Connect with us

Accurate Voltage on a Capacitor

Discussion in 'Electronic Design' started by Kuan Zhou, Mar 9, 2008.

Scroll to continue with content
  1. Kuan Zhou

    Kuan Zhou Guest

    Hi,

    We designed a circuit model that can maintain an accurate analog
    voltage on the capacitor through the positive feedback. Whever there is a
    current input to the current source, the analog voltage will
    increase/descrease correspondingly and this analog voltage can be maintained
    for several days with no attenuation. The initial purpose of this circuit is
    to emulate the brain working memory. We are wondering now what are the
    potential engineering applications for such circuits. Can any guys give us
    some suggestions?

    Kuan
     
  2. Guest


    If you ever build a time machine, you'll have a great circuit for 1959.
     
  3. D from BC

    D from BC Guest

    Sounds like a transconductance amp + sample and hold.


    D from BC
    British Columbia
    Canada.
     
  4. Robert Baer

    Robert Baer Guest

    "A few days" ... reminds me of some Intel PMOS RAM that required 3
    voltages..once programmed, power off, unplugged, they could reliably
    hold data for days.
     
  5. D from BC

    D from BC Guest

    I'm still thinking about brain function..
    The brain stores in terms of varying levels??? Kinda cool..
    Lots of people are probably assuming the brain is like a computer with
    1's and 0's.


    D from BC
    British Columbia
    Canada.
     
  6. The way I have heard it, things are learned by the
    production or elimination of synapses, that alter the
    interconnections of neurons and that way, vary the way they
    trigger each other. Think old plug-board programming a la
    ENIAC.

    http://en.wikipedia.org/wiki/ENIAC
     
  7. Tim Williams

    Tim Williams Guest

    Nah nah, base e. 0, 1, 2, ... 10(e) ~= 2.7182818284590 (dec). ;-)

    Incidentially, Pi (base e) ~= 10.1010020200021.

    Tim
     
  8. BobW

    BobW Guest

    That's good to know, but it still doesn't explain the universal significance
    of the number 42.

    Bob
     
  9. Tim Williams

    Tim Williams Guest

    42 (dec) = 2001.20121000000 (e)

    Hm, I'm sure there are more digits, but that's all of it i've got. Kinda
    funny that it's 2001 though.

    Tim
     
  10. Tim Williams

    Tim Williams Guest

    42 (dec) = 2001.20121000000 (e)

    Hm, I'm sure there are more digits, but that's all of it i've got. Kinda
    funny that it's 2001 though.

    Tim
     
  11. Tim Williams

    Tim Williams Guest

    42 (dec) = 2001.20121000000 (e)

    Hm, I'm sure there are more digits, but that's all of it i've got. Kinda
    funny that it's 2001 though.

    Tim
     

  12. See? That's what's wrong with dimbulb. He always comes up with
    666...


    --
    Service to my country? Been there, Done that, and I've got my DD214 to
    prove it.
    Member of DAV #85.

    Michael A. Terrell
    Central Florida
     
  13. There is no such accurate voltage at a capacitor, the charge varies on
    capacitance and current you supply dummy. Whatsamatta with you Chinese?
    Can't you figure it out on your own?


    --
    Service to my evil master? Been there, Done that, and I've got my DD666 to
    prove it.
    Member of DAV #666.

    Michael A. Terrell
    Central Florida
     

  14. ONLY A DUMBASS PUTS 00000 AFTER A DECIMAL NUMBERS, AND WORSE THAN THAT A
    Zero E (Exponent with no signification). Goddamn you idiots wasting
    natural resources, making Bogus War etc...


    --
    Service to my evil master? Been there, Done that, and I've got my DD666 to
    prove it.
    Member of DAV #666.

    Michael A. Terrell
    Central Florida
     
  15. Tom Bruhns

    Tom Bruhns Guest

    Hello,

    How large a capacitor? That is to say, what is the effective leakage
    current? We have operational amplifiers with input bias currents on
    the order of 10^-14 amps readily available and rather inexpensive, and
    certain capacitors I've tested have extremely long self-discharge time
    constants, so that the primary errors within a week's time are
    capacitance variation with temperature and dielectric absorption
    effects. I'm wondering if your circuit provides advantages beyond
    that.

    These days, it is quite possible to accomplish the function digitally
    with essentially zero droop rate; the only problems are component
    failure and perhaps external "noise" such as gamma rays or severe
    electromagnetic pulses that disrupt the circuit.

    Many years ago, when the first inertial navigation systems were
    developed, making integrators with very low drift rate was a big
    deal. Now, I don't think it is, at least not at the "days" level.
    "Years" or "decades" may be interesting.

    Others who have posted to this thread mentioned buried FET gates that
    can be charged and will then hold their charge for many years, but I'm
    not sure they count, unless there's a mechanism to actually control
    the charge on the gate over a continuous range accurately and at will.

    Cheers,
    Tom
     
  16. JosephKK

    JosephKK Guest

    Crap you are getting really obvious, forger.
     
  17. Robert Baer

    Robert Baer Guest

    Varying levels is much more likely: neuron firing levels are adjustable.
     
  18. Robert Baer

    Robert Baer Guest

    Makes it sound obsolete...
     
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

-