Robinson's
Difference Engine #2
The author's
model of Babbage's Difference Engine #2 from standard
Meccano parts. This model was exhibited while it still a work in progress,
at the 2005
Vintage
Computer Festival
hosted at the
Computer
History Museum in
Mountain
View, California. The pictures below are from that event. It was
awarded 2nd place in the class "
Re-creation,
Emulation, or Contemporary
Enhancement", the special award "Best technology, Non-electronic", and
by popular vote of the attending public, received the "People's Choice
Award". Only two orders of differences were operational, so it was
demonstrated calculating a simple table of squares.
The completed machine was exhibited at the Maker Faire ,
April 22-23, 2006 in San Mateo, California, where it attracted constant
attention.
This model follows fairly closely Babbage's original
design from 1848, though the constraints of using only standard Meccano
parts
inevitably mean some aspects of the operation are somewhat different.
The model can handle decimal numbers with up to twelve digits,
and
up to four orders of differences. Babbage's original design called for
31 digit numbers and seven orders of differences. It was not actually
constructed until 1991 when the Science Museum in London built it to
mark the bi-centenary of Babbage's birth.
Unlike Difference Engine #1, in this design Babbage made a clear
distinction between control and data. This results in a dramatic
simplification of the per digit mechanism compared to the first
difference engine. Scaling the machine to larger numbers simply
involves extending the columns vertically. Adding more orders of
differences extends the machine horizontally, but neither entails any
significant change to the control. The model
calculates reliably, producing a result about every 7 seconds -
somewhat
faster than they can be read off and written down - and about the same
as the "original" at the Science Museum. I have no doubt that if the
Meccano of the 1920's had existed 100 years earlier, Babbage would have
been entirely successful in his quest. It may be amusing one day to attach a Meccano steam engine
to
drive the mechanism and therefore realize "computing by steam".
The picture
below shows the view from the rear with the control section at the left.
The calculating part
follows very closely Babbage's original drawings. The control section
however does deviate somewhat from the prototype. Babbage was able to
make cams of arbitrary profile to control the rods and levers that
distribute the motions through the machine. Meccano does not offer the
possibility of arbitrary cam profiles, nor is it strong enough the
handle the large forces that would be involved. Instead, the model uses
a stack of built up cams to control a bank of intermittent motion
mechanisms which transfer power directly from the source to the
actuating levers.
A short
video of the machine operating is available here
(23MB AVI, requires free DivX codec). The same sequence in four
separate
Quicktime
files (bigger download, but works better on slower CPUs) is here:
Part 1
front (23MB MOV, Quicktime)
Part
2 front, closeup (23MB MOV, Quicktime)
Part
3 rear (16MB MOV, Quicktime)
Part
4 control (17MB MOV, Quicktime)
If you have a very fast
connection and a fast CPU, a short high resolution video can be
downloaded
here (240MB, MPEG2).
Last modified: 7 May 2006
If you experience any problems with this site, please
contact the webmaster
©
2005-2006
Tim Robinson