Close
0%
0%

Folds and Hinges Technology to Make Mechanisms

Engraving by CNC a special composite, the Hylite, to create folds and hinges. Zatsit, so made, is on Kickstarter until oct 5, 2018. Hurry!

Similar projects worth following
I propose a new technology for Makers: it is about creating articulated mechanisms with live hinges, and various mechanical structures, by engraving by CNC a sheet of a special thin composite (1,2mm), the Hylite, whose facings are in aluminium alloy, and the core in Polypropylene (PP).

One knows the boxes that are formed by engraving an ALUCOBOND, or DIBOND composite, etc... (Aluminium + polyethylene), and bending along the engravings. With Hylite, we can do the same, but also something else very interesting: by engraving the two faces in concordance, we can obtain a precise hinge, in the zone where the PP will have been exposed by the engraving.

This technique allows to realize many kind of mechanism associating rigid parts, with folded stiffeners, boxes, etc..., and hinges, the whole in a very satisfactory aesthetics and robustness, and also a very good precision, since it is that of engraving by CNC.

This delta robot, Zatsit, is a very convincing first example.

The engraved Hylite technology

Hylite composite was originally developed to lighten body sheet metal in the automotive industry. Its ability to produce hinges, because of its polypropylene core, is cited by the manufacturer, but had not given rise to significant developments.

PP is very suitable for forming strong and durable live hinges, as shown by the millions of moulded boxes with integrated lids that are used everywhere. With the Hylite composite, such live hinges can be made in any complex geometry, by CNC engraving. By acting on the engraving profile, the mechanical characteristics of the hinge can be adjusted, for example the precision according to the bending stroke.

Moreover, the composite itself is very rigid, so that the deformable engraved areas are, by contrast, very precisely located. The articulated mechanisms thus created are thus precise, and without geometric uncertainty.

I have done some research, and many prototypes, to constitute a catalogue of engraving profiles adapted to different uses as stiffening folds, recesses, robust hinges capable of turning at 180°, or very precise hinges for limited strokes, etc.

It is this work that allows me to talk about a new technology, the engraved Hylite.

Durability of the engraved live hinges

I have performed extensive tests to ensure the durability of the live hinges, over more than 14 million bending cycles, without them being altered in any way. One of my prototypes has been running as a 3D printer for over two years with exactly the same mechanical precision. It's very reliable, and very strong! The tests are reported here: 

A convincing application :  Zatsit

The delta robot Zatsit is a first example entirely made up in this way. The technique of engraved Hylite makes it possible to obtain the totality of a very light delta mechanism, and geometrically very exact, starting from a simple sheet of Hylite. The result is extremely efficient, in terms of precision and repeatability of movements, resistance to vibrations, durability and ease of building.

Zatsit is just one example. This technique, consisting in engraving Hylite by CNC, to constitute technical or aesthetic, complex and multi-functional objects, is applicable everywhere. 

Zatsit as a generic delta robot

Zatsit was designed as a universal robotic platform, on a linear delta architecture. Its main qualities are low cost, ease of construction, and performance in speed, repeatability, and accuracy of movement.

The holder is hollow, to be able to install any tool smaller than a cylinder of Ø66mm. The fixing device consists of three M3 screws, arranged at 120° on a Ø70mm circle.

Adapting, or hacking, an existing tool, or a specific device, is therefore very easy. It is generally sufficient to make an adaptation board, either from a Hylite disc proposed for this purpose, or in 3D-printing an ad-hoc plate.

The lower the mass of the effector, the better the robot's performance. With lightweight effectors, such as Pick & Place manipulators, or 3D print heads, speed and acceleration can be spectacular. 

With heavier tools, acceleration must be reduced. The articulated mechanism is robust, but due to its relative flexibility it is not suitable for high forces, as in the case of milling. On the other hand, engraving is possible with a light spindle.

The screw mounting system is completed by a three-magnet system, useful for mounting light accessories, e.g. cameras, air diffusers, sensors, etc., in addition to fixed tools.

As an example, a tool for engraving glass by vibration/percussion was adapted very simply, by 3D-printing an adaptation...

