Shop OBEX P1 Docs P2 Docs Learn Events
Mosfet circuit — Parallax Forums

Mosfet circuit

ArchiverArchiver Posts: 46,084
edited 2004-06-14 07:02 in General Discussion
Can someone tell me if the following mosfet circuit will work. I want to
turn a battery operated circuit on with a momentary ON pushbutton and have the
Stamp keep it on.

Battery plus to mosfet drain
Battery neg to system ground
Mosfet source to VR In
Mosfet gate to drain via pushbutton. Gate has 10K pulldown.

Press PB for 1 or 2 seconds, gate goes high, mosfet conducts, supplies power
to VR, then to Stamp. Stamp Pin 15 is held high by program, goes to gate and
keeps mosfet conducting. Later P15 goes low, turns off mosfet, turning off
battery.

Could this possibly work?

Sid


[noparse][[/noparse]Non-text portions of this message have been removed]

Comments

  • ArchiverArchiver Posts: 46,084
    edited 2004-06-12 11:37
    In a message dated 6/12/2004 10:23:06 AM Eastern Daylight Time,
    laurasdog@w... writes:


    > Even with a logic level FET, you will need to pull
    > the gate several volts higher than the source to
    > turn it on. If the source is connected to VR, where
    > will the drive signal come from that is a higher
    > voltage than that? Not from a stamp output pin...

    The drive signal will come from the battery when I press the pushbutton.

    Sid


    [noparse][[/noparse]Non-text portions of this message have been removed]
  • ArchiverArchiver Posts: 46,084
    edited 2004-06-12 11:48
    Correct,

    The gate needs to be several volts higher than the source to keep the device
    on, just as Steve mentioned.

    Dave recently brought up this subject. I thought he settled on a voltage
    regulator with an enable pin.

    Ken

    ===================
    Sid,
    No I don't think that will work...
    Even with a logic level FET, you will need to pull
    the gate several volts higher than the source to
    turn it on. If the source is connected to VR, where
    will the drive signal come from that is a higher
    voltage than that? Not from a stamp output pin...

    You can put the FET on the ground side of the
    circuit and make it work.

    Steve


    [noparse][[/noparse]Non-text portions of this message have been removed]
  • ArchiverArchiver Posts: 46,084
    edited 2004-06-12 11:57
    In a message dated 6/12/2004 10:54:21 AM Eastern Daylight Time,
    laurasdog@w... writes:


    > But when you release the button, the drive will come from a stamp
    > pin that is less than 5 volts above ground. If the FET source is
    > connected to V+ on the stamp, than the gate will need to be
    > higher than V+ by several volts to stay on.
    > Perhaps I'm not visualizing the circuit you propose correctly

    I think you see the circuit correctly. I did not know that the gate voltage
    has to be X volts higher that V+.

    Yoiu said something about using it to complete the ground circuit. How would
    I do that?

    Sid


    [noparse][[/noparse]Non-text portions of this message have been removed]
  • ArchiverArchiver Posts: 46,084
    edited 2004-06-12 12:08
    As several have stated, the design will not work as posted. I assume since
    you are running a stamp the voltage regulator is 5 volts, correct?

    And what is the voltage of the battery supply? Nine volts? 12 volts?

    If the battery voltage is high enough to provide a MINIMUM of about 10.5
    volts (5 volts plus Vce sat) one possible solution:

    Of course, the pushbutton momentary switch is needed and I don't see any
    problem there.
    To keep the gate of the logic level mosfet at 10 volts you could have the
    stamp ground the base lead of a PNP (2n3907).

    Emmitter of the 3907 to battery plus, collector of 3907 to the gate of the
    mosfet and the base of the 3907 to at stamp output pin that will go low to keep
    it on.

    I wonder though, when you make the stamp go high to turn things off, how it
    will perform.......hmmmm I am going to build one up right now.....if it works I
    will post a shematic in the files section of the basic stamps group.

    Ken
    Can someone tell me if the following mosfet circuit will work. I want to
    turn a battery operated circuit on with a momentary ON pushbutton and have
    the
    Stamp keep it on.


    [noparse][[/noparse]Non-text portions of this message have been removed]
  • ArchiverArchiver Posts: 46,084
    edited 2004-06-12 12:26
    Dohhhhhhhhhhh,

    3906

    k
    To keep the gate of the logic level mosfet at 10 volts you could have the
    stamp ground the base lead of a PNP (2n3907).

