RepRap Project
From Wikipedia, the free encyclopedia
The RepRap Project is creating a partially self-replicating rapid prototyping machine. Simplistically a rapid prototyper is a 3D printer that is able to fabricate three dimensional artifacts from a computer-based model. RepRap is described as 'self-replicating' because it is being designed with the ability to reproduce the components necessary to build another version of itself.
Due to the replicating ability of the machine it should be possible to cheaply distribute them to people and communities enabling them to create (or download from the internet) complex products and artifacts without the need for expensive industrial infrastructure. This gives RepRap the potential to become a powerful disruptive technology.
From a theoretical viewpoint it is attempting to prove the following hypothesis:
- Rapid prototyping and direct writing technologies are sufficiently versatile to allow them to be used to make a von Neumann Universal Constructor.
A universal constructor, or clanking replicator, is a machine which uses self-replication to create new generations of itself. It is speculated that the RepRap will eventually demonstrate evolution in this process as well as being able to increase in number exponentially.
|
[edit] Current status
On 13 September 2006 the RepRap 0.2 prototype successfully printed the first part of itself which was subsequently used to replace an identical part originally created by a commercial 3D printer.
An short production run (~1 km) of 3 mm polycaprolactone filament suitable for use in the Mk II extruders used in several operational and near-operational reprap prototypes has been produced for the project. With 1km of polycaprolactone filament, the project will be able to produce larger artifacts and also test the candidate systems for days instead of for hours, as was the case with hand-produced filament. Samples of the material have been airmailed to researchers with operational Mk II extruders for confirmation of the product's suitability before major amounts are shipped and larger production runs of filament ordered.
[edit] Background
Dr. Adrian Bowyer[1], a Senior Lecturer in mechanical engineering at the University of Bath in the United Kingdom, invented and initiated RepRap, and leads the project team. The project itself uses rapid prototyping (specifically fused deposition modeling), and will make the results available under the GNU General Public License at no cost, allowing other investigators to work on the same idea and evolve it as well.
Currently commercial low-end commercial 3D prototypers cost about US$30K (Z-Corp). Prototypes made by these low-end commercial machines costing around US$2 per cubic centimetre to fabricate. The Reprap Project is on track to produce an open source 3D prototyping machine and accompanying software that costs about US$400 to build and which can fabricate objects at a cost of about US$0.02 per cubic centimeter.
[edit] Hardware development
[edit] Enabling technology
While the RepRap project was formally launched in March of 2005, the rate of hardware development exploded after the development of the Mk II fused deposition extruder by Dr. Bowyer in November 2005.[2] The Mk II represents a major advance in the state-of-the-art in that it can operate in a room temperature environment rather than a closed box heated to a temperature just short of the melting point of the polymer being extruded. This greatly reduces both the cost of producing a system and makes specifying materials used in such systems much less critical.
[edit] Full RepRap prototypes
[edit] RepRap 0.1
Originally assembled in February 2006 as of 10 May the first full prototype replicator designed by Vik Olliver in New Zealand soon began to extrude forms. By mid-July objects of up to 15 layers in depth and 50 mm in diameter were extruded by Da Witch, Vik Olliver's prototype RepRap machine.
The polymer components of the 3D positioning system and polymer extrusion head were created using a commercial FDM machine. The system was subjected to intensive shakedown exercises by Vik Olliver and Simon McAuliffe to integrate the software, firmware and hardware components.
[edit] RepRap 0.2
As of September 2006 Da Witch has been rebuilt to make it portable and a second extruder head has been added. The modified RepRap 0.2 has been dubbed Zaphod. Vik Olivier has delivered a lecture about and demonstration of Zaphod and the RepRap research programme in Vienna at the Paraflows 2006 digital arts congress there during 9 September-16 September.
[edit] Other prototypes
Other full prototypes at various stages of development are presently being built by Higgs, Sells and Bellmore. Higgs' prototype uses the RepStrap concept originally articulated by Simon McAuliffe of the RepRap team.[3] Briefly, repstrapping involves building an intermediate machine out of readily available local materials. This RepStrap is then employed to fabricate parts for a full RepRap machine.
