Warning, your Z axis seems to be inverted.

Moreover the points are not super dense, but that's ok to get close to what you are looking for.

I rotated your cloud so that the vertical (Z) goes upward, and I got the following result:

volume.jpg (150.55 KiB) Viewed 67144 times

i.e. 5679 instead of 5672.

[DA523]

I am aware about Z inversion
Ok , What about volume to plane Z=26 ( which I know =13546 cu)
How to get this volume by CC?

[daniel]

Since the plane Z = 26 is way above (or below ;) the mesh, which volume do you expect?

CloudCompare will consider some vertical walls joining the shape and the plane. This gives roughly 16000 cu:

volume2.jpg (128.54 KiB) Viewed 67095 times

Which is about right since the shape projected surface is about 790 u^2 (787.4 exactly), and if you multiply it by (26 - 13) = 13, you get 10236 cu. And if you had ~5670 cu which is the volume below Z = 13, you get 10240 + 5670 ~ 15910 cu.

Not sure how you obtain 13500 cu?

[DA523]

I am sorry plane level is Z=23 not Z=26 (my mistake)

however precise volume is V =13546 cu
I can get almost the same volume by CC V= 13778 cu
Thank you for help

[daniel]

Ok, it makes sense.

Mind that if you sample more points on your mesh (e.g. 1 million at least), you'll be able to reduce the grid cell size and get a better estimate of the volume.