    Emmitter of the 3907 to battery plus, collector of 3907 to the gate of the
    mosfet and the base of the 3907 to at stamp output pin that will go low to
    keep


    [noparse][[/noparse]Non-text portions of this message have been removed]
  • ArchiverArchiver Posts: 46,084
    edited 2004-06-12 12:33
    In a message dated 6/12/2004 11:17:02 AM Eastern Daylight Time,
    smartdim@a... writes:


    > Dave recently brought up this subject. I thought he settled on a voltage
    > regulator with an enable pin.
    >

    I know, but that won't work. A user could reset the device anytime he wants
    without waiting for a mandatory 10 minutes pause, so we gave that up.

    Sid


    [noparse][[/noparse]Non-text portions of this message have been removed]
  • ArchiverArchiver Posts: 46,084
    edited 2004-06-12 12:57
    Connect the source to battery ground. Connect the drain to the stamps ground.
    ==============

    Keep one thing in mind. If the current that runs through the mosfet ground
    path (mosfet) rises enough, there will be a voltage drop across the FET and then
    the ground potential of the stamp is no longer at ground. A mosfet with low
    on resistance will eliminate that potential.

    Example, a IRL520 on resistance is about .25 ohms and at .5 amps that will
    raise the ground potential .125 volts.

    Depending how the external components are configured that may or may not
    cause a problem.

    One other point to consider is with the cicuit will be "at positive
    potential" when not enabled and any kind of ground that contacts the circuit where it
    is "live" could lead to smoke.

    Ken


    [noparse][[/noparse]Non-text portions of this message have been removed]
  • ArchiverArchiver Posts: 46,084
    edited 2004-06-12 13:16
    In a message dated 6/12/2004 11:40:41 AM Eastern Daylight Time,
    laurasdog@w... writes:


    > I use a lot of 30N06L FET's (digi-key), but any of the IRL series FET's
    > will
    > work. But not the IRF series....
    >
    >

    Thanks, Steve.

    The 30N06L is a 30A TO-220. Do you know of something in the 1A TO-92
    style tht would work? What about the 2N7000?

    Sid


    [noparse][[/noparse]Non-text portions of this message have been removed]
  • ArchiverArchiver Posts: 46,084
    edited 2004-06-12 13:21
    My idea below does not work......

    With the stamp powered off, the impeadance of the stamp pin that was to
    control the 2n3906 is low enough to activate the circuit.

    Ken
    Emmitter of the 3906 to battery plus, collector of 3906 to the gate of the
    mosfet and the base of the 3906 to at stamp output pin that will go low to
    keep
    it on.

    I wonder though, when you make the stamp go high to turn things off, how it
    will perform.......hmmmm I am going to build one up right now.....if it works
    I
    will post a shematic in the files section of the basic stamps group.

    Ken
    Can someone tell me if the following mosfet circuit will work. I want to
    turn a battery operated circuit on with a momentary ON pushbutton and have
    the
    Stamp keep it on.


    [noparse][[/noparse]Non-text portions of this message have been removed]
  • ArchiverArchiver Posts: 46,084
    edited 2004-06-12 13:22
    In a message dated 6/12/2004 12:12:08 PM Eastern Daylight Time,
    smartdim@a... writes:


    > Keep one thing in mind. If the current that runs through the mosfet ground
    > path (mosfet) rises enough, there will be a voltage drop across the FET and
    > then
    > the ground potential of the stamp is no longer at ground. A mosfet with low
    > on resistance will eliminate that potential.
    >
    > Example, a IRL520 on resistance is about .25 ohms and at .5 amps that will
    > raise the ground potential .125 volts.
    >

    I would think our max current will be about 70ma.

    Sid


    [noparse][[/noparse]Non-text portions of this message have been removed]
  • ArchiverArchiver Posts: 46,084
    edited 2004-06-12 13:24
    Doesn't you current proposed set up also have a momentary pushbutton switch
    to turn the circuit on the first time?

    When you say reset the circuit, do you mean to turn it on with the momentary
    switch?

    And if so, what prevents the useer from resetting the device anytime he wants.
    I know, but that won't work. A user could reset the device anytime he wants
    without waiting for a mandatory 10 minutes pause, so we gave that up.

