I am working with a small group on an interactive structure made entirely out of wood. We are still in the design phase.
The structure itself is a wooden dodecahedron made up of regular pentagonal panels (each with side lengths of 5'). Each pentagon will join with adjacent pentagons via interlocking wooden connectors. The idea is to make this structure entirely out of wood, no metal.
The initial draft is below
People will be standing inside of the structure, but in case someone decides to climb on top, etc, we want to make sure each panel is as strong as possible.
Since the CNC router can only handle a certain size, we have to break each panel down into smaller elements. Obviously this will make the overall panel weaker. But what is the best way to break this up? My sketch is below (sorry for my poor drawing)
For drawing #1, A shows how we initially intended to divide the panels. But we think it would be better to have the cut go in the same direction as the polyhedra (B)
For drawing #2, assuming we use cut B for all panels, to join individual panels with an adhesive is it better to use a finger joint? How much weaker will the panels be if we joined them flush?
We still have issues to solve with the actual connections and the placement themselves but I figured I'd ask this question first and go from there. Thanks for any advice you are able to offer.
I did not make the drawing myself, a friend did it in sketchup. This is not being presented anywhere, it is just for our own reference.
Menona, thanks for the hinge idea - I will run this by the group.
May I get an opinion on the placement of the connector pegs? Currently, we have them located at this position:
But I am wondering if it would be more stable to do this instead? With more located closer to the points of the polyhedron. Does this make it more stable? If so, is there a formula we can use to figure out the exact spacing of the connectors that would be ideal?
why use a CNC router? Just bust out a set of regular hand held saws and go to town on full sheets. Really, there's no point to CNC unless the shapes are wacky enough to support it.
that' my two cents from the backside of the theater.
I got over big machines a long time ago. It's a ton more interesting to build big things with simple cheap tools.
Start asking yourself these questions:
1) how can I make 2 d shapes from flat material that fold, slot into one another, interlock or fasten to each other simply that...
2)...build a 3d structure that is much more interesting and strong than the shape of the 2d pieces alone would suggest?
and
3) ...that the shape of the 2d pieces is ALSO interesting
and
4) ...that the final structure is way more visually interesting AND functional than something that's built in a standard way?
I mean, all you've done is created a really fancy potting shed with a completely leaky roof and no walls, that's so structurally unstable that you're worried if someone climbs on it. Any one can go down to home depot and buy a shed for $800 that runs circles around yours in terms of functionality.
You can cut the panels on a table saw, and use biscuits to join them. Simple, easy.
It is the connection between units that will get you into trouble, and you haven't even started thinking about them. There isn't a lot of material there to make the joint with, if you're adamant about using only wood.
A well glued biscuit joint should suffice for joining the individual units. We're talking about a single thickness of 3/4" plywood, probably, that needs to support the weight of 2 or 3 people? That bolted connection above is clunky and bad. It will only encourage the students to use the CNC, which will give them a stepped edge, rather than a clean saw blade edge. Less surface area meeting means weaker glue joint.
This thing is riddled with obvious structural and fabrication problems. The thought process shows little sensitivity to the nature of the materials or understanding of the fabrication techniques necessary to make it.
Trying to produce this via CNC (or anything else for that matter) without first working out all the details in prototypes will result in a big disappointment. Fabrication is a critical part of the design process.
The best way for you to figure this out is to stop talking and start mocking up full scale joints to test, where you will quickly find out what doesn't work. Make sure you model at least a full corner as the angles will make assembly of various joinery systems problematic, to say the least.
What I see here (based on my studios as a student, AND my years teaching studio)
1) I see a professor pushing CNC technology on the students. This is dangerous, but it seems to be the way academia is doing it these days.
2) I see students trying hard to please the professor without understanding the point of it. Heck, I can't really see the point of it either AND I've used CNC fabrication as part of my business before.
3) I see that design is hard and the lightbulb hasn't turned on for the students yet. It probably hasn't for the professor either, but the professor needs to take the extra step and explain a couple of things to the students:
a) Architecture must be - at its basis - be dry, functional, safe and efficient
b) Beautiful architecture must operate on at least 2 planes of creativity in order to go beyond the banal.
c) Innovative construction technique MUST do both of the above AND create things that cannot be created by other means. I personally think Frank Gehry's work (for example) fails on point a, which makes points b and c moot.
