Back to Synergetics Gallery
For many of the tensegrity structures I've designed, I've provided here the datasheets which I used to assemble them. For the most part, the information the contain is raw input to or output from the software I've written to design the structures. (The software is currently not publically available, but the logic behind it is available in A Practical Guide to Tensegrity Design.) The datasheets are put together somewhat crudely and hastily, but I've gone back and tried to put in information to allow other people to interpret them. In addition to these datasheets, there are some over with the press releases. These are listed in the second group below. Also I sometimes put data on member lengths and relative prestress forces directly in the main narrative describing the structure. These are listed as the third group below.
These datasheets hopefully will give people who are interested in building the structures a handle on the information necessary. Information on the dimensions of the final structure is usually not given.
The following datasheets are mixed in with the press releases:
The following files contain data on member lengths and forces as part of their narratives:
Gerald de Jong once asked (email February 1, 2008) about my approach to building tensegrity structures. The relevant extract from my reply is (email February 1, 2008 — cross posted to several public boards, see the Index for the ones I use — I've changed and added some words here and there to clarify things and provide more detail):
“My design procedure is iterative, but not incremental. I start out with a whole of some sort based a geometry I have in mind. I know enough geometry that I can rough things in pretty well. Then the minimization techniques take over to bring it into a feasible configuration. Or the design may just collapse or go off into some configuration that doesn't look interesting so I have to jiggle some parameters or rethink how I'm connecting things.
“Definitely my design procedure is not Tensegritoy, that is it doesn't involve physical prototyping though certainly experience from previous models I've built is important. When I build, I don't use highly elastic tendons. Braided nylon is about as far as I will go as far as elasticity. My computational procedures are flexible enough that I've never felt the need to use any physical prototyping technique. Certainly I've seen other people come up with some pretty interesting things using prototyping with elastic materials, it's just not my thing. And I think elastic physical prototyping definitely has its limitations. Snelson has an interesting approach using bead chains and slotted struts. That's the only inelastic physical prototyping technique I've seen.
“Once the design is complete and looks feasible, then I think on my materials and draft a realization according to the materials. At this point, I leave the computer for the most part, except I might refer to one of the viewers when I can't quite see how things fall from the 2D snapshot I'm working from, but a carefully chosen 2D snapshot with point labels is fairly adequate for the most part. Most of my realizations involve braided nylon for the tendons (nylon twine for larger structures) and wood of some sort for the struts.
“My method for attaching tendons to the struts varies. For my initial models I almost always used screw eyes screwed into the end of the wooden-dowel struts. Lately I've been favoring single nails driven into each end of the dowel. Splitting of the dowel when driving in the nail has been a problem with this approach, but it seems to allow for slimmer struts and more elegant looking models, at least by my aesthetic standards. For larger-scale outdoor structures I drill holes directly into the garden stakes I use, and I may try this approach at some point for smaller-scale indoor structures.
“My approach to assembly depends on the model. For domes I tend to take a modular approach and build some subassemblies first and then tie the subassemblies together. For prisms I start by assembling one end and work my way to the other end. Same way with masts. It can be very difficult to get the topologies right, and keep track of which tendon should be what length, and more frequently than I would like I find I've tied something wrong and I need to re-tie. Sometimes I use a jig to hold things temporarily, but things have to get pretty confusing for me to resort to such a thing. I try to tie the tendons I know will be tight (according to calculations) as early as possible when things are looser. I try to get all the struts tied in as early as possible.
“It's nice when some tension develops because that's when things start moving from being a shapeless mass to something definite. Of course then I have the problem with having to tie the tendons in a tensed state. That can be a plus and a minus. The minus is that I have to hold the nodes together somehow while I do the tieing. The plus is that many times a knot will stay tied if its always under tension, but may come unraveled in a loose state where it is constantly being jostled. So very often the knot I use depends on whether I'm tieing under tension or not.
“Going for the whole too early can result in a shapeless mess that's hard to deal with. I'm thinking in particular on a torus where I usually proceed as with a mast and it doesn't develop into a complete loop until the last.
“I have to constantly use my imagination to build a tensegrity in the early stages, since it can take awhile for it to start looking how I expect it to, and in the beginning it may diverge quite a bit from the final configuration.
“I don't consider construction my forté. I figure anyone could manage somehow given a design, and putting together accurate specifications was the main hurdle for me when I first started assembling my own tensegrity designs. I probably could have gone far with physical prototyping techniques if making models had been my only interest, but my main interest was engineering large-scale domes, and elastic modeling techniques seemed hopelessly cumbersome and inexact to me, so much so that I didn't even try them.
“Construction technique may be a hurdle in getting tensegrity into the mainstream. I don't think construction is too difficult ex post. With a major project, I think putting together a prototype would teach me a lot. Once the structure is understood, and I know what materials I will be using, a construction sequence can be formulated easily enough.
“For a major project, safety is a big consideration since if a tendon lets go, it, or the strut it releases, can pack a big whallop. That's one thing I like about my dome technology is that enough tendons are involved that one doesn't make a big difference. So flying struts are not likely to be a problem. But just a wire cable moving at a good clip can cause quite a bit of damage, so make sure the fasteners you are using can take what you are dishing out and then some. I once tried to quiz Snelson on this and basically what he had to say was ‘it depends on where you're standing.’ Yes. Don't stand in the wrong place.
“Early on I used metal wire for tendons, and attached them by putting the wire tendon through a screw eye and then twisting it to secure it in place. That seems faster than tying the nylon, but the nylon structures are so much more pleasant to have around. Wire can be very handy in assembling a nylon structure as a way to tie two nodes close together so I can get the nylon in place. Once the nylon is in place, the wire can be removed and reused to help tie the next tendon. However, with the split-level prism I just assembled I didn't need wire and just pressed the two nodes together by putting one on a hard surface where it wouldn't slip and pressing the other toward it while I tied the tendon.”
Your comments or questions regarding these datasheets are welcome and may help improve their presentation. Experiences you have had assembling these structures and photos of the results are also of interest. You can send email to email@example.com or Postal Service mail to:
Cambridge, MA 02142-0021
Back to Synergetics Gallery