    Sid


    [noparse][[/noparse]Non-text portions of this message have been removed]
  • ArchiverArchiver Posts: 46,084
    edited 2004-06-12 13:30
    In a message dated 6/12/2004 11:20:24 AM Eastern Daylight Time,
    alw@a... writes:


    > Keep in mind that not all FETs will fully switch on with 5V Vgs (the IRL
    > series will, however).
    >
    >

    Thanks, Al. I think now I understand how to use the Mosfet - I'll just have
    to be careful about which one if pick.

    Sid


    [noparse][[/noparse]Non-text portions of this message have been removed]
  • ArchiverArchiver Posts: 46,084
    edited 2004-06-12 13:37
    In a message dated 6/12/2004 12:25:41 PM Eastern Daylight Time,
    smartdim@a... writes:


    > Doesn't you current proposed set up also have a momentary pushbutton switch
    >
    > to turn the circuit on the first time?
    >

    We'e down to pushing a button to turn on the mosfet to turn on the Stamp.
    The Stamp program has a 10 minute pause, after which it takes the gate low
    and turns its self off. The system can then be restarted.

    Sid


    [noparse][[/noparse]Non-text portions of this message have been removed]
  • ArchiverArchiver Posts: 46,084
    edited 2004-06-12 13:49
    I must be missing something here.

    If you activate the mosfet by pushing a momentary switch, the stamp then
    takes over and keeps the mosfet turned on for 10 minutes. After 10 minutes has
    past, the user can again push the momentary switch to start over. Is my
    understanding correct?

    If my understanding is correct.....have the user turn on the voltage
    regulator with a momentary switch, then the first command for the stamp is to send a
    high (through 4.7k or so) to the same enable pin of the voltage regulator.After
    10 mnutes has past, disable the voltage regulator enable pin via the stamp.

    Am I missing something? I don't see any difference between enabling a mosfet
    or enabling a voltage regulator?

    Ken

    ================

    We'e down to pushing a button to turn on the mosfet to turn on the Stamp.
    The Stamp program has a 10 minute pause, after which it takes the gate low
    and turns its self off. The system can then be restarted.

    Sid


    [noparse][[/noparse]Non-text portions of this message have been removed]
  • ArchiverArchiver Posts: 46,084
    edited 2004-06-12 14:23
    In a message dated 6/12/2004 10:14:00 AM Pacific Standard Time,
    laurasdog@w... writes:

    The IRL520 is rated (I think) at 100 volts and has a fairly high Rds(on)
    =======================
    0.18 ohms


    [noparse][[/noparse]Non-text portions of this message have been removed]
  • ArchiverArchiver Posts: 46,084
    edited 2004-06-12 14:58
    Sid,
    No I don't think that will work...
    Even with a logic level FET, you will need to pull
    the gate several volts higher than the source to
    turn it on. If the source is connected to VR, where
    will the drive signal come from that is a higher
    voltage than that? Not from a stamp output pin...

    You can put the FET on the ground side of the
    circuit and make it work.

    Steve

    At 06:38 AM 6/12/2004, you wrote:

    >Can someone tell me if the following mosfet circuit will work. I want to
    >turn a battery operated circuit on with a momentary ON pushbutton and have
    >the
    >Stamp keep it on.
    >
    >Battery plus to mosfet drain
    >Battery neg to system ground
    >Mosfet source to VR In
    >Mosfet gate to drain via pushbutton. Gate has 10K pulldown.
    >
    >Press PB for 1 or 2 seconds, gate goes high, mosfet conducts, supplies power
    >to VR, then to Stamp. Stamp Pin 15 is held high by program, goes to gate and
    >keeps mosfet conducting. Later P15 goes low, turns off mosfet, turning off
    >battery.
    >
    >Could this possibly work?
    >
    >Sid
    >
    >
    >[noparse][[/noparse]Non-text portions of this message have been removed]
    >
    >
    >
    >To UNSUBSCRIBE, just send mail to:
    > basicstamps-unsubscribe@yahoogroups.com
    >from the same email address that you subscribed. Text in the Subject and
    >Body of the message will be ignored.
    >
    >Yahoo! Groups Links
    >
    >
    >
    >


    [noparse][[/noparse]Non-text portions of this message have been removed]
  • ArchiverArchiver Posts: 46,084
    edited 2004-06-12 15:53
    >
    >The drive signal will come from the battery when I press the pushbutton.
    >
    >Sid

    But when you release the button, the drive will come from a stamp
    pin that is less than 5 volts above ground. If the FET source is
    connected to V+ on the stamp, than the gate will need to be
    higher than V+ by several volts to stay on.
    Perhaps I'm not visualizing the circuit you propose correctly...
    Got a schematic?