To date virtually all RepRap prototypes have been repstrapped. Vik Olliver's prototype, which has evolved into RepRap 0.2, began largely as an assembly of Meccano parts,[4] and to give a notion of the range of materials that can be utilized the Higgs prototype was almost entirely built of poplar.
The materials chosen reflect local conditions and the skillset that the individual team member brings to the project.
[edit] Software development
Reprap has been conceived as a complete replication system rather than simply a piece of hardware. To this end the system includes CAD (computer-aided design) in the form of a 3D modeling system and CAM (computer-aided manufacturing) software and drivers that convert RepRap users' designs into a set of instructions to the RepRap hardware that turns them into physical objects.
[edit] 3D modeling (CAD)
The RepRap team has, for now, identified the open source Art of Illusion (AoI) 3D modeling system as the front end for the RepRap system. AoI is well-suited for this role both because of its power to model 3D objects and because it is written in the ubiquitous multiplatform Java programming language. The basic AoI modeling platform is being tailored to the special needs of the RepRap project via scripts.
[edit] Drivers (CAM)
Work on RepRap's CAM system also being written in Java is currently in an advanced state of development. That work is being is being pursued by both Simon McAuliffe and Adrian Bowyer and is being tested as part of the RepRap 0.1 shakedown exercises.
[edit] Materials
RepRap will have multiple extruder nozzles for applying various materials in the creation of parts. Potential materials include, but are not limited to:
[edit] Structural materials
- Thermoplastic polymer - which is what is being concentrated on currently. This is generally used to create structure. Polycaprolactone is currently the target material though tests with [PLA] will be undertaken in the next several months.
- Ceramic slurries - to create very hard and strong ceramic structure. Silicon nitride is currently being contemplated in this role, though several similar refractories are being considered.
[edit] Support structure materials
- plaster/cellulose mixes - can be washed away from the finished product with lukewarm water.
- Icing sugar - can be washed away from the finished product with lukewarm water.
- polycaprolactone/marble dust mixes - can be peeled away from the finished product.
[edit] Gasket materials
- Silicone polymer - for gaskets, seals and flexible parts.
[edit] Electroconductive materials
- Wood's metal or Field's metal - low-melting point metal alloys (lower than the melting point of the plastic) to incorporate electrical circuits into the part as it is being formed.
- Silver-filled polymers - are commonly used for repairs to circuit boards and are being contemplated for use for electrically conductive traces.
[edit] Whimsical materials
- Chocolate - chocolate has been proposed as an extruded material. This could allow the manufacture of complex 3D Easter eggs and other such items.
[edit] Controversy
Although it appears likely that RepRap will be able to autonomously construct much of its mechanical components in the near future using fairly low-level resources, it would still require an external supply of many sophisticated components such as sensors, stepper motors, cameras, or microcontrollers. Barring a major, global technological breakthrough, there are no conceivable methods for RepRap to manufacture these items out of raw resources. The project team acknowledges this and specifically excludes construction of the aforementioned items from project goals.
In light of the above, some critics argue that, by the virtue of its total dependence on man-made, externally supplied devices, RepRap cannot be called a true clanking replicator, and many of the stated goals of the project are simply unattainable - for example, the lifelike, unattended expansion of the population, or the evolution of individuals (particularly in regard to the robotic brain) is unlikely to be observed beyond the obvious and expected patterns.
The project can be defended by stating that a certain level of parasitic dependence on generic man-made components or man-initiated interaction is allowed. By that criteria, however, a robot built to carry its own blueprints, and force encountered humans to build copies to these specs at a gunpoint, may also deserve to be called a self-replicating device.
[edit] RepRap team rebuttal
The project team has always acknowledged that a certain percentage of such devices will have to be produced independently of the RepRap self-replicating process. The goal is, however, to asymptotically approach a 100% replication over a series of evolutionary generations. As one example, from the onset of the project the RepRap team has explored a variety of approaches to integrating electrically conductive media into the product. Success on this initiative, which will hardly require a "major, global technological breakthrough" will open the door to the inclusion of connective wiring, printed circuit boards and even motors in RepRapped products. Variations in the nature of the extruded, electrically conductive media could produce electrical components with different functions than pure conductive traces not unlike what was done John Sargrove's sprayed-circuit process in the 1940's.