To further criticize the design - it's structurally unstable because you've built in MANY failure points and you have weak connection locations. Actually, the strong point of any geodesic dome is in the solid connection between the connection points - the connection points are located on the intersections of the points of the form, NOT along the ribs of the form. You've inverted the connections. Go back and take a close look at Buckminster Fuller. You might also look at Konrad Wachsmann.
I think the design skill to do this type of design successfully is difficult to learn - it takes a good understanding of structure, of materials, of fabrication techniques and some trial and error. I was experimenting with some of this years back - some "cut this fancy shape on a CNC" then "tab A inserts into slot A and is fastened" and "tab B inserts into slot B and is fastened" along with "bend along this dotted line" type of stuff. It can create some really neat stuff that's easy to build and looks great and functions great. Scares the crap out of most contractors and fabricators though, because they don't understand it, and it's therefore very expensive and unrealistic in most real world architecture. Some of your professors probably do it, because they have access to free machines that cost the rest of us hundreds of dollars an hour. Heck, I'd do it if I had those machines too LOL.
can you biscuit joint something at an odd angle like that? the cut for the biscuit would have to go in at the right angle. seems like that would be difficult, but i suppose someone who knows what they are doing could figure it out. i also don't see how you would clamp it together for the glue to dry, if that's something you were going to do.
it seems to me, as someone who doesn't know what they're doing, something like a dovetail might be more forgiving. it wouldn't look as clean though.
also, i agree with all of the general 'that's a bad idea' comments.
The connectors shown are actually lap joints.I know they arent very descriptive in the image but I can upload a better picture of the detail a little later. This is just for someone to sit it, it will be chained to the ground and remain stationary. It is just as an exercise in making something completely out of wood.
have you thought about rolling it down a hill while someone is sitting in it? would be way more cool. maybe you don't want to tell them it will be rolling until they're already seated.
if there was some fire, too, that would be awesome. much better than rehashing the youtube one that's already been done.
If I was teaching and a student went online posting "how to make" questions I'd fail them on the spot. Figure it the fuck out, that's what school is for. And life.
Guys, this isn't for any class. This is a personal project a few friends are working on.
I appreciate the input but the design isn't changing, we will just have to work with what we have.
Maybe I should rephrase my question and make it a bit more direct. Forgetting all about the actual design of the structure itself.
Say you have a dodecahedron (no context, no reason, this is in a vacuum ok, lol). Each plane of the dode is its own unit. If you needed to keep them in place with a connector, where would be the best place to affix the ties along the edges and why?
put your connections along the entire edge, each edge, spaced 6"-8" apart.
that of course makes no sense because i don't know what kind of connection you're using. if you just glue it all together, then it's a continuous connection right? that might be a good option.
there's a you-tube video on this thread where someone already did this. why not just do what they did?
as a side note, below is also an entirely wood project, where someone can sit in it. you could easily chain it to the ground if you so desired. seems to meet the performance requirement, if not the dodecahedron requirement:
Sheesh you architects can't answer simple question without debating the purpose of the structure she is building, or the merits of digital fabrication tools in modern curriculums, or what have you.
Kozumelle: that's what makes us architects not just builders or artists. We think about things holistically because we can't help it. To not consider purpose, structure, and aesthetics all at once is impossible. That's how our minds work.
I looked up an FEA image of a dodecahedron, I got a few search results like this:
Which leads me to think that there is additional force at the points/nodes of a polyhedra, Which means it needs extra support there somehow? Or am I way off base here?
There you go, Paula G, all done! Just send that to you 3D printer or whatever and go have lunch.
We think about things holistically because we can't help it. To not consider purpose, structure, and aesthetics all at once is impossible. That's how our minds work.
It's not that we can't help it:it's what we are supposed to do. That's what all the training is supposed to be for. And that's what makes threads like this (and architects, architectural school and students, as well as people in general like this) so absolutely infuriating.
Since there;s nothing riding on this other than failure and shattering your sense of knowing, just experiment. You'll learn much more through trial-and-error than any one of these programs can ever teach you at this point.
