Designing a geodome from scratch

[Spindoctor]

Hello!

After a several months break I am about to jump in my next FreeCAD adventure.

I would like to try to design a Kruschke 4/9 geodome from scratch.

I am aware of the geodome generator macro which might be helpful. However, I want to improve my skills, therefore I want to try it on my own.

In my previous works in FreeCAD, the Sketcher tool was very handy. Using this, I managed to design the geodome's base as well as an "unfold" version of the geodome.

However, I don't know how to take this part to the third dimension. It seems like if I would follow the sketcher/part design path, I would need to know/calculate the angle of each pane. This doesn't seem very useful to me - I wonder if there is another way to design the dome without knowing the exact angles. Instead, the angles of the panes should be a result of combining the triangles which I can define completely.

Can you help me find a best practice to construct this object in FreeCAD?

Thank you!

[Spindoctor]

... as I still try to figure it out, I *bump* this topic.

Hope this is not regarded impolite...

[bejant]

I don't consider it impolite but I didn't reply before because it was already discussed a few times. Maybe you did a forum search using the wrong term? I used "geodesic":
https://forum.freecadweb.org/search.php ... s=geodesic
Does that help?

[Spindoctor]

Thank you for your reply!

I have to admit, I searched the forum only after my first post (but before my bump). I think I read all relevant threads. Most of them suggested to use the macro, which seems to be awesome but wouldn't help me understand FreeCAD better.

The most helpful thread was this one:
https://forum.freecadweb.org/viewtopic. ... sic#p74658
where whirlybomper recommends to use a bottom up approach: design the triangles first and then assemble them to build the dome.

I tried that and designed the triangles in the part designer. That was easy. However, I didn't succeed to assemble them the right way. I didn't find a way to snap the edges together and aligning them by hand seemed an impossible task...

That's where I am stuck. However, maybe there is a better approach anyway?

[bejant]

Thank you for your reply!

You're welcome!

tried that and designed the triangles in the part designer. That was easy. However, I didn't succeed to assemble them the right way. I didn't find a way to snap the edges together and aligning them by hand seemed an impossible task...

Assemble them after they are solids?
Assemble triangles or each leg of the triangles?

You could try the Part WB > Edit > Alignment tool:
https://www.freecadweb.org/wiki/index.p ... _Alignment

[ulrich1a]

n my previous works in FreeCAD, the Sketcher tool was very handy. Using this, I managed to design the geodome's base as well as an "unfold" version of the geodome.

However, I don't know how to take this part to the third dimension.

There exists a "simple" recipe to create such highly symmetric objects in FreeCAD. It is based on a projection of the object along one of its faces with the main symmetry axis staying vertically. If you could manage to create a sketch of the projection, the next steps are straight forward. See these threads:

https://forum.freecadweb.org/viewtopic. ... 157#p57885
https://forum.freecadweb.org/viewtopic. ... =20#p57611

Ulrich

[Spindoctor]

@ ulrich1a
Thank you for your input. I will try to get my head around this in the next couple of days. I am however not sure I will succeed

@ bejant
This was a very helpful hint! To be honest, I tried the alignment tool before, however I didn't use solids but just faces (created with the part workbench). This didn't succeed.

This time I used solids instead and got quite a big step further (see file "geodome align.fcstd").

The next step would be to add part "Pad CCD" just in between the other two triangles (the clones of "PadCCB"). I thought, the Assembly 2 workbench might be helpful. However, adding the last constraint (an axial constraint between "Pad CCD" and "Clone of PadCCB001" takes really a long time to calculate (see file geodome constraings.fcstd").

Do you think I am on the right path? If yes, is there a way to "help" the Assembly 2 workbench to come to a solution (best would be the right solution) in a reasonable amount of time?

[NormandC]

Hello,

I thought, the Assembly 2 workbench might be helpful.
(...)
Do you think I am on the right path?

No, you're not. You're not using Assembly2 as it was intended. The specific workflow is to create a new document (let's call it the assembly document), then import part into it from separate, independent FreeCAD documents.

Having everything in the same document was never intended, who knows what can happen.

The next step would be to add part "Pad CCD" just in between the other two triangles (the clones of "PadCCB"). I thought, the Assembly 2 workbench might be helpful. However, adding the last constraint (an axial constraint between "Pad CCD" and "Clone of PadCCB001" takes really a long time to calculate (see file geodome constraings.fcstd").

You cannot move a Pad except by moving its underlying sketch. I'm pretty sure that's what causing the issue with Assembly2.

As I said, Assembly2 requires that parts be imported from external files. What it does then is to create a dumb copy of the imported part so there is no restriction on its movement.

The next step would be to add part "Pad CCD" just in between the other two triangles (the clones of "PadCCB").

This can easily be done by moving then rotating the underlying sketch using Draft Move then Draft Rotate. But you'll soon find that the triangle doesn't fit in the space. It overlaps one of the other triangles.

[Spindoctor]

Thanks for your reply!

@ NormandC:
Thank you for pointing this out. So I will give it one more try with Assembly 2 in the way it was intended.

The form I am trying to achieve looks like this, so in the right angle it should be possible to fit the triangle inside.

I thought Assembly 2 would help to figure out the right angle...

@ ulrich1a:
I thought about your approach a bit more. It would be easy to make a projection on the "ground" of the geodome. However, if I get you right I would need to make a projection on one of its faces. This seems next to impossible to me - at least without knowing the exact equations for building the geodome (which I have to admit I don't know and even hoped, that I don't need to know them exactly).

[ulrich1a]

I thought about your approach a bit more. It would be easy to make a projection on the "ground" of the geodome. However, if I get you right I would need to make a projection on one of its faces.

You need an orthogonal projection, just the same as on a drawing. You wrote you have already all triangles drawn. Then you have all information.
The projection is just a polygon. The polygon lines are either a side of a triangle or the heights line of a triangle. The points where the sides of triangles meet, are lying on a half-circle. The points where heights of triangle meets have a certain distance to the half circle. You can easily get this distance, if you draw the base line of the height as a chord of the half-circle in the same sketch.

But I should have done it myself first, before talking about it. The before suggested method of joining the extrudes profiles fails in case of a geodesic dome. The only thing you can get by just rotating the profile, are some points in space to start the aligning of the triangles.

at least without knowing the exact equations for building the geodome

Here you go: https://groups.google.com/forum/#!topic ... G4Mm__cVHI

Ulrich