Read more »

  • 3 × Makerslide profile 600 mm long
  • 2 × CNC engraved HPL plate with built-in necessary threads
  • 1 × Hylite generic holder to be customized for any hacking
  • 1 × Gimballed support for a Titan flystruder 2 hylite plates and an elastic string
  • 1 × Water cooling system for the hot-end Tank jar, pump, silicone tube, electronic step down PSU

View all 13 components

  • Zatsit on Kickstarter, now!!!

    Marc Peltier10/06/2018 at 22:00 0 comments

    Hi!

    I am really happy and proud to announce that the Kickstarter campaign has been launched!

    Here is the link: Zatsit on Kickstarter

    I had a great joy in designing and building this machine. I see another great joy, it is to initiate a community of Zatsit users, all around the world...

    Thanks you, Friends!

  • Zatsit, Zat's FAST!

    Marc Peltier09/05/2018 at 15:03 0 comments

    Hi!

    In this log, Zatsit shows its ability to run quite fast.

    The purpose of this test is to highlight the quality/speed ratio.

    There are many printers that can print a satisfying object by going very slowly, and many can also go very fast by doing just about anything. 

    What characterizes a quality printer is to go fast without losing quality.

    This test is made possible by a remarkable filament, PLA+ uDiamond from Carbodeon, which is loaded with nanoparticles, in order to change the thermal properties. Print speeds can be tripled with the same hot-end.

    Even with this filament, we have to stop at 560 mm/s, whereas the mechanics could go up to 1m/s!

  • Another ultra-light printing head

    Marc Peltier07/25/2018 at 17:11 4 comments

    I continue to explore the compact 3D print heads, with the integrated extrusion function.

    This one weighs only 29g, and is more accurate and robust than the previous one published.

    But it is not easy to build at the prototype stage, because of a difficult silver solder. In mass production situations, however, things could be simpler with another technology.

    This version could be declined in several extrusion paths, but only one extrusion motor, remote. If I can keep it light enough (I hope about 40g), it would be a major advance for delta printers, which will also be able to print objects in several materials including TPU.

  • Ultra-light extrusion : toward the Graal...

    Marc Peltier06/12/2018 at 12:23 1 comment

    Zatsit works even better when the effector is light, especially as a 3D printer. So I try to find ultra-light solutions, without any compromising on the very best functionality.

    The extruder is one of the components that most influences the quality of 3D printing. Ideally, the filament should be pushed by the extruder as close as possible to the area where it will be melted, so as to avoid inaccuracies in the extrusion caused by the elasticity of the various components present on the guide path, including the filament itself. Delta 3D printers are at a disadvantage in this regard, as it is not possible to install a sufficiently powerful extrusion assembly directly on the printhead, due to the associated weight.

    In addition, the extruder must drive the filament in a powerful, efficient and precise manner, which is best achieved by pinching it between two symmetrical hobbed gears.

    The prototype system I propose today meets these two criteria: two symmetrical hobbed gears are driven by the same worm gear, in a compact and very light unit, located just above the water-cooled Hot-End. The filament is pushed within 25mm of the melting zone. 

    The worm gear axis is driven by a flexible shaft connecting it to a stepper motor, flexibly suspended, and gimballed, between the three carriages of the delta linkage. In this position, the center of gravity of the stepper moves very little, and its inertia is thus neutralized ("Flystruder").

    The extrusion mechanism is mounted floating on a Hylite support which also carries the water-cooled Hot-End. It is the filament that automatically centers the extruder mechanism on the extrusion path.

    This device is therefore very close to the Graal of Delta 3D printers: a powerful, precise, and above all DIRECT extrusion head, that is also ultra-light.

    Here, the whole unit weighs only 32g, including the support, the extruder, the Hot-End, and the cooling of it! Yes, 32g! This is about one-tenth the weight of a conventional direct-drive extrusion assembly, which does not always have a double hobbed gear drive!

    As a result, the ratio between speed and print quality is still improving...

