Dear all,
I am pretty new to grasshopper so I can't figure this out on my own.
I need to create a diagrid structure for curved surfaces similar to Norman Foster's british museum canopy or that office building in London.
I have tried diagrid GH script shown on Grasshopper Primer but I wasn't able to created curved diagrid members. I also want rectangular members instead of tubes.
So, Does anyone know how to create, curved smooth diagrid members for curved forms???
you'll also find more examples w/ tubes because they can just use the pipe component on the basic wireframe. Rectangular sections will be more involved. good luck!
i spent a while creating a grasshopper definition for diagrid on any surface last semester. One of the biggest challenge is to get the edge condition right...
With grasshopper, I don't think you can create members of other section shapes other than tubular.
thank you very much, stargazer.
Before posting this thread, I googled and did find some GH_scripts though they did not give me what i wanted. (otherwise, i wouldn't post this thread to waste your time.) In most cases, I got straight members with messy edges. I want these members to reflect curvature of the surface, similar to beatifully curved hull members you see inside ships. I want to achieve curved members without increasing the density of surface division.
So far, component on surface script worked best for me though I need to work out edge condition.
Btw, you mentioned you wanted curved members, which means you're glass would have a double curvature on it. £$£. The Gherkin is actually straight members, faceted to look curved from a distance. Paneling tools has some analysis functions to determine the deviation of components from you original surface.
Sorry I can't be of as much help as I'd like to with this, but figured I'd put my understanding of this structure down for what it's worth. I had no involvement in this project, so take into consideration my knowledge isn't first hand but relayed from word of mouth.
The Great Court Roof is an amazing piece of work by Fosters and Buro Happolds. I've actually been researching into the form of this roof as a potential follow up tutorial for my blog, http://geometrygym.blogspot.com
I thought the roof surface was form found, but it seems that there is a mathematical equation that defines it. But your question really relates to the grid arranged over that surface, and the work to derive this stunning grid is more sophisticated than most would undertake.
The grid isn't a mathematical or analytical solution, rather a segmented arbitrary grid topology was placed on the surface as a starting point. Then a "relaxation" was undertaken to achieve the fine curves and spiral paths that derive the grid. This nodes were free to glide around the surface until (I don't know the exact objective) the lengths were similar or angles of opposite elements were equal or something similar.
So I think it might be difficult to achieve a similar looking grid in Grasshopper without some programming. I'll look into it further and post it to my blog if I figure it out.
You might find some of the other blog tutorials worth studying, there is one on emulating the Swiss Re "Gherkin" using Grasshopper, and another on the Beiging Watercube. This includes parametric sweeping of hollow section (or similar) profiles on curves in Rhino.
Cheers,
Jon
A web link that might better explain the Great Court Roof strategy:
Thank you everyone
I did the same thing 18x32 suggested.
I know, I am fine with straight members. Having curved members wasn't my idea. it was my picky boss's idea who wants large curved structural members wrapping all around his design, rather than many small members emulating curvature.
I am very grateful to everyone for help.
You might be still interested, I've released the first mesh relaxation command into the Rhino plug-in that enables determining smooth straight curve transitions of a diagrid like the British Museum. Details (including the file for you to try for yourself) at http://geometrygym.blogspot.com
I hope to add even more power to this relaxation capability by adding dynamic relaxation to it some time soon.
Nov 11, 09 1:25 pm ·
·
Block this user
Are you sure you want to block this user and hide all related comments throughout the site?
Archinect
This is your first comment on Archinect. Your comment will be visible once approved.
Grasshopper-Help-Diagrid on curved surface
Dear all,
I am pretty new to grasshopper so I can't figure this out on my own.
I need to create a diagrid structure for curved surfaces similar to Norman Foster's british museum canopy or that office building in London.
I have tried diagrid GH script shown on Grasshopper Primer but I wasn't able to created curved diagrid members. I also want rectangular members instead of tubes.
So, Does anyone know how to create, curved smooth diagrid members for curved forms???
Thank you very much in advance.
