Connect with us

Article on analog computer

Discussion in 'Electronic Design' started by Ira Rubinson, Aug 30, 2003.

Scroll to continue with content
  1. Ira Rubinson

    Ira Rubinson Guest

    I'm looking for an article on analog computers.
    I want examples of problems that are more suited to analog than digital

    Thanks -Ira
  2. Robert Baer

    Robert Baer Guest

    One classical problem is to set up a sine wave generator at a given
    frequency, from the well-known derivitave equation x=-d(dx) or x = -x
    doubledot (different notation not easily done in ASCII).

    One i set-up in SLAC on a Systron Donner analog computer was for beam
    dynamics, for focusing the beam thru quadropole magnets: where one would
    focus in X and defocus in Y, and another would focus in Y and defocus in
    How many are needed, what spacing, what strengths for net focus at a
    given final length.
    This problem required relays (or the equivalent) to switch in and out
    the quadropole magnet parameters at the appropiate times (time being the
    analog of distance from source).
    And the integrators had to be "tuned" for low drift, and line spikes
    had to be scrupously filtered out.
    Turns out line spikes at a regular, repetitive rate were injecting the
    equivalent of a bend magnet.
    They came from the campus accelerator a few miles away - from there,
    way up in the hills to a main PG&E power feed, and back down to SLAC to
    where the analog computer was (estimated trip distance about 5 miles).
    Ordinary line filters did not help; had to design some and "split" the
    power into the S-D computer to two feeds and use my filters on each feed
    going in.
  3. Check Electronic Design magazine at under Bob Pease (or Bob's
    mailbox or Pease Porridge). He had a series of articles on analog
    computing. I think they maintain an archive of his articles.

    Steve B.
  4. James Meyer

    James Meyer Guest

    There are no problems that today are more suited to analog than digital

    In the past, analog was either the only way or the cheapest way to
    compute. Present technology has changed all of that.

  5. Especially when someone is likely to suggest implementing an analog computer
    with a PIC and some AD and DACs (or maybe the PIC has them onboard?).

  6. Leon Heller

    Leon Heller Guest

    About 40 years ago, the English Electric Co. Ltd. used a LACE analogue
    computer (I subsequently worked on it ) at their Kidsgrove plant to find
    out out why motors were burning out on the English Electric Deltic
    diesel-electric locomotive. This type of problem wasn't suitable for the
    digital computers available at the time. IIRC the cause of the problem was a

  7. Dave Baker

    Dave Baker Guest

    Back in my youth I use to calibrate the F/A-1-11 Terrain Following Radar
    computer. Must have been one of the most complicated analog computers in it's
    day. Took about 8 hours of twiddling pots to get it out the door.

    I believe a digital system took over.

  8. Leon Heller

    Leon Heller Guest

    The LACE analogue computer I mentioned (all valves/tubes of course) had a
    multiplier - lots of valves in a 19" rack unit. I was quite good at setting
    the thing up - about 10 interacting pots that had to be tweaked. It took
    about four hours.

    A nice feature was the detachable patch panel, we had several. It could be
    wired up for a particular problem and then removed, ready to be re-used when
    the same problem needed to be solved. Amplifiers and feedback components
    were in small plug-in units, with turret tags for soldering in the
    components - usually precision resistors and polystyrene capacitors.
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day