Back to Philip Stewart

Reply to first email from Philip Stewart

From: Robert W Burkhardt
Date: Thu, 8 Nov 2001 11:13:52 -0500
Subject: Re: tensegrity engineering

Hi Stu,

Thanks for your interesting email.

I believe the system I use is flexible enough to handle any tensegrity whatsoever, highly symmetrical, highly asymmetrical, etc. I can compute lengths and static stresses, but I'm in the dark about dynamic stresses.

Since I started working for Gamesville (now part of Terra Lycos) a couple years ago, I haven't had time to do much serious work with it though I do hope to get back into it. I had an email correspondence with Kenneth Snelson for a brief period. He was upset over the way I'd labeled the tensegrity mast, and I fixed the labeling to his satisfaction I think. In the course of our discussions, I asked him if he had coordinate values for things he'd put together. I wanted to try his values in my system and see what it had to say about stresses etc. and get his feedback on whether this was what he experienced in working with his structures. He didn't provide any data. I think he must have such coordinates to have been able to do the computer graphics he's done with his structures.

I'm inspired more by Bucky's engineering visions for tensegrity. I've been trying to develop a geometry suitable to the large-scale tensegrities Bucky envisioned (dome covering Manhattan etc., although I think I'd like to start first in a climate that doesn't have to deal with freezing weather).

I'm not a formally educated engineer. I'm OK with math, and know some basic physics. With the lack of torsion (at least in the macro sense) the stresses in tensegrity I think are rather simple to deal with it, so I don't think I'm being too outlandish trying my hand at computations here. But I think I do need feedback. I made proposals to the NSF a couple years back and they thought I should get the cooperation of a civil engineer which sounded like a good idea to me.

As you probably saw though, I've done some purely aesthetic explorations. One experiment, which is still incomplete, is to put a meta-procedure on top of my basic procedure to align the struts in a tensegrity in various ways. I think this is mostly a matter of developing the software. I attempted this years ago in a structure I called the aligned diamond tensegrity, but I didn't get the constraints right and it turned out it wasn't aligned the way I wanted after all though it was somewhat close. I think I know the right constraints now and I hope to go back to that problem and do it right.

If you have web material on your designing efforts I'd be interested to see it. The best address to reach me at is my webmail at I only check the Juno mailbox once a week or so.


Back to Philip Stewart