To LUGNET HomepageTo LUGNET News HomepageTo LUGNET Guide Homepage
 Help on Searching
 
Post new message to lugnet.roboticsOpen lugnet.robotics in your NNTP NewsreaderTo LUGNET News Traffic PageSign In (Members)
 Robotics / 20982
20981  |  20983
Subject: 
Mechanical Memory for Computing
Newsgroups: 
lugnet.technic, lugnet.robotics, lugnet.build, lugnet.announce.moc
Followup-To: 
lugnet.technic, lugnet.robotics, lugnet.build
Date: 
Wed, 25 Jun 2003 03:06:30 GMT
Viewed: 
233 times
  
   Inspired by the recent pneumatic computing thread I have created some
LEGO mechanical memory.  It is an adaptation of a possible design for molecular
level RAM.

   The folder is at http://www.brickshelf.com/cgi-bin/gallery.cgi?f=47549 with
details below.

   In this overall view of it
http://www.brickshelf.com/cgi-bin/gallery.cgi?i=436795 the sliding black #8
axles encode the individual bits by produding or not through the green technic
beams.

   The red #12 axle functions as a write-protect mechanism.  In the position
shown, the black axles are prevented from sliding and the byte is
write-protected.

   In this second overhead view
http://www.brickshelf.com/cgi-bin/gallery.cgi?i=436799 the red write protect rod
has been moved to its other position.  Note that the light gray pieces connected
to the red rod no longer obstruct the light grey pieces on the black sliders.
The byte is no longer write protected.

   The 4-bit byte is a function of #12 axles being the longest available, but
the design is extendable to link multiple bytes together into longer length
words.

   The white #4 axle running parallel to the red axle
http://www.brickshelf.com/cgi-bin/gallery.cgi?i=436803 serves to insure that the
light grey connectors on the red axle stay in the proper orientation by
preventing the red axle from rotating.

   In these partial dis-assemblies
http://www.brickshelf.com/cgi-bin/gallery.cgi?i=436809
http://www.brickshelf.com/cgi-bin/gallery.cgi?i=436811 the black bit-slider rods
and their guiding technic beams have been removed.  The half-stud offsets align
the holes in the technic beams properly.  The white bricks insure that the bit
slider rods have no opportunity to rotate.  In these views the red rod is in the
write-protect position.

   An overhead of the same dis-assembly
http://www.brickshelf.com/cgi-bin/gallery.cgi?i=436813 but with the red rod in
the write-allowed position.

   More dis-assembly http://www.brickshelf.com/cgi-bin/gallery.cgi?i=436824
better shows the construction of the write-protect mechanism.  The two 1x2 black
technic beams have 2 holes each.  The long black friction pin holding the yellow
pieces together gives a possible attachment point for an external means of
moving the write-protect rod.  To link bytes together into multi-byte words
requires removing the yellow connectors and white rod and using and axle
extender to join the bare red axle to the red axle of the adjoining byte.  The
adjoining byte loses the 1x2 black technic beam with 2 holes in order to
accomodate the axle extender.  The far left byte retains the yellow connector
assembly and moves the write-protect rod for the entire word.

   Finally, the byte is actually oriented this way
http://www.brickshelf.com/cgi-bin/gallery.cgi?i=436826 in 3D space, with the
black bit-sliders vertical and the green technic beams at the bottom.  The
reason for this is so that the byte can be reset
http://www.brickshelf.com/cgi-bin/gallery.cgi?i=436829 via gravity assist when
the write protect is turned off.  If the byte were remain in the horizontal
position it would require the assistance of rubber bands.

   The molecular level RAM design included multiple bytes stacked up and a
method to select which one was being read/written, but it involves additional
structure containing flexible rods for which there is no suitable LEGO
counterpart.  I am working on alternate methods, but I am not hopeful at this
point in time.

   Please let me know what you think.

Brian



Message has 3 Replies:
  Re: Mechanical Memory for Computing
 
Brilliant! Positively brilliant! >> Mark (21 years ago, 25-Jun-03, to lugnet.technic, lugnet.robotics, lugnet.build)
  Re: Mechanical Memory for Computing
 
(...) <SNIP> (...) Do you happen to have any links to the molecular RAM design you are referencing? It would be very cool to compare and contrast your excellent model with the "real world" counterpart. Jona (21 years ago, 25-Jun-03, to lugnet.technic, lugnet.robotics, lugnet.build)
  Re: Mechanical Memory for Computing
 
Very cool design. (...) Question: Why are you calling 4 bits a byte? 4 bits is a nibble, while 8 bits would be a byte. Obviously the design can be extended to 8 bits like you mentioned. They you would have your byte. Michael (21 years ago, 26-Jun-03, to lugnet.technic, lugnet.robotics, lugnet.build)

7 Messages in This Thread:



Entire Thread on One Page:
Nested:  All | Brief | Compact | Dots
Linear:  All | Brief | Compact
    

Custom Search

©2005 LUGNET. All rights reserved. - hosted by steinbruch.info GbR