Connect with us

3.3V Logic-Level MOSFET

Discussion in 'Electronic Design' started by Ted Lechman, May 21, 2004.

Scroll to continue with content
  1. Ted Lechman

    Ted Lechman Guest

    Has anyone already found what I'm looking for?

    I'm looking for a high side MOSFET to control power to a low power ( <
    100ma ) device. The voltage to be switched is 3.3V and the control
    (gate) voltage to the MOSFET is also 3.3V (not 5V). the RDSON is not a
    major factor. Do you know of a MOSFET part that will work at these
    gate and input voltages?

  2. Tim Wescott

    Tim Wescott Guest

    Vishay/Siliconix has a number of them. Just look for "logic level" on
    their selector guide (then read the data sheet with care).
  3. Lots of them (assuming you're looking for a p-channel MOSFET). For
    example, the IRF7425 which has 0.013 ohm Rds(on) at -2.5V. About 85
    cents US in 100's.

    Look for "logic" parts with a very low Vds (something like -20V) and
    many of them will fit your specifications if you want to search for an
    optimal part.

    Best regards,
    Spehro Pefhany
  4. Tim Wescott wrote...
    Many small power MOSFETs (which can switch 1A or more with the usual
    5V gate drive) can work quite well with 3.3V drive (which is a TTL
    "HI" level). For example, the common 2n7000 and 2n7002 will switch
    up to 50mA just fine with 3.3V gate voltage. Perhaps not offically,
    according to the data sheet, but in fact they do, as can be shown
    using a mathematical analysis of the data-sheet numbers. Switching
    100 - 200mA with 3.3V is a bit tougher for legacy parts, but as Tim
    says, there are lots of new "logic-level" parts that work well. For
    example the Supertex TN0702, available in the TO-92 package, and the
    TN0604, available in TO-92 and SMT. But sadly most of the attractive
    new logic-level parts are *only* available in SMT packages, like the
    Fairchild nds331 and nds335, the Philips pmv30, pmv31 and pmv56, and
    the Vishay Si2302, which are all in SOT-23 packages. These parts can
    switch several amps.

    - Win

    (email: use hill_at_rowland-dot-org for now)
  5. Tim Wescott

    Tim Wescott Guest

    So, has Supertex been around for years and me too dim to notice, or did
    they have a different name last year?
  6. 1976.
  7. Tim Wescott wrote...
    Supertex has long been a favorite of mine, initially because they
    habitually offered each die in an array of different packages. But
    they're not as well known as other companies, and now most of their
    interesting parts (e.g. low-capacitance high-voltage FETs in TO-220
    power packages) have been discontinued. <sigh>

    BTW, I misread the O.P.'s question, and all my suggested parts are
    N-channel types. Spef suggested a P-channel part, the IRF7425,
    causing me to carefully re-read the question, which asked for a
    "high side MOSFET" switch. Oops! Sorry about that!

    - Win

    (email: use hill_at_rowland-dot-org for now)
  8. Joerg

    Joerg Guest

    Hi Winfield,

    Most of the time I found it easier to use PNP transistors in cases like this
    where it is just about switching on power to a device. In a Darlington setup if
    there wan't enough drive power. At least then you are guaranteed that only a
    few hundred millivolts get dropped. FETs are often too fickle when operated so
    close to their steep region. Suddenly you receive a shipment that is on the
    high side of Ugs and things get toasty.

    Regards, Joerg
  9. Joerg wrote...
    This is eminently sensible. These days we love power MOSFETs. But we
    may seek to use them even if they're less than ideal for an application.
    That's not good. Driving a PNP transistor with 4 to 10mA in order to
    switch up to 100mA makes good sense. The PNP is cheaper and easier to
    get. And it's very well understood. Low Vgs FETs are becoming widely
    available, and are dependable batch to batch. But they're expensive,
    and may go out of stock or even be discontinued. Not a happy scene.

    - Win

    (email: use hill_at_rowland-dot-org for now)
  10. Joerg

    Joerg Guest

    Hi Winfield,

    With "cheaper" you actually chose a really important word here. Bipolars beat FETs
    almost all the time under a few hundred mA. Higher up it can be different.

    Academia is really a bit behind on this topic. They do not teach cost thinking well
    enough, in my times not at all. There should be seminars where students a presented
    with a huge list of parts or for that matter with the URL for Digikey. Then they'd
    have to create a certain circuit and get graded on what it costs. Of course it also
    has to work.

    The part suggestions in Art of Electronics are great, especially for folks getting
    started. Maybe a new edition could contain rough budgetary pricing. That would
    teach engineers a lesson or two on the penalties of going an easy route. We almost
    always can get that panacea chip or dream transistor that does most of the grunt
    work but at the end of the day that may blow the budget.

    Regards, Joerg
  11. Joerg wrote...
    OK, thanks for the thought, good idea. And it comes at a good
    time in our writing.

    - Win

    (email: use hill_at_rowland-dot-org for now)
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