I would just google it. There's probably a definition on a blog somewhere you could just download--the diagrid is pretty popular these days :)
I would just google it. There's probably a definition on a blog somewhere you could just download--the diagrid is pretty popular these days :)
triple post... nice
you'll also find more examples w/ tubes because they can just use the pipe component on the basic wireframe. Rectangular sections will be more involved. good luck!
right LML, i'm sure there is already a plug-in for rhino for surface conversion to diagrid.
yeah, I bet there's something in the Paneling Tools plugin as well that would accomplish the same thing
i spent a while creating a grasshopper definition for diagrid on any surface last semester. One of the biggest challenge is to get the edge condition right...
With grasshopper, I don't think you can create members of other section shapes other than tubular.
thank you very much, stargazer.
Before posting this thread, I googled and did find some GH_scripts though they did not give me what i wanted. (otherwise, i wouldn't post this thread to waste your time.) In most cases, I got straight members with messy edges. I want these members to reflect curvature of the surface, similar to beatifully curved hull members you see inside ships. I want to achieve curved members without increasing the density of surface division.
So far, component on surface script worked best for me though I need to work out edge condition.
did you ask mr. Grasshopper himself at the forum? That's where the pro's are...
stasis, right, it really requires a clear mind to sort out the shifts in the algorithm and their effect on the edge condition of the 4 sides.
it's kind of fun to solve the algorithm, if you have problem of doing it, i'm happy to help
The Paneling tools plugin will help.
http://en.wiki.mcneel.com/default.aspx/McNeel/PanelingTools.html
But failing that, Grasshopper should be it. http://designreform.net/2009/07/rhino-grasshopper-parametric-truss/
Btw, you mentioned you wanted curved members, which means you're glass would have a double curvature on it. £$£. The Gherkin is actually straight members, faceted to look curved from a distance. Paneling tools has some analysis functions to determine the deviation of components from you original surface.
Oh sorry, you mentioned the BM, not the Gherkin. Anyways, same story.
Hi,
Sorry I can't be of as much help as I'd like to with this, but figured I'd put my understanding of this structure down for what it's worth. I had no involvement in this project, so take into consideration my knowledge isn't first hand but relayed from word of mouth.
The Great Court Roof is an amazing piece of work by Fosters and Buro Happolds. I've actually been researching into the form of this roof as a potential follow up tutorial for my blog, http://geometrygym.blogspot.com
I thought the roof surface was form found, but it seems that there is a mathematical equation that defines it. But your question really relates to the grid arranged over that surface, and the work to derive this stunning grid is more sophisticated than most would undertake.
The grid isn't a mathematical or analytical solution, rather a segmented arbitrary grid topology was placed on the surface as a starting point. Then a "relaxation" was undertaken to achieve the fine curves and spiral paths that derive the grid. This nodes were free to glide around the surface until (I don't know the exact objective) the lengths were similar or angles of opposite elements were equal or something similar.
So I think it might be difficult to achieve a similar looking grid in Grasshopper without some programming. I'll look into it further and post it to my blog if I figure it out.
You might find some of the other blog tutorials worth studying, there is one on emulating the Swiss Re "Gherkin" using Grasshopper, and another on the Beiging Watercube. This includes parametric sweeping of hollow section (or similar) profiles on curves in Rhino.
Cheers,
Jon
A web link that might better explain the Great Court Roof strategy:
http://www.burohappold.com/BH/NWS_1999Engineeringtheglassandsteelrooftot.aspx
Thank you everyone
I did the same thing 18x32 suggested.
I know, I am fine with straight members. Having curved members wasn't my idea. it was my picky boss's idea who wants large curved structural members wrapping all around his design, rather than many small members emulating curvature.
I am very grateful to everyone for help.
Sounds like you got it sorted, glad to hear it.
You might be still interested, I've released the first mesh relaxation command into the Rhino plug-in that enables determining smooth straight curve transitions of a diagrid like the British Museum. Details (including the file for you to try for yourself) at http://geometrygym.blogspot.com
I hope to add even more power to this relaxation capability by adding dynamic relaxation to it some time soon.
Block this user
Are you sure you want to block this user and hide all related comments throughout the site?
Archinect
This is your first comment on Archinect. Your comment will be visible once approved.