Connect with us

rectifier

Discussion in 'Electronic Design' started by Abstract Dissonance, Dec 30, 2005.

Scroll to continue with content
  1. I was wondering if there is any type of circuits that use transistors in
    place of the diodes in a rectifier circuit?

    I was thinking that maybe one could use them in some way to smooth out a
    source by using some type of negative feedback or something... or maybe
    there could be some other use?

    Just wondering,

    Thanks,
    AD
     
  2. MOSFETs are used instead of rectifying diodes because they have less
    resistance, IIRC.

    lg, Bernhard
     
  3. That's very clever!

    i
     
  4. What do you mean by backwards?

    i
     
  5. The trick is that you have to use them backwards because of the body
    diode.


    Best regards,
    Spehro Pefhany
     
  6. Pooh Bear

    Pooh Bear Guest

    Check out synchronous rectification.

    Graham
     
  7. Graham, what you showed me is just plainly fantastic! It's just what I
    need for a certain application (AC voltage measurement).

    i
     
  8. When the MOSFET is conducting the source will be positive wrt the
    drain on an n-channel device. The body diode is thus in parallel with
    the channel, and hopefully not conducting significantly. The opposite
    polarity when it's blocking.


    Best regards,
    Spehro Pefhany
     
  9. Spehro, maybe you can educate me a little bit. I am hoping that I
    could use a mosfet with its gate not energized (connected to drain),
    as a diode with low voltage drop, acting, rather, like a resistor. It
    would conduct from drain to source.

    Is it correct that I can do something like that?

    If so, what mosfet would you suggest for low voltage, low current applications?

    thanks

    i
     
  10. Pooh Bear

    Pooh Bear Guest

    Pleased to have been of assistance. :)

    I'm intruiged now !

    Graham
     
  11. No, it's not that easy. The MOSFET won't really start to conduct until
    the gate voltage is several volts, so it can't really be a 2-terminal
    device and still be better than a silicon diode. You have to drive the
    gate from some other circuitry. What you'd get if you did what you
    suggest is just a slow and expensive diode with normal voltage drop.

    For measuring applications often we use a "precision rectifier". There
    are a few ways of doing it, most of which involve an op-amp and a
    diode or two. Rather than drone on (in the Larkinsian sense), I'll
    suggest something you might not find-- If you just need to indicate AC
    mV on a moving-coil panel meter, you can do something like this:


    +---|>|----+----|<|-----+
    | + | |
    | ,---. |
    | | M | |
    | '---' |
    | - | |
    +----|<|---+-----|>|----+
    | |
    | .-.
    | | |
    | +V | | Rb
    | |\| '-'
    +--------|-\ |
    | | >-----------+
    .-. +---|+/
    | | | |/|
    Ra | | | -V
    '-' |
    | o input
    |
    ===
    GND

    Ra sets the sensitivity, Rb limits the coil current and
    causes the op-amp to have to slew faster.



    Or maybe this where the extra diodes shunt current away from
    overloads and damp the movement.


    +---|>|----+------+-----+-- -|<|-----+
    | + | | | |
    | ,---. | | |
    | | M | | | |
    | '---' | | |
    | - | --- --- |
    | | / \ \ / |
    | .-. --- --- |
    | Rs| | | | |
    | | | | | |
    | '-' | | |
    | | | | |
    +----|<|---+------+-----+-----|>|----+
    | |
    | |
    | |
    | +V |
    | |\| |
    +--------|-\ |
    | | >------------------------+
    .-. +---|+/
    | | | |/|
    Ra | | | -V
    '-' |
    | o input
    |
    ===
    GND




    Best regards,
    Spehro Pefhany
     
  12. Pooh Bear

    Pooh Bear Guest

    Ah - you mean you want a precision rectifier for measurement purposes where
    there's no Vf from the diodes ?

    Spehro has posted one example but it can be made to provide a ground
    referenced output too.

    Graham
     
  13. Jim Thompson

    Jim Thompson Guest

    As in "FullWaveRectifier.pdf"

    on the S.E.D/Schematics page of my website.

    ...Jim Thompson
     
  14. Pooh Bear

    Pooh Bear Guest

    http://www.edn.com/archives/1997/030397/05DI_04.htm

    Graham
     
  15. no, I didn't want anything really... I was just wondering if there were such
    things ;) I figured it would be "generalization" of sorts and might be
    useful... just wondering if there were such circuits that were similar but
    with other components replaceing the diodes. I think the wheatstone bridge
    is of similar design but with resistors or whatever?

    AD
     
  16. the Jackal

    the Jackal Guest


    Slow diodes can take even hundreds of microseconds to turn off though
    popular fast turn off/on diodes like the 1N914 (used in computers) go
    over their expected dc level upto even 0.5V for 10 to 20ns before the
    diode turns on. Using the emitter base junction of BJTs like the
    2N3904 as diodes negligible overshoot voltages are observed and turn
    on times of 0.1ns are observed. Their leakage current is lower then
    diodes too.
    Disadvantage is they can withstand very low reverse voltages.

    You'll find diode circuits that need speed in switchers , fast pulse
    detectors and some voltage clamps.


    In the field of audio I've heard of transistors used as diodes
    exploiting the 2nd harmonics generated by BJTs to give
    better/different sound outputs compared to diodes, but as you know
    this may just be another audio myth!
     
  17. Tim Williams

    Tim Williams Guest

    My experiments with 2N4401 show the B-E junction has recovery time
    comparable to the storage time, which makes reasonable amounts of sense
    (IANA semiconductor physicist).

    Tim
     
  18. Tim Williams wrote...
    I prefer the base-collector diode.
     
  19. Joseph2k

    Joseph2k Guest

    yes, but in that case the breakdown voltage is usually limited to about 5 V;
    and the CB junction is usually not a fast. it is tradeoffs.
     
  20. John Larkin

    John Larkin Guest

    We have one product whose operation involves deliberately forward
    biasing a power diode to +48 volts for a short time.

    John
     
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

-