  • Example 2

    Marc Peltier06/02/2018 at 18:08 0 comments

    Remote control of the focus on a telescope


    This was, for me, the very first application of engraved Hylite technology: Create a Sarrus mechanism, in Hylite, to be able to focus a small telescope at a distance, without having to touch it.

    Cheap amateur telescopes, such as this venerable 114mm newtonian, often have mounts that are a little too frail. At high magnification, just touching them to adjust the focus makes the image shake in a way that makes focusing very difficult.

    This device includes a Sarrus mechanism, deformed by the action of two air bellows. A silicone tube connects these bellows to two syringes of different diameters, which the operator holds in his hand. 

    By acting on the pistons, without touching the scope, the extension of the bellows is varied, and thus the position of the eyepiece. The two syringes make it possible to obtain two ratios: coarse focus or fine focus.

    The improvement in use is magical! Very fine adjustments can be made with a perfectly stable image, as if it were a very high-end telescope.


    It works so well, that I'm thinking of updating this application, to make it a commercial product, after Zatsit.

  • Other examples

    Marc Peltier06/01/2018 at 13:53 0 comments

    Many objects can be made by engraved Hylite technology. I made some of them, apart from the Zatsit robot and its accessories.

    iStand, a folding stand for smartphones

    I realized this as an exercise of style and design: From a format equivalent to that of the smartphone itself, how to make a photo stand for this smartphone?

    On this occasion, I enriched this technology with a trick: the integration of round magnets in the Hylite, to make fastenings, active in two different configurations.

    Here is the object, folded :

    Then, unfolded

    And finally, upright as a photo stand. Tilting is adjustable! 

    Done!

View all 6 project logs

  • 1
    The detailed assembly of the Zatsit kit, in 7 videos

    1 - Frame

    2 - Mechanism

    3 - Integration

    4 - Motors

    5 - Hot-End (in french for now)

    6 - Flystruder (in french for now)

    7 - Accessories (in french for now)

View all instructions

Enjoy this project?

Share

Discussions

Michał wrote 02/10/2020 at 18:50 point

An extremely smart solution! Bravo!

  Are you sure? yes | no

Paul wrote 06/12/2018 at 14:38 point

This is brilliant. Flexures and linkages have a lot of potential for robotics and manipulators, and this is a very clever and effective way to manufacture them. Flat-pack robotics, essentially. Have you considered etching or laser engraving and cutting this material?

  Are you sure? yes | no

Marc Peltier wrote 06/12/2018 at 16:04 point

At the beginning, I had the water jet cutting tested on the Hylite material, with random results: the settings to avoid blistering are delicate, and the cutting is not very clean anyway.

The laser is not very indicated on bright aluminium.

In the end, CNC milling is the most flexible and gives the best results. Some engravings must respect a particular profile to optimize the hinge, and in this case, there is no choice: it is necessarily by CNC milling.

  Are you sure? yes | no

Paul wrote 06/12/2018 at 16:15 point

Understood. Thanks for taking the time to answer.

  Are you sure? yes | no

Barry wrote 06/01/2018 at 07:30 point

This is amazing! Can't wait to follow this development.

I don't know if you've considered a flex-shaft on a dremel style tool for engraving, but it might be a good way to keep the gantry light. Almost like a bowden setup.

Any ETA on a kit? :)

  Are you sure? yes | no

Marc Peltier wrote 06/01/2018 at 09:25 point

Remote powered engraving bit : indeed, why not ?

I have already installed stepper motors in flystruder situation, on a gimballed support. You can do the same thing with a Dremel...

For the engraving, I proceeded differently: I used a small brushless motor, whose axis was Ø3mm, and I quite simply replaced the axis by a Ø3mm engraving bit. The set weighs less than 50g, and it works very well!

P.S. I don't practice acronyms in English : What is ETA?

  Are you sure? yes | no

Barry wrote 06/01/2018 at 18:31 point

The brushless is likely a better way to do engraving as engraving only requires low torques - although it would be dreamy to use a diode laser or something with 0 side-load.

Have you tried any speed tests? It seems like the low weight of the gantry would allow speeds of perhaps 200mm/s or more (without unexpected forces, that is!)

