Hey! This thing is still a Work in Progress. Files, instructions, and other stuff might change!

Delta-T: Tensegrity 3D Printer

by LoboCNC, published

Delta-T: Tensegrity 3D Printer by LoboCNC Sep 19, 2015
0 Share
Download All Files

Thing Apps Enabled

Order This Printed View All Apps


Liked By

View All

Give a Shout Out

If you print this Thing and display it in public proudly give attribution by printing and displaying this tag.

Print Thing Tag

Thing Statistics

18154Views 1114Downloads Found in 3D Printers


This experimental printer uses fishing line to suspend and move the print head in the workspace. The lines are all kept in tension by a spring-loaded tension pole that pushes down on the print head carriage. It is, in effect, a dynamic tensegrity (https://en.wikipedia.org/wiki/Tensegrity) structure. It looks kind of like a delta printer, but the kinematics are a little different. This is a great project for anyone wanting to spend a great deal of time figuring out how to build (and improve upon) a unique 3D printer.

I know of a couple of other designs using a similar approach, so you might also want to check out:

Watch the video: https://youtu.be/J6X3oLVhHF4


These instructions are far from complete, but they'll serve as a starting point if you are interesting in trying to build this printer. Just post any questions you have in the comments, and I'll try to fill in the gaps as we go. Keep in mind: even though I managed to get my printer working, it will take a lot of ingenuity on your part to to make your own. This project is not for the faint of heart!

The frame is formed by laser-cut 6mm acrylic top and bottom plates (see DXF files) separated by 3/4" dia aluminum tubes (11" long) and clamped together with internal 1/4-20 threaded rod. The build platform is actually just the internal cutout from the bottom plate rotated by 60 degrees. It rests on three 10-32 leveling screws that are topped with 3D printed adjustment knobs and round rubber bumpers.

Print Head Carriage
This printer uses a Bowden-type extruder, but instead of the filament being fed in-line with the extruder nozzle, it comes in at a right-angle. This is necessary because the tension post moves around in the area above the carriage and would interfere with the Bowden tube. The carriage plate is 1/16" aluminum plate (see DXF files). I won't get into the hot-end/extruder details here, but I can provide more information up request.

This is the most complicated part of the printer. It uses 30# test Spectra fishing line which has very little stretch. Each of the 3 sides of the carriage plate is supported by two separate lines, similar
to a delta-style printer. For any given side, both lines wrap onto a single 10mm dia. spool on the motor shaft. For one side, the line goes from spool, thru the eyelet on the upper plate, down through
the eyelet on the carriage plate, back thru the same eyelet on the upper plate, and then finally to an adjustment screw. It takes quite a bit of patience to get this strung up. Examine the photos and
video carefully to see how the lines are routed. Note that each line has an effective pulley reduction of 2:1 which doubles the positioning resolution.

Eyelets & Adjustment Screws
The eyelets are formed from 0.035" dia hardened steel wire. I started by making a U-shape in the wire and then shoving the ends of the"U" through the holes in the carriage and upper plates. I then placed a thin spacer inside the "U" and folded the legs over to make sure that all the eyelets had exactly the same height. Finally, I cut off the excess wire with a Dremel cut-off blade.

To prevent abrasion of the fishing line as it goes thru the eyelets, I slid a small piece of 0.022" ID teflon sleeve over the line at each eyelet. The teflon sleeve stays in place while the fishing line
slides through it.

The very end of the fishing line is clamped between 2 nuts on the end of a 1" long 4-40 screw that is threaded thru the upper plate. Rotating the screw will wind up (or unwind) any excess fishing line.
This is the mechanism used to fine-tune the leveling of the carriage plate. The adjustment screw can then be locked in place with another nut against the upper plate.

Tension Pole
The tension pole is a 0.292” dia, 18” carbon fiber tube that presses down onto the carriage plate to keep the strings in tension. At the bottom, it has a 1/2” dia. Nylon ball glued on that fits into a 3D printed cup. At the top, it has a cap with places to attach the tension springs. For springs, I've used 3” (active) lengths of 0.187” dia surgical tubing. Steel springs would probably be better, but the tubing is much easier for doing trial and error adjustments. I have the tubing stretched to about 6” when the print head is at the top.

