Connect with us

Help understanding N-Channel MOSFET body diode

Discussion in 'Electronic Design' started by [email protected], May 18, 2007.

Scroll to continue with content
  1. Guest

    I tried this question in the BASICS forum but did not receive a
    response. I thought I would try here next. I appologise in advance
    for cross-posting.

    Is there a rule of thumb stating how much power an N-Channel MOSFET's
    internal body diode (anode at source cathod at drain) can handle vs.
    the normal use drain to source power rating of the device?

    Does this make sense?


    I am using an Nch MOSFET to switch a load to ground in a battery
    operated project. The design does not allow me to place a rectifier
    diode in series with the load to prevent current flow through the
    body
    diode in the event of reverse battery connection. I was wondering if
    the battery was connected in reverse, would the load limit current
    enough to save the device? Unfortunately I never see the power
    rating
    (or forward voltage drop of this diode) published in manufactures
    data
    sheets so I have no way to tell.


    Thank you in advance,


    Ge0
     
  2. Phil Allison

    Phil Allison Guest

    ** The intrinsic body diode of a mosfet has a fairly high conduction
    voltage - from 1 to 2 volts - making heat dissipation in the device much
    higher than is typically the case for saturated switching in the normal
    direction.

    Work out the load amps and estimate the dissipation in watts to see if your
    case is OK.

    If not - add a parallel ( Schottky low forward drop ) diode with sufficient
    ratings and heatsinking to do the job instead.



    ....... Phil
     
  3. Guest

    Thanks for the prompt response Phil. Great idea adding the Schottky.
    I guess I'll need to get/create a forward voltage drop graph showing
    V(f) throughout the temp span and at the worse case fault I(f) to see
    if I need to go to these measures.

    Cheers,

    Ge0
     
  4. You might want to look at devices like the Harris HUF75337G3, and Hitachi
    do similar transistors, which have got proper specifications for the
    diode, and are designed to allow this to be used. Generally, if the diode
    doesn't have a specification in the data sheet, then reckon that it may
    well have quite poor characteristics. Specifically, the conduction voltage
    may well be high, risking damage to other components.
    Remember also that you can generate a diode with a lower voltage drop than
    a Schottky diode, using a MOSFET, for series connection, if it is the drop
    that is the problem for a series diode.

    Best Wishes
     
  5. John Larkin

    John Larkin Guest

    I once blew up a lot of Mororola mosfets in a half-bridge motor
    controller application. The motor back-emf forward-biased a substrate
    diode, then the opposite fet turned on, and the substrate diode made a
    step-recovery spike so big it blew out its own gate. I wonder if their
    reverse recovery is better controlled these days.

    John
     
  6. Yes, in some.
    However it is an amazingly common failure mode, with people 'routinely'
    relying on the internal diode, without realising how poorly suited it may
    be...
    Unfortunately, there is a tendency to 'see' the diode shown in the data
    sheet, and assume it is well matched for the application, without
    realising just how slow it is.

    Best Wishes
     
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

-