I am asking because we don't have much time left to produce it. I thought this was a place to get help and suggestions? How helpful is it to basically tell someone to F off and go google it themselves? Coming here for suggestions was a last resort.
Let's get this clear: because you procrastinated and had no other recourse, you came here looking for immediate help. Hoping for a simple fix-all solution to conceptual problems in your project that hasn't been addressed.
Paula G, this thread should serve as a learning experience for knowing the right time to ask questions and knowing the right questions to ask. IMHO, you missed on both of those accounts.
You are asking the questions at the wrong time. You should have started asking questions a lot earlier in the design process when you weren't so invested in making one design work the way you think it should.
You are asking the wrong questions. It isn't as simple as what connection or joint will make your design work. You need to figure out the right questions to ask and I think a lot of the posters here (regardless of their level of rudeness) are trying to point you to the right types of questions.
My honest opinion is that you should make the structure and the panels distinct and not try to make one thing work for both. Create a dodecahedron frame using dimensional lumber, or steel, or whatever. Attach panels to this frame to hide the structure. The level of detail and the types of connections and all the other stuff after that is up to you and how far you want to get into the design and fabrication and the refinement of that.
If this isn't for a class and is just a bunch of friends working on it ... what's the rush? Why not take the time you need?
How helpful is it to basically tell someone to F off and go google it themselves?
Aside from the fuck off part (which I don't recall anyone saying), isn't that exactly what you were doing? LOL
Coming here for suggestions was a last resort.
Obviously. The sense of entitlement that some people have is astounding. When the center of the universe is finally discovered, more than a few people are going to be shocked to learn that they are not it.
I believe I nailed this one with my first post (#3).
"I was not expecting anyone to do anything for me, merely suggestions or maybe a straight forward answer."
There's a link in here with the exact life-sized thing you wanted to build. Already made. Might as well just copy it and understand why they did it that way.
This is obviously an installation for Burning Man. Pack in, pack out, so this needs to be put together on the Playa, and using a CNC for multiple parts will make it easier to put together in the desert, because all the parts are identical. And Labor Day isn't too far away, especially if you are still in design. Mockups and experimentation is the only way to go. Good luck.
The problem is there *isn't* a straightforward answer to a question like this, because there are far too many variables that none of us can glean from how you've presented the question. As Everyday Intern said, you're not asking the right questions.
No, the problem here is that she didn't get the answer she wanted.
At least she appears to have gone away - some people just obsessively persist in posting the same thing, hoping that eventually they will get the answer they want.
Advice needed on wooden dodecahedron structure we are designing
I am working with a small group on an interactive structure made entirely out of wood. We are still in the design phase.
The structure itself is a wooden dodecahedron made up of regular pentagonal panels (each with side lengths of 5'). Each pentagon will join with adjacent pentagons via interlocking wooden connectors. The idea is to make this structure entirely out of wood, no metal.
The initial draft is below
People will be standing inside of the structure, but in case someone decides to climb on top, etc, we want to make sure each panel is as strong as possible.
Since the CNC router can only handle a certain size, we have to break each panel down into smaller elements. Obviously this will make the overall panel weaker. But what is the best way to break this up? My sketch is below (sorry for my poor drawing)
For drawing #1, A shows how we initially intended to divide the panels. But we think it would be better to have the cut go in the same direction as the polyhedra (B)
For drawing #2, assuming we use cut B for all panels, to join individual panels with an adhesive is it better to use a finger joint? How much weaker will the panels be if we joined them flush?
We still have issues to solve with the actual connections and the placement themselves but I figured I'd ask this question first and go from there. Thanks for any advice you are able to offer.
This should be good.
Don't worry about finger joints - just add a biscuit or dowels to a flush joint - will be just as strong as a finger joint.*
Option B is definitely better/ easier to fabricate and will be stronger.
*Archanonymous is not a structural engineer. The extent of my knowledge involving joints is rolling em and smoking em.
What/Why is it?
What about creating some sort of long piano hinge out of wood for each connection?
A long oak dowel run through several "flangie-type-pieces"
It would look something like this...-ish. (search "wood hinge" on the interwebs for pics)
You could output all of the hingie-pieces via CNC as well.