To keep the springs at more or less a constant tension, they are rigged with addition fishing line similar to the carriage plate lines. The line for each spring starts at the motor spool and is suppose to wrap up as the carriage lines unwrap. From the spool, the line goes thru an eyelet at center spacer, up thru the clip at the bottom of the spring, back down thru the same eyelet at the center spacer, and then to a clamping screw on the upper plate.

I've used a Printerboard running Marlin for the controller. I configured Marlin for a Delta printer, but I modified the kinematic equations for this printer. Normally, the parameter DELTA_DIAGONAL_ROD is the length of the rigid arms for a delta printer. In this implementation, this parameter is the starting length of the strings when the print head it in the top-center position. marlin_files.zip contains the modified configuration.h, pins.h and marlin_main.cpp files I used.

The one big unsolved issue is homing because there's no good place to put limit switches. My hack has been to use square knobs on the back of the motors. Before starting each print, I position each knob manually to exactly the same orientation so that the carriage is at its top-center position. This is the origin, and I set the Z-offset in my slicer to distance the table is below the nozzle tip (a negative number).

More from 3D Printers

view more

All Apps

Auto-magically prepare your 3D models for 3D printing. A cloud based 3D models Preparing and Healing solution for 3D Printing, MakePrintable provides features for model repairing, wall thickness...

App Info Launch App

Kiri:Moto is an integrated cloud-based slicer and tool-path generator for 3D Printing, CAM / CNC and Laser cutting. *** 3D printing mode provides model slicing and GCode output using built-in...

App Info Launch App
KiriMoto Thing App

With 3D Slash, you can edit 3d models like a stonecutter. A unique interface: as fun as a building game! The perfect tool for non-designers and children to create in 3D.

App Info Launch App

Print through a distributed network of 3D printing enthusiasts from across the US, at a fraction of the cost of the competitors. We want to change the world for the better through technology, an...

App Info Launch App

Quickly Scale, Mirror or Cut your 3D Models

App Info Launch App

Treatstock is an online platform that offers decentralized manufacturing services such as 3D printing and CNC machining for business-to-business and business-to-consumer sales all over the world. W...

App Info Launch App

3D print your favourite design with NinjaPrototype, a professional 3D manufacture with consistent quality and speed.

App Info Launch App

This is a super interesting design, I'm surprised how well the prints came out. Good job for being creative :)

You venture through the countless cubes toys and other items created to say you can do better then learn and create something you think is pretty useful and cool.. and assume you have grasped the complexity of both design and the machine hardware.. Then you come across a guy like you that blows your mind with the concepts.. the designs, foresight and skill.. You raise the bar pretty high.. Yours is the best collection I have seen so far.. LOBOCNC by far..

Thanks for your kind comments! Keep in mind that I've been engineering for close to 40 years, and these days, I have way too much time on my hands:)

I absolutely love this idea. I'm curious though, could you use a telescoping spring loaded center pole instead? thereby eliminating the height requirement? you could also run all the wires and filament down the center of it which removes the wagging tail most deltas have.

I really want to make one of these now, this seems so much more logical than using the verticals as rails/tracks for motion.

You might be able to do it with a telescoping pole, but what you really want is a constant (more or less) downward force, and that'll be hard to get with a telescoping pole. Also, given the relative flexibility of the Spectra line, you really have to be careful to minimize any friction.

I can't wait your next design! You are so creative, respect to you.

I saw this and I was like no. Then I watched the video and I saw the wires and bands tightning to make the hot end move and I was convinced. This has to be the coolest Delta printer ever

Thanks! As mentioned in some of the comments below, this particular printer is not hugely practical, and all of the strings are a bit of a nightmare. However, I'm mulling over a much simpler, more practical design. Stay tuned...

This is just ludicrously clever. My hat off to you, the fact that you conceived and created this, and got it to work as well as you have just blows my mind. Granted this is obviously not your first rodeo but even so, I'm gobsmacked. Would never have thought it was possible.

Hey, thanks! Like most ideas, I've found that if you beat on them long enough, you can get them to work. But also like most ideas, once you finally get it to work, you realize it's not quite as great an idea as you first thought. I don't think this'll ever be a practical printer, but it was fun to get it working.

