Scenario: Micro Machines

On route to Nanotechnology we can anticipate Micro Machines, tiny precision engineered devices which are nevertheless very low cost. The low cost will arise because:
 * The material costs are low, due to the small quantities used.
 * The labour costs in Manufacturing are low because of automated lights-out manufacture.

First Embodiment
The design shown below is for a full sized robot grip. Similar principles could be used at the 1cm scale.




 * Progressive Miniaturization of robot Hands And Grips leads to a precision robot arm and hand that is driven by a desktop computer. The hand is about 1cm across.
 * To achieve the design without a huge investment, the design uses conventional parts as far as possible. The precision engineering is confined to the hand itself.
 * The 'tendons' that articulate this hand return via wire cables. Off the shelf brake cables are used, attached at the 'big end' to normal sized solenoid actuators, as used in factory automation.
 * The first hand is a simplified six finger robot grip, with the fingers arranged in a circle, each finger identical in construction. This simple design simplifies the challenge of self manufacture. Fewer kinds of new part are needed.

On June the 1st 2010 the first successful replication of the grip and arm is achieved entirely under computer control. A complete hand and arm takes 38 hours to manufacture, using just one hand.

By August 30th 2010 when the team go public they have around 200 correctly manufactured precision 'hands'. They have 12 complete rigs up and running, including actuators and driving computers.

Although the hands are very simple, these 12 rigs are working round the clock, manufacturing more hands, and the first products.

First Products
The image below is an illustration of a tiny proof-of-concept toy produced using conventional manufacturing technology. A strip of memory metal runs down the centre of the 'snake body'. When current is applied the eyes flash (there is an LED behind the eye lenses) and the snake slowly resumes it's original shape.



As well as 'toys' such as the one shown, in 2011 micro-manufacturing is producing products such as the following, at very low cost:


 * Surgical instrument for key hole surgery. In particular the precision devices for stabilising manipulator arms against vibration.
 * Optics for key hole surgery - the rigs are being used to assemble and calibrate the delicate lenses and fibre optics used to view surgical work.
 * Air-turbine 'motors' used in dentist's drills.

Patents?
Readers of the announcement in the Economist are stunned to see that the design files and software for manufacturing the precision hands is being LGPL'ed. It turns out the research program was entirely funded by a charity set up to bring the best parts of the future to realisation sooner.

Following steps
By August 2011 numerous tools have been miniaturized using the robot hands by both hobbyist enthusiasts and commercial enterprises. There are precision micro tools available for
 * Casting metal
 * Bending sheet metal
 * Stones for grinding and polishing.
 * Oxy-acetylene torches
 * Knives, pincers, mallets, scissors, pop-riveters, screwdrivers.
 * Miniature lathes

Construction times range from around an hour for an oxyacetylene torch to three days for the miniature lathe. The cost of these precision parts is extremely low due to the method of manufacture.

Effect on General Manufacturing
Engineers are now falling over themselves to find ways to make smaller consumer goods, since the cost of precision engineering is now gone. Smaller consumer goods make sound economic sense as that minimises material costs.


 * Kettles are a thing of the past. A flash-heater in the tap can give you water at exactly the temperature you want.
 * Radiators now consist of a few heating fins around the pipe - which carries super heated steam, not water. Rooms can be heated in moments.
 * Piezo electric fans take the place of rotating blade fans.
 * Lighting is replaced by LED arrays.
 * Self repairing, smart 'garbage disposal units' in sewage pipes are able to guarantee not to block or jam, allowing faster-flowing narrower-bore pipes to be used for sewage.

Custom manufacturing becomes the norm, when so much is manufactured under computer control.

Remaining Challenges
Moving from tethered micro machines which derive their power and computation from a central server doesn't happen all in one step.


 * One of the early steps is 'relay' machines, which manage a small team of micro machines. The relay splits a turbine air supply, oxyacetylene, electrical power between its team and adjusts voltage, pressure, oxyacetylene blend, as well as providing quick response to failures.
 * 'Smarts' gradually move down into the relays. The relays already contain small storage units to buffer electrical and pressure surges. These allow for limited detached action. Radio communication with the server allows for physical separation whilst still under central control.

These semi-independent micro machines still need to hook up to pipelines for power and materials. However the model is now more like a refuelling stop than being continuously connected to base.

Design at the 1cm scale
An interview for Technology Magazine with one of the project team

Q: Can a drill wielded by a micro machine achieve the same results as a full sized industrial drill?


 * Yes. It can do better. It can cut a hole using less power and generating less heat. The industrial drill completely pulverises the waste material (concrete) or creates many pieces of swarf (metal). The micro machine drill cuts a core out, scoring an ever deepening circle on metal, using a pick to remove tiny chunks from concrete. The principles do differ. The circle-scoring drill bit is constantly alternated one way and then the other to keep its edge sharp. It self-sharpens against the hole that is being cut. The drills typically operate with much less inward force than a conventional drill as a surface to push against may be a great distance away.

Q: How does vibration affect the precision achievable at this scale


 * It is a huge factor. Micro machines overcome it through computer directed feedback. Rather than building rigidity at this scale, you adapt to the movement. It will be much worse as we scale down.

Q: Which leads me to the next question, what are the prospects for scaling down further.


 * Ah well now. That's the million dollar question.