I did not make the drawing myself, a friend did it in sketchup. This is not being presented anywhere, it is just for our own reference.
Menona, thanks for the hinge idea - I will run this by the group.
May I get an opinion on the placement of the connector pegs? Currently, we have them located at this position:
But I am wondering if it would be more stable to do this instead? With more located closer to the points of the polyhedron. Does this make it more stable? If so, is there a formula we can use to figure out the exact spacing of the connectors that would be ideal?
why use a CNC router? Just bust out a set of regular hand held saws and go to town on full sheets. Really, there's no point to CNC unless the shapes are wacky enough to support it.
that' my two cents from the backside of the theater.
I got over big machines a long time ago. It's a ton more interesting to build big things with simple cheap tools.
Start asking yourself these questions:
1) how can I make 2 d shapes from flat material that fold, slot into one another, interlock or fasten to each other simply that...
2)...build a 3d structure that is much more interesting and strong than the shape of the 2d pieces alone would suggest?
and
3) ...that the shape of the 2d pieces is ALSO interesting
and
4) ...that the final structure is way more visually interesting AND functional than something that's built in a standard way?
I mean, all you've done is created a really fancy potting shed with a completely leaky roof and no walls, that's so structurally unstable that you're worried if someone climbs on it. Any one can go down to home depot and buy a shed for $800 that runs circles around yours in terms of functionality.
You can cut the panels on a table saw, and use biscuits to join them. Simple, easy.
It is the connection between units that will get you into trouble, and you haven't even started thinking about them. There isn't a lot of material there to make the joint with, if you're adamant about using only wood.
Why make it? What is it?
Get some ribs on the inside of the edges. Don't trust dowels or biscuits only.
maybe something like this for the joints. i think it could accommodate the odd angles.
A well glued biscuit joint should suffice for joining the individual units. We're talking about a single thickness of 3/4" plywood, probably, that needs to support the weight of 2 or 3 people? That bolted connection above is clunky and bad. It will only encourage the students to use the CNC, which will give them a stepped edge, rather than a clean saw blade edge. Less surface area meeting means weaker glue joint.
Failure will happen where the units meet.
I take back what I said. I overlooked the 5' length of the sides.
Your project is ill conceived, and will hurt someone if built. It will also be a waste of trees. Don't do it.
++ gruen
This thing is riddled with obvious structural and fabrication problems. The thought process shows little sensitivity to the nature of the materials or understanding of the fabrication techniques necessary to make it.
Trying to produce this via CNC (or anything else for that matter) without first working out all the details in prototypes will result in a big disappointment. Fabrication is a critical part of the design process.
The best way for you to figure this out is to stop talking and start mocking up full scale joints to test, where you will quickly find out what doesn't work. Make sure you model at least a full corner as the angles will make assembly of various joinery systems problematic, to say the least.
What I see here (based on my studios as a student, AND my years teaching studio)
1) I see a professor pushing CNC technology on the students. This is dangerous, but it seems to be the way academia is doing it these days.
2) I see students trying hard to please the professor without understanding the point of it. Heck, I can't really see the point of it either AND I've used CNC fabrication as part of my business before.
3) I see that design is hard and the lightbulb hasn't turned on for the students yet. It probably hasn't for the professor either, but the professor needs to take the extra step and explain a couple of things to the students:
a) Architecture must be - at its basis - be dry, functional, safe and efficient
b) Beautiful architecture must operate on at least 2 planes of creativity in order to go beyond the banal.
c) Innovative construction technique MUST do both of the above AND create things that cannot be created by other means. I personally think Frank Gehry's work (for example) fails on point a, which makes points b and c moot.
To further criticize the design - it's structurally unstable because you've built in MANY failure points and you have weak connection locations. Actually, the strong point of any geodesic dome is in the solid connection between the connection points - the connection points are located on the intersections of the points of the form, NOT along the ribs of the form. You've inverted the connections. Go back and take a close look at Buckminster Fuller. You might also look at Konrad Wachsmann.