I don't think it matters much whether it's practical, what matters is that a concept was conceived, implemented, developed and assessed. Through this, all human progress occurs. I'm amazed at how fast you managed to remix emmett's Iris Box. Granted this is my first ever attempt at SCAD-tweakage, but it took me a healthy 20+ minutes to work out how to shrink the clearance between the outer spheroid and the base, plus another couple hours of trial and error tweaking the value and re-rendering at 120, all so I could generate a version that would print neatly at 150% with a clean stable junction. How on earth you managed to turn his design into your design in a couple of days I have no idea.

All of which sounds a bit gushy, no matter. Credit where it's due, you're an inspiration to an old programmer playing with new toys.

Actually, I used Solidworks to redesign the Iris Box from scratch. I don't know how to use SCAD, and even if I did, it would probably taken me a long time to modify it. The emmett's basic design is beautifully simply, so using a powerful design tool like Solidworks makes life infinitely simpler.That said, sometimes I really wish I understood SCAD because sometimes Solidworks can be maddeningly frustrating.


I like your Delta-T design - great work - and I have some questions.
I'm interested in the compact ultra-cheap end of the spectrum.
But maybe a little compromised with improved portability.

What I think is especially great on this design is that it save so much mass and thus is potentially one of the most portable designs possible. By using bamboo for the rods and balsa-wood for the plates or 3D-printed fractal low infill trusswork for the whole structure its mass may be reduzible even further.
Do you know the approximate weight ratios in your current design (frame:motors:electronics)?
Is there a chance the motors get to be the weight bottleneck?
With the motors at the top do you think the high up center of gravity could become a problem?

The one thing that I see is most problematic on the Delta-T for portability is the looong fragile pole sticking out of the top.
Seeing that the central pole embodies the central tensegrity design idea I'm wondering if that can be compactified so that nothing sticks out without adding excessive complexity.
A difficult request. Do you think that could be possible?

I guess attempting to making the whole printer foldable would not be a too great idea due to too much complexity and potential printer de-tuning. Do you think so too?

By using an anti-prism (a stretched octahedron) instead of a prism for the printers frame the mass increases a bit but sturdiness against external mistreatment (also paramount to portability) should increase quite a bit. If you had used an anti-prism frame would the posts have gotten in the way?

About the hot end: It seems you use 1.75mm diameter filament.
Do you think it would work with a 3mm bowden too even though the side-wards load by the supply tube increases?
Do you have a quick sketch of the cross-section of the 90° heating block assembly you've made that you could upload?

I have old cut open bicycle tires lying around. do you think stripes of that could work as springs?
If you hang 1kg (e.g. bottle of water) on 50cm of 0.187” dia surgical tubing how much does it stretch?
That is: whats the spring constant you're using?

The fishing line I have lying around (0.58mm 60kg & 0.33mm ?kg) seems to be waxed probably to reduce friction (in unloaded state it stays kinked when bent over a finger nail). Does the your the 30# spectra fishing line you're using behave the same way?

Edit: I fount that the similar bot Flying SkyDelta has a antiprismatic frame so that seems to work.
The author talks about "frameless" ceiling mounting.
There the problem that the pressure pole can't overstand into the ceiling would occur too.

Here's a forum discussion about the printer that might help answer some of your questions:

Thanks, but I already had read through the whole thread - thus the edit.

Could you take a few more pictures of the cable routing? Does it wrap 2 stings or just one?

This is a really interesting design. Are you going to continue working on it? How would you say this compares print-quality and build-cost wise to something like a kossel mini?

The print quality to cost ratio is great because the costs are down to the bare minimum (motors, controller, hot-end, some structural parts & hardware) and the print quality ended up pretty decent (although not stellar). However, the design is finicky enough that I wouldn't want to use it as an everyday printer. I designed it mostly to see how cheaply you could make something that actually printed.

Very clever design ! well done! is there any special marlin setting needed or this setup to work ??

Thanks! The Marlin files I had to modify are included in marlin_delta-t.zip. There are configuration settings in Configuration.h and I had to change the pin assignment for the fan for the Printrboard. Lastly, in Marlin_main.cpp, I modified the calculate_delta function for the different kinematics.