OS Fab Lab Proposal

See corresponding blog post - http://openfarmtech.org/weblog/?p=1254

Introduction

Fabulous Friends Smari, Kyrah, Olle, Erik, Edmund, Henri, and James X. Jones:

I would like to follow up on the concept of the OS Fab Lab. You have expressed interest in it, and I'd like to take a moment to see how you might want to participate. After you read this, please let me know your level of interest. In particular:

1. Who else should we recruit for the organizational team?
2. Who and how to contact others to find more collaborators?
3. How much time could you devote to this?
4. Can you serve on the organizational team?
5. Can you set up a website for this?
6. How can we enlist communicators, PR people?
7. Besides the crowds, who can we tap for funding?
8. Etc, etc.

The basic concept is that a set of advanced tools like the MIT Fab Lab – deployed at 1/10^th the cost and ten times the functionality – could be a great boon to liberating Industry 2.0. This means the simple concept of: global design collaboration towards local fabrication. Transformative economic potential is huge, as personal fabrication takes over the work of large factories and slave labor. This is around the corner in the 21^st century – if we want to make it happen.

The facts are that we are likely to get 10-fold cost reduction over the standard Fab Lab price of $100k. We've already proven 10-fold cost reduction with The Liberator (CEB press), and the same could be done with the MIT Fab Lab – which would include laser cutting at 100W, an induction furnace (20kW, 300 lb metal melt per hour), metal casting and rolling, plus heavy machining (mill, drill, press with 20 hp interchangeable hydraulic motor), metal press (also shears and punches holes). To top it off, let's add a robotic arm. Also, include technologically-recursive development of OS high power supplies for welder, induction furnace, and plasma cutter. Here's a summary of the tools included in the OSFL, their description, and their Bill of Materials (not including labor). This strategy involves a hydraulic power source for high torque and power applications, and 20kW of shop power.

These tools can get you from melting scrap steel, casting and rolling for parts, cutting, machining, lasing, routing, 3D printing, metal working, and circuit fabrication. This gets to all electromechanical devices known to humans, and can get us to just about all technology up to about 1980 or so. Combined with Arduino, it gives you automated control of CNC fabrication. The goal here is all the capacity of the MIT Fab Lab - plus much more: metal melting from scrap, heavy duty fabrication, and robotics. This is just a suggested approch. If OS, I don't see trouble getting to the prices mentioned.

I am hoping we could fund it by crowdsourcing from those who gain access to the designs. We can motivate donations by offering use of the developed equipment for making copies of the machines at ridiculously low costs. I would support this myself if I were to gain access to fabrication facilities and training.

Toolset

TOOL	DESCRIPTION AND DEPLOYMENT STRATEGY	WT	COST ($)
3D printer	RepRap printer in plastic; plans available; self-replicating for all of its joints and plastic parts; head interchangeable for a small router for circuit fabrication	1 m3, 50 lb	300
CNC torch table, router	RepTab prototype available – results indicate that it can be scaled successfully to router applications with 280 lbs of moving torque by using 4 stepper motors on the x and y axis; high power router can use hydraulic motor (3000 RPM, 10-20 hp); self-replicating for all structural parts; use RepRap motor drivers and controls	300 lb	1500 for 280 lb torque version
Drill-mill-lathe	Interchangeable hydraulic motor (0-650 rpm, 20 hp, $250); off-shelf chuck ($150) and off-shelf x-y table ($200); off-shelf spindle and collet ($200); true drill press – hydraulic cylinder ($100) moves spindle up and down; large motor can handle drilling up to 1.5”; mill funcionality via x-y table; CNC drive can be retrofitted onto x-y table; use RepTab motor drivers and controls	500 lb (welding table serves as base for added wt)	1100 plus CNC
Welding table	Sheet of 1/2”, 4x8' mild steel	640 lb	300
MIG Welder	This involves opensourcing the power supply, and using a commercial gun/wire feeder	200 lb	400
Plasma Cutter	This involves opensourcing the power supply, and using a commercial gun	50 lb	300
Induction Furnace	20 kW, water-cooled coils; involves opensourcing the power supply; the rest is a melting container, pouring mechanism, and insulation	1000 lb	2000
Metal casting, rolling	Casting involves simple molds; rolling involves high power rollers, using 20 hp hydraulic motors above; start by rolling bars from hot billet	1000 lb	2000
CNC laser cutter	Utilize existing x-y table, and use a stationary laser; build a laser from a $500 CO2 laser engraver tube of 80W; sufficient to cut ¼ inch wood and acrylic, and thin metal – perhaps up to 1/8” in a large number of passes; larger laser may be built from scratch by creating a tube at later phases of recursion	300 lb	2000
Metal press, shear, and hole puncher	Up to 1” holes in 1” metal; shears 3” wide 1” metal; relies on a large cylinder ($260)	1000 lb	1000
Cold cut metal saw	Uses existing hydraulic motor	100 lb	100 plus blade
oscilloscope	Develop OS computer oscilloscope	10 lb	50
Robotic arm	6 degree of freedom robotic arm for welding or other applications; use hydrauilic motors ($900) with encoders	500 lb	2000
Spectroscope	Microwave/X-Ray spectroscopy tool	?	?
TOTAL		5500 lb	$13,100