Apologies - ETA=Estimated Time of Arrival - I would love to buy (or beta-test) a kit for the delta machine, it looks too cool :)

  Are you sure? yes | no

Marc Peltier wrote 06/02/2018 at 06:34 point

Without filament, with motors 0.9° under 24V, 1.6A, I tested the speed up to 600mm/s, with 3G acceleration, without step loss. But no one can extrude at this speed, and such acceleration only serves to shake the machine unnecessarily!

What limits the speed is the volume flow of the molten plastic, and even more, the cooling of the plastic that has just been deposited.
With a Titan extruder in Flystruder position, a 0.4mm nozzle and a layer height of 0.2mm, under-extrusion defects appear at about 250mm/s. I printed at 420mm/s, but with a layer height reduced to 0.08mm, to limit the volume flow.

Speed in absolute terms doesn't mean much. What characterizes a quality printer is the ratio between speed and print quality.

  Are you sure? yes | no

Barry wrote 06/02/2018 at 08:25 point

Speed still opens the door to some interesting applications like pick & pluck or coordinate measurement, it can be useful! Have you been able to measure the precision/repeatability? I imagine that's another benefit in eliminating mechanical components, oftentimes flexural components are used in nanopositioning for that reason, I'm curious how well the principle scales up?

  Are you sure? yes | no

Marc Peltier wrote 06/02/2018 at 11:05 point

About precision and repeatability :

Delta printers often have good repeatability, but are often less gifted for precision, because of the geometrical requirements which are rather severe (equidistance and parallelism of the linear guides, length of the arms).

Zatsit provides an efficient answer: all the geometry is derived from CNC cutting and engraving, so that no inaccuracies can be introduced during assembly.

I haven't done any specific tests in metrology, but as a matter of principle, any 3D printer is an implacable "micrometrograph": we appreciate very well, with the naked eye, any misalignment between layers. In my opinion, this visual appreciation has a sensitivity of 10µm, or less. In this respect, Zatsit excels. The alignments on the vertical walls seem perfect, when the extrusion is regular. The irregularities which can occur are always due to malfunctions of the extruder, or to the wet filament.

Moreover, unlike other delta machines, Zatsit is exact: the dimensions of printed objects are respected to within 0.01mm (along X, Y or Z), if the extrusion is well calibrated.

This geometric quality is confirmed by the procedures offered by firmwares in order to correct the inaccuracies of delta printers. For all my prototypes, the correction proposed was about zero, whatever the parameter.

About Crowdfunding:

You're right, but Zatsit is different.

The kit only concerns mechanics. And besides, I only have two subcontractors for the whole production. So things are easier to control.

During the design, I always thought of these questions: How am I going to get this or that done, by whom, and how can I avoid problems?

Target price :

I hope to be able to propose the whole kit around 300€ (as usual, some better deals for "Early birds")

  Are you sure? yes | no

Marc Peltier wrote 06/02/2018 at 06:42 point

ETA : I should launch a Kickstarter campaign this month, hoping there's enough interest...

  Are you sure? yes | no

Barry wrote 06/02/2018 at 08:48 point

I'd be happy to help you spread the word, I've been enraptured by compliant mechanisms for about a year and it's nice to see them getting some attention.

Launching a 3d printer on kickstarter will be an uphill battle because of how noisy the market is, people need to have a lot of confidence in your scale-ability. Everything else in this project is really well thought-out, but expect that burden of proof of manufacturing viability to be considerable.

Have you decided on a price-point yet?

  Are you sure? yes | no

Castellin.elie wrote 05/31/2018 at 05:07 point

a real futuristic project! Innovation and research in lot of field! 

  Are you sure? yes | no

Jouis wrote 05/30/2018 at 15:03 point

Project of exceptional quality with a great capacity for evolution and customization!

  Are you sure? yes | no

Marc Peltier wrote 05/30/2018 at 16:09 point

Thanks ! Indeed, so many things to do this way !

  Are you sure? yes | no

Similar Projects

Does this project spark your interest?

Become a member to follow this project and never miss any updates