rescue files from svn.openmoko.org

old-boom/CHARACTERISTICS

@@ -0,0 +1,140 @@
+*** This is an older draft of the concept - differs sometimes from the way
+    things are done in gta02-core. ***
+
+
+BOM matching
+============
+
+BOMs are matched with inventories in the following way:
+
+- a .lst file with the BOM is generated by KiCad
+
+- using a ruleset, component characteristics are translated to a
+  canonical format and default values may be used for unspecified
+  characteristics
+
+- part catalogs are searched for matches with the canonical component
+  descriptions. This yields a list of supplier-specific part numbers
+  for each component.
+
+  Parts can be characterized by either specifying their properties or
+  by equating them to another part. E.g., a Digi-Key part may be
+  defined as an NXP part which in turn is equivalent to a TI part.
+
+- this list is then matched against inventories, using a suitable
+  optimization strategy (e.g., prioritize inventories and try to
+  pick as many suitable components as possible higher priority ones
+  before moving to lower priority ones)
+
+  E.g., local stock could be the first-level inventory, followed by
+  more distant warehouses, followed by distributors, followed by
+  manufacturers.
+
+  Inventories could also include pricing information.
+
+- TBD: it would be good if parameters gathered in the matching process
+  could be fed back into KiCad (as some sort of annotations, similar
+  to the expanded view of schematic symbols), such that under-specifed
+  parts yielding mismatches can be spotted by manual review.
+
+
+Catalog
+=======
+
+A catalog contains part characteristics and the reference number(s)
+assigned to them.
+
+
+Basic syntax
+------------
+
+Catalog entries consist of "words" in the sense that each word does
+not contain any whitespace and words are separated from each other by
+whitespace. Whitespace can be included in a word if it is enclosed in
+double quotes.
+
+Each entry begins in the first column of a line. If an entry needs
+more than one line, the words on the continuation line(s) must be
+indented by whitespace.
+
+Trailing whitespace is ignored, and so are comments beginning with a
+hash mark. Blank lines end any entry and are also ignored.
+
+Each catalog entry begins with the part number followed by a part type
+designator. 
+
+Characteristics have the form <field>=<value>, where
+the fields follow the pattern outlined below. The value is some
+description of the value of that characteristic, typically a number
+and a unit (e.g., 4.7uF) or a name (e.g., X5R).
+
+Numbers use a decimal point where necessary. Mantissas are normalized
+such that they fall into the range 1 <= n < 1000. E.g., instead of
+0.1uF, write 100nF. There is no space between number and unit. The
+Omega of Ohm is written as "R".
+
+
+Fields
+------
+
+Each 
+
+General fields
+- - - - - - -
+
+FP	Footprint
+H	Height (overrides any height implied by footprint)
+TOL	Tolerance, with percent sign. Split tolerances are indicated as n/m%
+DSC	Free-format description
+
+
+Resistors
+- - - - -
+
+RES	Part type designator
+R	Resistance, with unit
+P	Maximum power dissipation
+V	Maximum volatage
+
+
+Capacitors
+- - - - -
+
+CAP	Part type designator
+C	Capacitance, with unit
+M	Material, e.g., TANT, NP0, X5R, etc.
+V	Maximum voltage
+ESR	ESR, with unit
+
+
+Inductors
+-- - - -
+
+Diodes
+- - -
+
+DIODE	Regular diode
+STKY	Schottky diode
+
+Vf	Maximum forward voltage
+Vr	Maximum reverse voltage
+If	Maximum forward current
+Ir	Maximum reverse current
+C	Capacitance
+
+LED	Ligh-emitting diode
+
+COL	Color, multiple colors are separated by /, e.g., blue/red
+ARRAY	If multiple diodes form an array, this parameter describes
+	its structure: CA = common anode, CC = common cathode,
+	SEQ = tap-A-C-tap-A-C-tap sequence
+
+ZENER	Zener diode
+
+Vz	Zener voltage
+
+TVS	Transient voltage suppressor
+
+Vac	Working voltage, AC
+Vdc	Working voltage, DC
+E	Energy