One note – the laser seems like the most difficult item here. However, given access to laser tubes from laser engravers, this is not that difficult. Many amateurs even blow glass to make their own tubes. On the robotic arm – I haven't searched yet – but there must be tons of people knowledgeable and interested in it.

Other model tool sets:

• ultimate tool buying guide
• electronics workbench
• fablab inventory
• techshop inventory
• emachineshop inventory

Proposal Outline

Here is an Outline of a more complete proposal which could deploy this campaign.

1. Introduction to the Open Source Fab Lab (OSFL)
   1. Problem Statement
   2. OSFL Collaboration Concept
      1. Downloading open source hardware over the web
      2. Collaborating with SKDB (it's basically this project except Marcin doesn't know about it)
   3. Industry 2.0 Problem Statement
      1. Design repositories – SKDB, Smari's work, Thingiverse, Google 3D warehouse,
      2. Local fabrication – Fab USA, Fab (Your Country) program
   4. Universal Constructors
      1. Examples – RepRap, CubeSpawn, OSE work, Erector Set, Box Beam, Makerbot, Fab@Home, CandyFab, P3P, Contraptor, MIT Fab Lab, Bildr
   5. Design Specifications
      1. Low cost, multipurpose, robust, OSE spec
      2. Scope of production covered
      3. Size, cost, and weight of equipment set
      4. Open Standards
   6. Applications of OSFL
2. Required Functions
   1. Fabrication functions
   2. Scope of production covered
3. Proposed Components and Their Functions
   1. Fab Lab template
      1. Scope of application
      2. Missing functions
   2. OS Fab Lab template
   3. Technological recursion level
      1. Off-shelf components
      2. OS power electronics
4. Proposed implementation
   1. Utilization of hybrid electric and hydraulic drive
   2. Power sources – grid and generator
   3. Bill of materials for OSFL
5. Bill of Materials for the OSFL
6. Proposed collaboration
   1. Hackerspaces
   2. Fab Labs
   3. Economic development organizations (1^st and 3^rd world)
   4. Other libre and open source, universal constructor programs
7. Funding, PR, Resource Development
   1. Website and funding basket
   2. Allocation procedure
   3. Resource development
   4. Required PR materials
   5. Donation strategy - based on use of FeF facilities for low-cost replication
8. Inventory of Existing Hackerspace Equipment
   1. FeF collaboratory equipment inventory
   2. Available Fab Lab facilities and equipment
   3. Hackerspaces
9. Available design and engineering resources
   1. Encyclopedias - Fabripedia, mechanisms, industrial processes, chemical processes, food processing, agricultural equipment, equipment design, how things work (collections of mechanisms)
   2. Engineering and formula handbooks in all fields
   3. Available free software
   4. List and Evaluation of collaborative engineering platforms
   5. OSE Dedicated Project Visits
      1. Nature of visits
      2. Infrastructure development
10. Organizational Team
11. Summary

skdb

RepLab will need a robust way to represent and track designs, builds, edits, forks and all the other information management challenges of an open-source distributed hardware project.