I think the design skill to do this type of design successfully is difficult to learn - it takes a good understanding of structure, of materials, of fabrication techniques and some trial and error. I was experimenting with some of this years back - some "cut this fancy shape on a CNC" then "tab A inserts into slot A and is fastened" and "tab B inserts into slot B and is fastened" along with "bend along this dotted line" type of stuff. It can create some really neat stuff that's easy to build and looks great and functions great. Scares the crap out of most contractors and fabricators though, because they don't understand it, and it's therefore very expensive and unrealistic in most real world architecture. Some of your professors probably do it, because they have access to free machines that cost the rest of us hundreds of dollars an hour. Heck, I'd do it if I had those machines too LOL.
can you biscuit joint something at an odd angle like that? the cut for the biscuit would have to go in at the right angle. seems like that would be difficult, but i suppose someone who knows what they are doing could figure it out. i also don't see how you would clamp it together for the glue to dry, if that's something you were going to do.
it seems to me, as someone who doesn't know what they're doing, something like a dovetail might be more forgiving. it wouldn't look as clean though.
also, i agree with all of the general 'that's a bad idea' comments.
Are you going to make someone stand in there and then make them roll down a hill?
The connectors shown are actually lap joints.I know they arent very descriptive in the image but I can upload a better picture of the detail a little later. This is just for someone to sit it, it will be chained to the ground and remain stationary. It is just as an exercise in making something completely out of wood.
Take a look: http://makezine.com/2015/05/18/plywood-dodecahedron-snaps-single-piece-construction/
Well if it is just to make something cool, then make it cool. I have no experience with this, just googling stuff.
Credit to this Etsy store https://www.etsy.com/listing/211042671/dodecahedron-star-ornament-made-from?ref=related-0
++ tinnt's link not only shows that it's been done but how some of the problems were addressed.
Best quote from the video: It's one of those things, we just made it because we could.
Alas.
i'm hearing you say, 'stationary.'
have you thought about rolling it down a hill while someone is sitting in it? would be way more cool. maybe you don't want to tell them it will be rolling until they're already seated.
if there was some fire, too, that would be awesome. much better than rehashing the youtube one that's already been done.
run a simulation and it will tell you where most of the stress occurs. That is where you want to put a connector
Hmmm. I'm trying to decide if that's the stupidest thing I've ever heard. Koz, you should move to Texas and run for congress. You be a shoe-in.
And thanks for living up to my expectations for this thread.
Well, I feel like a dumbass for misreading the drawings - but, the drawings don't really describe what they are trying to show very well.
I think that Miles has it right - mock from a distance, because otherwise there's a pretty good chance of looking foolish, like me, or you, Koz.
The right answer for connector location: rethink the whole goddamn project.
^yes, rethink the whole project because you are running into a plagiarism problem, no doubt your classmates and professors have seen the fablab thing.
If I was teaching and a student went online posting "how to make" questions I'd fail them on the spot. Figure it the fuck out, that's what school is for. And life.
Guys, this isn't for any class. This is a personal project a few friends are working on.
I appreciate the input but the design isn't changing, we will just have to work with what we have.
Maybe I should rephrase my question and make it a bit more direct. Forgetting all about the actual design of the structure itself.
Say you have a dodecahedron (no context, no reason, this is in a vacuum ok, lol). Each plane of the dode is its own unit. If you needed to keep them in place with a connector, where would be the best place to affix the ties along the edges and why?
*sigh*
put your connections along the entire edge, each edge, spaced 6"-8" apart.
that of course makes no sense because i don't know what kind of connection you're using. if you just glue it all together, then it's a continuous connection right? that might be a good option.
there's a you-tube video on this thread where someone already did this. why not just do what they did?
as a side note, below is also an entirely wood project, where someone can sit in it. you could easily chain it to the ground if you so desired. seems to meet the performance requirement, if not the dodecahedron requirement:
Sheesh you architects can't answer simple question without debating the purpose of the structure she is building, or the merits of digital fabrication tools in modern curriculums, or what have you.
Kozumelle: that's what makes us architects not just builders or artists. We think about things holistically because we can't help it. To not consider purpose, structure, and aesthetics all at once is impossible. That's how our minds work.
I looked up an FEA image of a dodecahedron, I got a few search results like this:
Which leads me to think that there is additional force at the points/nodes of a polyhedra, Which means it needs extra support there somehow? Or am I way off base here?
