Design levels for (advanced) APM systems provide a general blueprint for relevant software package architecture.
They are somewhat similar to the assembly levels which provide a general blueprint for a concrete physical layout.
Back: technology level III
- 1 Design levels
- 2 Related
- 3 External references
This topic is basically about modeling methods and modeling software that matches the assembly levels plus software for overall system level management (design,simulation,...).
Tooltip level design
Here the goal is to find reliable and closed loop tooltip systems that support increasingly many moieties / elements.
Highly accurate quantum mechanical simulations are necessary but the systems are small enough to be handlable well by currents computing systems. Some results can already be verified experimentally.
[Tooltip cycle; DC10c;...] tooltip chemistry
- NanoHive@Home’s Published Results: Analysis Of Diamondoid Mechanosynthesis Tooltip Pathologies Generated Via A Distributed Computing Approach
- DC10c: Design and Analysis of a Molecular Tool for Carbon Transfer in Mechanosynthesis tip based nanofabrication
Simple force field approximations are suitable for all but the core mechanosynthesis processes of advanced APM. For these more accurate simulations (Ab initio quantum chemistry methods) are necessary Gaussian can be used to analyze specific problems of tooltip chemistry
A list of quantum chemistry computer programs.
Atomistic mechanic level design
Engineering of DMEs is a completely new field. The software tool Nanoengineer-1   was developed to make this area more accessible. It's development is currently idle (2014). It was extended to support more near term structural DNA nanotechnology. There is also the cadnano for structural DNA nanotechnology.
Nanosystems 14.6.3. Automated generation of synthesis and assembly procedures - a. Retrosynthetic analysis of chemical syntheses.
Lower bulk limit design
Bigger structures where atomic detail may matter less or which are simply not simulatable yet because of limited computation power may be designed with conventional methods of solid modelling.
A few issues have to be thought about though:
- Since we operate on the lowermost size level there needs to be set a minimum wall thickness that must not be deceeded
- surfaces should be kept parallel to the main crystallographic faces such that they will not create random steps when auto-filled with virtual atoms.
[todo add links to demo collection]
Nanosystems section 9.3.2 and 9.3.3 (bounded continuum)
Somewhat related here is the degree of applicability of macro 3D printing for nanomachine prototyping.
Programmatical CAD Software Tools for volumetric shape description are needed: (Makro and Meso)
- OpenSCAD - limited language capability + quick preview
- ImplicitCAD - no quick preview yet - native language interface could be better + high language capability
- miniSageCAD - very slow + nice interface + educative experiment
- Proposal: creation of an Solid Body Construction Tool SoBoCoTo that combines the best aspects of the above and extended by the capability of intelligent atomic tessellation. (Binding to DME design tools.)
Purely graphical 3D-modelling tools seem less suitable (see WYSIWYG vs WYSIWYAF).
One of the more severe problems that came from purely graphical 3D-modelling tools is the concept of grouping of geometries.
Nanosystems 14.6.4. Shape description languages and part arrays
System level design
Main topics are:
- organisation of diamondoid metamaterials
- nanofactory system design
Diamondoid metamaterials (and more heterogenous microcomponent subsystems) are of high importance since they form the basis for all advanced AP products an applications.
Examples for what metameterials we might want to design are:
- elasticity emulation
- infinitesimal gear bearings
- legged block mobility for e.g. live self repair
- DME recycling
It is desirable to organize these metamaterials in microcomponents which are designed such that they allow adjustable inter-mixture of standalone subsystems. Examples for intermixture of sub-systems:
- infinitesimal bearings + chemomechanical converters + energy storage cells = chemical interfacial drive
- infinitesimal bearings + electromechanical converters + electric distribution system = electrical interfacial drive (non-mechanical)
- interfacial drive + self repairing systems + hirachical heterogenous comutation systems = very advanced APM product
Note that this functional composition has it's limits. Some especially fancy functions might exclude a whole set of others.
In AP manufacturing systems system level design determines the mapping of the abstract assembly levels into a concrete three dimensional layout of a nanofactory.
Today (2013) it is rather difficult to do work on this area. Lots of questions need to be answered.
(yet speculative) advanced metamaterials.
The main topics can each be further subdevided into:
- three dimensional placement of huge amounts of standard components
- topological interconnections
- temporal organisation in a dynamic setting
- IO logistics of all the media (materials, information, energy,...) to handle
- emulation of physical (especially mechanical) properties
A big problem at this design level is that the sizes of the diverse functional components and the locations of their connection points are yet unknown.
Helpful may be a software capable of crystallographic space subdivision (space groups) and piece-wise connection of different crystal structures with compatible 2D cross-sections (plane groups). Scale invariant symmetries (fractal symmetries) are also of high relevance especially in redundancy design that is e.g. needed in artificial motor-muscles design.
- DME system simulation
- 3D Layout generation from schematics
Nanosystems 14.6.5. Compilers - b. Design compilers -- detailed info
- Production control .. sub product part paths
Nanosystems 14.6.5. Compilers - a. Assembly compilers
Nanosystems 14.6.3. Automated generation of synthesis and assembly procedures - b. Hierarchical decomposition of larger structures. [move reference?]
- Auto Assembly Code Generation - CNC path generation - slicing - gcode - retro-synthetic analysis
[Todo: improve section]
Software plumbing and "Frankenstein systems"
Today (2017) the software solutions for subsystems that are being tied together often where originally never built to work together because historic reasons like:
- Before the individual solutions where created the combined problem wasn't even known to exist.
- In case the combined problem was known to exist from the start it might just have been to hard or unprofitable to tackle as a whole for the later creator of some individual sub-solution(s).
In CAD & BIM the most used software packages are often proprietary and closed source. This allows little to no internal restructuring by third parties which could mend the software plumbing nightmare at least a little.
When having the opportunity for doing system design from scratch (which may or may not be the case for the development of gem-gum factories) one very much wants to avoid a repetition of the creation of such "Frankenstein systems".
Beside the right preconditions (TODO: elaborate here) this will require fundamental paradigm shifts in programming. Specifically: A switch will be necessary to new programming languages that have higher compositional power than languages that follow the object oriented paradigm.
Examples for some "Frankenstein systems" today (state 2017):
- proprietary BIM software conglomerates (to investigate)
- open source CAD software conglomerates (e.g. FreeCAD ...)
- ... there are most certainly many more ...
BIM (building information management)
While the modeling of the whole life cycle of buildings (BIM) on the outside is a pretty different problem compared to the design of gem-gum factories there are several parallels. Supply of energy, raw materials and information, transport logistics (elevators), waste removal, ... Just as with the the parallels to biological systems the metabolism analogies are only superficial.