Digital control over reversibly composable units of matter

From apm
Revision as of 19:54, 8 June 2023 by Apm (Talk | contribs) (Related)

Jump to: navigation, search

Digital control over matter or digital mattercontrol for short.
Not to confuse with digital manufacturing.

This does not mean digital control over manufacturing machinery.
This does mean digital control over (recomposable) matter.

Essence being discrete pieces & pick-and-place (scale independent)

One very advanced form of digital control over matter is to have the capability
to control the placement of individual atoms FAPP fully deterministically.

But digital control over matter is also possible at the macroscale.
Prerequisite for that is that there are pre-made standard building blocks that are reversibly recomposable.
These can be pre-produced via various manufacturing methods including 3D-printing, casting, injection molding or whatever is feasible.
Note that while the machines for the preproduction can and likely will be under digital control this is not digital control over matter.
What is digital control over matter is the subsequent reversible pick and place assembly and recomposition of these pre-built parts.

Bits are structurally reversible (not necessarily energetically reversible). They can be discretely written and deleted.
Digital control over matter thus needs to have its pieces discretely addable and removable.

Actually due to the facts that …

  • physical pieces cannot be magically deleted (or spawned) "POOF" and
  • removal implies reversibe connection mechanisms

… digital control over matter implies an analogy not just to digital logic
but to reversible digital logic.
(Side-note1: Reversible logic features the potential for energetic reversibility.)
(Side-note2: Note that immutability is compatible with reversibility as there are is uncomutation in retractile cascades.)

See related page: Tracing trajectories of component in machine phase

Error prevention

Self centering structures can help funneling parts to the places where they are supposed to go.
Macroscopically without heavy vibrations there is a super exponential drop in error rate.
At some point the error rate is literally zero

At the nanoscale what represents the funneling in piezomechanosynthesis are the attractive spots on the crystal lattice.
Relevant here for the error rates is the quantity of lattice scaled stiffness.

Self-centering funnels to redirect entropy

Conceputually one can think of it like this:
A system fully in machine phase has zero entropy. Everything is known.
A placement mechanism has some placement error that "intends" to introduce some entropy into the system.
The self centering funnel or crystal lattice point attraction restores the entropy to zero.
Entropy can only grow. So where does the entropy go you ask? Sound and heat of course.
Energetically this is not relevant at macroscale the annoying clanking sound is though (clanking replicator).
At the nanoscale the energetic loss will eventually become relevant though and subject to optimization (see idea: Dissipation sharing).
What one essentially does is pumping out the entropy to the impulse space before any of it can seep into position space.

Optimization; minimization of loss of energy (dissipation)

Stiffer robotic structures and slower operation speeds give lower ingress of entropy per assembly step. Less wiggle to funnel.
This lowers dissipation and allows (but does not necessitate) for smaller funneling structures or materials with lower lattice scaled stiffness.
Related to stiffer structures and funneling the wiggle: Combining advantages of different selfassembly technologies

Error correction

Error rates in piezomechanosynthesis or bigger scale pick-and-place processes can be kept low by repeatedly testing results.
This can be as simple as by touch, not necessarily a need for cameras that wouldn't work at the nanoscale.

Repeated testing for correct assembly scales exponentially essentially allowing FAPP perfectly error free nanocomponents like crystolecules and microcomponents.
pressing down the error rate that is already low to begin with die to other means.

Of course there is radiation damage at bigger scales.
This can be taken by whole system redundancy and more advanced things like self repair
(also strongly based on the digital control of matter foundation).

What does digital mean concretely

In digital control over matter the base parts are digital in the sense that error correction applies.
But not in the sense of a single bit. Rather in the sense of a set of bits for each part.

  • Their positions are exactly known at all times (machine phase). E.g. via coordinates or connection topology.
  • Their type within a finite discrete (but potentially large) set of options is exactly known. An ID.

Link between digital mattercontrol and mechanical metamaterials

Having (due to machine phase) reversibly re-composable base is an excellent basis for metamaterials
as properties are defined by structure and the structure can be reversibly recomposed.

Relation to 3D printing

3D printing alone is not "digital control over matter"!
It is one of many pre-producing process that can make base parts for "digital control over matter".

The revolution in 3D printing is rather merely that

  • BIG: Materials needed for building the 3D-printer machines are cheap (initiated by RepRap self-replication!) aiding in local at home workshop access and a largely unnoticed Cambrian explosion of technological progress
  • BIG: Production processes are vastly simplified (no complex toolpaths and no toolchanges needed)
  • MINOR: novel complex geometries became possible (this is rarely fully exploited)
  • MINOR: additive rather than subtractive meaning net-shape no waste chips
    (many prints fail and are immediately disposed of plus plastic is much les recyclable than metal of so what)
  • DETRIMENTAL: In place printing hype. Saving (manual) assembly effort by printing everything in one piece.
    In place printing of big monolithic parts deeply goes against "digital control over matter". That beside adding additional design constraints like e.g.
    more overhang limitation, more layer strenght direction constraints, worse backlash in rolling or sliding interfaces, ...

Related


  • Common misconceptions – "Thermodynamics prevents one from having every atom at the place we want it" - wrong for practical scales

  • Error correction
  •  ?? parity bits and more complicated redundancy schemes ??
  • Shannon information theory


  • Digital manufacturing – usually just referring to digital control of machines that have analog control over matter