There you go, Paula G, all done! Just send that to you 3D printer or whatever and go have lunch.
We think about things holistically because we can't help it. To not consider purpose, structure, and aesthetics all at once is impossible. That's how our minds work.
It's not that we can't help it: it's what we are supposed to do. That's what all the training is supposed to be for. And that's what makes threads like this (and architects, architectural school and students, as well as people in general like this) so absolutely infuriating.
Point and click, zero understanding. TaDa!
Since there;s nothing riding on this other than failure and shattering your sense of knowing, just experiment. You'll learn much more through trial-and-error than any one of these programs can ever teach you at this point.
Wow, some of you guys are unnecessarily rude.
I am asking because we don't have much time left to produce it. I thought this was a place to get help and suggestions? How helpful is it to basically tell someone to F off and go google it themselves? Coming here for suggestions was a last resort.
Let's get this clear: because you procrastinated and had no other recourse, you came here looking for immediate help. Hoping for a simple fix-all solution to conceptual problems in your project that hasn't been addressed.
Did you expect a welcoming parade?
Anybody suggest hot-glue for the joints?
I was not expecting anyone to do anything for me, merely suggestions or maybe a straight forward answer. So far everyone has just made rude comments.
Nevermind, I was clearly mistaken in what kind of posts are welcome on this board. Thank you for your time anyway.
So far everyone has just made rude comments.
i said put your joints at 6"-8" o.c.
you couldn't even clarify how you were making the joints.
tintt even gave you clear examples of successfully completed precedent projects for you, so didn't even have to take the 2 minutes to google.
jeez. talk about rude. i ask you one simple question and it's just nag nag nag?
what connection? how you putting the pieces together?
Paula G, this thread should serve as a learning experience for knowing the right time to ask questions and knowing the right questions to ask. IMHO, you missed on both of those accounts.
You are asking the questions at the wrong time. You should have started asking questions a lot earlier in the design process when you weren't so invested in making one design work the way you think it should.
You are asking the wrong questions. It isn't as simple as what connection or joint will make your design work. You need to figure out the right questions to ask and I think a lot of the posters here (regardless of their level of rudeness) are trying to point you to the right types of questions.
My honest opinion is that you should make the structure and the panels distinct and not try to make one thing work for both. Create a dodecahedron frame using dimensional lumber, or steel, or whatever. Attach panels to this frame to hide the structure. The level of detail and the types of connections and all the other stuff after that is up to you and how far you want to get into the design and fabrication and the refinement of that.
If this isn't for a class and is just a bunch of friends working on it ... what's the rush? Why not take the time you need?
The problem with the design is that the points of the hexagons should touch.
and the form should be skeletal made with straight members....2x2s or something.
Done.
How helpful is it to basically tell someone to F off and go google it themselves?
Aside from the fuck off part (which I don't recall anyone saying), isn't that exactly what you were doing? LOL
Coming here for suggestions was a last resort.
Obviously. The sense of entitlement that some people have is astounding. When the center of the universe is finally discovered, more than a few people are going to be shocked to learn that they are not it.
I believe I nailed this one with my first post (#3).
"I was not expecting anyone to do anything for me, merely suggestions or maybe a straight forward answer."
There's a link in here with the exact life-sized thing you wanted to build. Already made. Might as well just copy it and understand why they did it that way.
This is obviously an installation for Burning Man. Pack in, pack out, so this needs to be put together on the Playa, and using a CNC for multiple parts will make it easier to put together in the desert, because all the parts are identical. And Labor Day isn't too far away, especially if you are still in design. Mockups and experimentation is the only way to go. Good luck.
No, the problem here is that she didn't get the answer she wanted.
At least she appears to have gone away - some people just obsessively persist in posting the same thing, hoping that eventually they will get the answer they want.
we did everything but build it. . .
if this kid is the future, i'm going to stop recycling
Curt, gold star for that comment.
The sIdes of a dodecahedron are pentagons. Did you know that the ratio of the diagonal of a pentagon to its side is the Golden Ratio? Really!
There's some magic in that dodecahedron, I tell you.
This is the only dodecahedron who contains magic.
This one has magic. Roll it to see if you escape the evil dragon-goat
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.