The point about "a robot built to carry its own blueprints, and force encountered humans to build copies to these specs at gunpoint, may also deserve to be called a self-replicating device," belies a poor knowledge of biology, though it is technically correct in that such a parasitic machine could - theoretically - prosper.
However, the RepRap machine is obviously intended to be symbiotic with people, not parasitic upon them - it will be capable of making a wide variety of useful products as well as itself, which will give people a ready incentive to assist its reproduction. The analogy with the symbiosis between the flowering plants and their pollinating insects is a close one: the RepRap machine produces goods (just as plants make nectar), and so people will help it to reproduce (as insects carry pollen).
It is worth remembering that the goal of the RepRap project is not to produce a pure self-replicating device for it own sake but rather to put in the hands of individuals anywhere on the planet, for a minimal outlay of capital, a desktop manufacturing system that will enable the individual to manufacture many of the artifacts used in everyday life. The self-replicating nature of RepRap will also facilitate its viral dissemination and may well facilitate a major paradigm shift in the design in manufacture of consumer products from one of patented, factory production to open-source, personal production. Opening up product design and manufacturing capabilities to the individual should greatly reduce the cycle time for kaizen improvements to products and support a far larger diversity of niche products than the factory production run size can support.
[edit] Project members
- Sebastien Bailard, in Ontario.
- Brett Bellmore in Michigan.
- Dr. Adrian Bowyer, Senior Lecturer in the Mechanical Engineering Department University of Bath.[1]
- Michael S. Hart, creator of Project Gutenberg, in Illinois.
- Dr. Forrest Higgs, Brosis Innovations, Inc. in California.
- James Low, undergraduate in the Mechanical Engineering Department at the University at Bath.
- Simon McAuliffe in New Zealand.
- Vik Olliver, Diamond Age Solutions, Ltd. in New Zealand.[5]
- Ed Sells, postgraduate in the Mechanical Engineering Department at the University at Bath.
- Zach Smith, in the United States.
[edit] Sponsors
- Reece Arnott
- The Bath University Innovative Manufacturing Research Centre[6]
- The Engineering and Physical Sciences Research Council[7]
- The Fluorocarbon Co. Ltd.[8]
- Michael Ingram
- Lukasz Kaiser
- The Nuffield Foundation
- Carl Witty
[edit] See also
[edit] References
- ^ a b Adrian Bowyer. University of Bath. Retrieved on 2006-06-04.
- ^ Version2OfThePolymorphExtrusionHead. ReprapDocs. Retrieved on 2006-06-04.
- ^ RepStrap. ReprapDocs. Retrieved on 2006-06-04.
- ^ Olliver, Vik (27 April 2005). Construction of Rapid Prototyping Testbeds Using Meccano (PDF). Retrieved on 2006-06-04.
- ^ Diamond Age Solutions. Retrieved on 2006-06-04.
- ^ The Engineering Innovative Manufacturing Research Centre, University of Bath. Retrieved on 2006-06-04.
- ^ EPSRC Website. Retrieved on 2006-06-04.
- ^ Fluorocarbon Co. Ltd.. Retrieved on 2006-06-04.
[edit] External links
[edit] RepRap project websites
- Official RepRap site.
- RepRap Wiki site.
- RepRap object repository
- The RepRap Project, Reports and Documentation
- Construction of Rapid Prototyping Testbeds Using Meccano, Final Report, 27-April-2005, PDF, 2471 KB
- RepRap Official Blog. - Official Developers Blog
- RepRap Bootstrap Blog. - Building a bootstrap RepRap
[edit] Articles
- '3D printer to churn out copies of itself', Celeste Biever, NewScientist.com news service 18 March 2005
- 'The machine that can copy anything', Simon Hooper, CNN, Thursday, June 2, 2005
- Self Replicating Robots And The Developing World., KnowProSE.com, Sunday June 5, 2005.
