geodesic dome 2

[fanoulos]

Hi all,Im trying to design a geodesic dome with the shape of a hemisphere with 5 rings and a truss for structure support,any ideas???I was thinking of using sketch with splines to design a section of the dome and then to pattern it....any help welcome!!!

 

[kdem]

Here is a site on creating a geodesic dome. If you are designing it similar to this it might be possible pattern but there's five patterns that need to be created if it can be done that way (I haven't looked at the dome in detail so my experice with it is limited).


[url]http://sci-toys.com/scitoys/scitoys/mathematics/dome/dome.ht ml[/url]

 

[michael3130]

Kdem, that is pretty cool, definetely a guy with too much time on his hands but cool none the less

I'm gonna get my kids to give it a go.

 

[fanoulos]

many thanks kdem but I meant to design it with proengineer 5!Any ideas??

 

[jraquet]

A cheap trick might be to create a hemisphere and save the file as an .STL. This will create a model with triangular facets - they might not be equilateral but you can work with the point set. If you decrease the chord height it will increase the resolution and you can pick and choosethe points you want.


View attachment 4928


View attachment 4929


have fun.

 

[jraquet]

You can also try an assembly.


Lock down the first triangle then add each of the first five triangles with a connection and using the sixth to lock it together.


About the same amount of work as building it with straws or toothpicks.

 

[kdem]

Kdem, that is pretty cool, definetely a guy with too much time on his hands but cool none the less

I'm gonna get my kids to give it a go.

Yep, no kids = too much time to do other things.

I didn't actually try building it, don't really have a desire to, but thought it was interesting structuregeometrically.

many thanks kdem but I meant to design it with proengineer 5!Any ideas??

I realize you meant you wanted to model it in ProE I was just posting the site as a reference because it shows some pictures of actual models and some of the intermediate steps which might be helpful in modelingone in ProE. The construction paper image shows what I mean. From the way this dome is colored you can see the patterns that would need to be created if it can be patterned. From a true construction techniqueI don't think you can create say a quarter section and pattern it from what I've read.

 

[michael3130]

It was really interesting and I'd like to actually know how he worked it out mathematically.


I assume the football model that has been posted on this forum was worked out using similar mathematics!!??

 

[jimshaw]

Fanoulos,


I've never done it, but looking at kdem's link gave me an idea. Use those triangular templates at the bottom of his link to sketch a sheetmetal flat, then just bend up the triangles!


May not be the quickest, but it'd be a fun excercise!


Jim Shaw
http://www.linkedin.com/in/shawengineering

 

[kdem]

Here is alink to a calculator that will calculate strut lengths for 3dome types. Check the illustrations section. It givesmore information on what the designations mean.


http://www.byexample.com/library/calculators/geodesic_dome_c alculator/


Here is another one with pages of information. If you look under Dome you will findGeodesic Dome Notes.


http://www.simplydifferently.org/
Edited by: kdem

 

[jraquet]

I gave it a try with the dims from the sci-toys page.


View attachment 4930


It looks like it works but am not sure how useful it would be. this is a triagular surfaces to hex/pent sub-assys to geo-assy.


I also tried a blend pent and plend hex that seem to work: this way you can define the "depth" of the dome for each sub-section. Fun if you have time to waste...but I had been looking for a decent "soccer ball" model for an AM machine to build.

 

[michael3130]

I have a soccer ball model which I downloaded off MCAD if u want it.

 

[jraquet]

Thanks Michael, I would appreciate that.

 

[michael3130]

2011-02-10_105517_soccerball_grey.zip

 

[jdurston]

A cheap trick might be to create a hemisphere and save the file as an .STL. This will create a model with triangular facets - they might not be equilateral but you can work with the point set. If you decrease the chord height it will increase the resolution and you can pick and choosethe points you want.


View attachment 4979


View attachment 4980


have fun.

This is nice, but it doesn't represent the geometry of a traditional geodesic dome. Most standard geodesic domes are based on the icosahedron (20 sided shape composed of all equilateral triangles). Then triangles of the icosahedron are subdivided into triangular grids, but the vertices touch the edges of the sphere. The idea is that the triangles are about as equal-length as possible and still give curvature of the shape. (geometrically, an equilateral triangle grid is by definition on a flat plane; so, the struts can be exactly equal.) Geodesic calculators will give you the lengths of the struts (calculated using spherical trigonometry). The gridding of the triangle breaking the edge down into equal parts will determine the dome's "frequency".

I think the strategy would be to try and get a pattern of points representing the vertices of a icosahedron.

Not easy to do, but if you guys figure it out, please post a reply.

 

[jdurston]

I gave it a try with the dims from the sci-toys page.


View attachment 4981


It looks like it works but am not sure how useful it would be. this is a triagular surfaces to hex/pent sub-assys to geo-assy.


I also tried a blend pent and plend hex that seem to work: this way you can define the "depth" of the dome for each sub-section. Fun if you have time to waste...but I had been looking for a decent "soccer ball" model for an AM machine to build.

This looks better. Nice job.

 

[rmatija]

Hallo


Here is exsample


regards


2011-03-17_024944_geodesic_dome_suppressed.prt.zip