Ben Lipkowitz and Bryan Bishop, among others, have put significant work into a system called SKDB. SKDB is a method for sharing hardware over the internet. "Hardware" means not just designs for circuit boards, but also biological constructs, scientific instruments, machine tools, nuts and bolts, raw materials, and how to make them. SKDB simplifies the process of searching for free designs, comparing part compatibility, and building lists of materials and components and where to get them. You could even say SKDB is "apt-get but for real stuff".

In SKDB, hardware is organized into packages. Packages are a standard and consistent way for programs to find data. Packages may contain CAD files, CAM parameters, computer-readable descriptions of product specifications, product-specific code, and bill of materials. For each part in a package there are a number of interface definitions, which describe how the part can connect with other parts, even parts from other packages. Each package also lists dependencies which have to be bought or built in order to successfully carry out a project. For example a drill press is required to make holes with a certain level of accuracy. SKDB downloads all of the dependencies automatically and compares them to your existing inventory, and generates instructions for your CNC machinery if you have any.

• git repository: [[1]]
• [[2]] (web view) http://adl.serveftp.org/skdb/
• presentation: http://adl.serveftp.org/lab/presentations/updates-from-austin.pdf
• email: openmanufacturing@googlegroups.com
• IRC: #hplusroadmap on irc.freenode.net
• main wiki: http://adl.serveftp.org/dokuwiki/skdb

SKDB is new but extremely promising. Sites like Thingiverse are a great beginning, but 90% of the designs are for a RepStrap or a laser cutter, which both have the virtue of being reasonably 'plug and play'. With multiple tools displaying dependencies and requiring non-trivial human interaction, something more robust is clearly needed.

RepLab needs a representation format, and SKDB needs a meaningful test case to develop its functionality. At a minimum, all data for the projects should be described in YAML wherever possible.

Team

NYC Resistor response

• contact@nycresistor.com
• date Wed, Nov 18, 2009 at 10:45 PM

We are not currently working on an opensource laser cutter. Some of our members are active with open source 3D printing, namely the RepRap and the Makerbot.

I do know that hacklab.to rebuilt a broken epilog and then taught it to sing (http://hacklab.to/archives/another-musical-variation/). They might have a couple of folks interested in opensourcing a laser cutter design.

Best, Max

Comments

Comment here. Click edit on top.

• Q:What happened to using the Multimachine design?
• A:What design? Those are concepts, and valuable ones, bu not particular implementations. If you can point me to an actual, doucumented design that is in any way replicable, we'll be glad to assess it.

• Q: Have you looked at the Gingery books? The lathe and shaper can replace most of a machine shop either with or without CNC capabilities. They are small machines, but sufficient for lots of needs, including bootstrapping.

Have you decided on a target skill level? RepRap produces snap together objects from CAD drawings, but most of the machines you're talking about require skill/training to setup and run. That's fine but it doesn't change the manufacturing paradigm. We need to be clear about what we are trying to achieve. How about this: The RepLab will be a facility capable of manufacturing 90% of it's own parts when directed by a skilled user. It will be able to manufacture 80% of items used on a large farmstead when used by a patron who has only minimal skills or receiving brief training and moderate assistance. Such skills might include: Soldering, mechanical assembly, 3D CAD modeling, measuring, aligning and clamping.

Also I wonder about the need for an induction furnace; aluminum/bronze foundry practice is pretty Open and incredibly versatile. I would focus more on automating safety than finding new ways to generate heat. An OS furnace which could pour without removing the crucible, and into a locked-in-place mold would make casting much more attractive to the general public as well as vastly safer. I am also not clear on the reasons for the rolling mill, oscilloscope, robotic arm and spectroscope.

Laser cutter

There are instructions for a homemade laser cutter here.

Open-source oscilloscope

http://www.elechouse.com/elechouse/index.php?main_page=product_info&products_id=396