    Steve


    [noparse][[/noparse]Non-text portions of this message have been removed]
  • ArchiverArchiver Posts: 46,084
    edited 2004-06-12 16:06
    Hi Sid,

    Look at http://www.al-williams.com/fetrly.htm

    Keep in mind that not all FETs will fully switch on with 5V Vgs (the IRL
    series will, however).

    Regards,

    Al Williams
    AWC
    * Add floating point math to any processor (now on sale):
    http://www.awce.com/pak12.htm




    Original Message
    From: Newzed@a... [noparse]/noparse]mailto:[url=http://forums.parallaxinc.com/group/basicstamps/post?postID=xQ7C8gWVXI2GA-WZe2k6BZlFI26D5l8Gvs4213tNocyvdIZo1uEnDcJrQC60zOpncpsAWOZzwx4]Newzed@a...[/url
    Sent: Saturday, June 12, 2004 9:58 AM
    To: basicstamps@yahoogroups.com
    Subject: Re: [noparse][[/noparse]basicstamps] Mosfet circuit


    In a message dated 6/12/2004 10:54:21 AM Eastern Daylight Time,
    laurasdog@w... writes:


    > But when you release the button, the drive will come from a stamp pin
    > that is less than 5 volts above ground. If the FET source is connected
    > to V+ on the stamp, than the gate will need to be higher than V+ by
    > several volts to stay on. Perhaps I'm not visualizing the circuit you
    > propose correctly

    I think you see the circuit correctly. I did not know that the gate voltage

    has to be X volts higher that V+.

    Yoiu said something about using it to complete the ground circuit. How
    would
    I do that?

    Sid


    [noparse][[/noparse]Non-text portions of this message have been removed]



    To UNSUBSCRIBE, just send mail to:
    basicstamps-unsubscribe@yahoogroups.com
    from the same email address that you subscribed. Text in the Subject and
    Body of the message will be ignored.

    Yahoo! Groups Links
  • ArchiverArchiver Posts: 46,084
    edited 2004-06-12 16:26
    At 07:57 AM 6/12/2004, you wrote:

    >In a message dated 6/12/2004 10:54:21 AM Eastern Daylight Time,
    >laurasdog@w... writes:
    >
    >
    > > But when you release the button, the drive will come from a stamp
    > > pin that is less than 5 volts above ground. If the FET source is
    > > connected to V+ on the stamp, than the gate will need to be
    > > higher than V+ by several volts to stay on.
    > > Perhaps I'm not visualizing the circuit you propose correctly
    >
    >I think you see the circuit correctly. I did not know that the gate voltage
    >has to be X volts higher that V+.
    >
    >Yoiu said something about using it to complete the ground circuit. How would
    >I do that?
    >
    >Sid

    Connect the battery + to the stamp V+ like normal.
    Connect the source to battery ground. Connect the drain to the stamps ground.

    Now turning the FET on will complete the ground path to turn the stamp on.
    If you use a logic level FET, you will need a gate signal that goes to 3-5
    volts
    ^above ground* to turn it on, which the stamp can provide.
    Many "regular" FETs require 8-10 volts of gate drive to turn fully on, but
    Logic
    Level FETs work at standard 5v logic level, perfect for use on a stamp.

    I use a lot of 30N06L FET's (digi-key), but any of the IRL series FET's will
    work. But not the IRF series....

    Steve





    [noparse][[/noparse]Non-text portions of this message have been removed]
  • ArchiverArchiver Posts: 46,084
    edited 2004-06-12 17:39
    As a general rule, Rds(on) is related to the voltage rating of the
    device. FETs with lower voltage ratings will likely have lower Rds(on)

    The IRL520 is rated (I think) at 100 volts and has a fairly high Rds(on)

    The 30N06L is rated for 30 volts and has a Rds(on) of less
    than 0.05ohm. So at 1 amp you're only looking at a drop of 0.05 volts
    across the FET.

    Sorry, I don't have any part #'s for TO92 sized FET's... when I need
    a FET I'm usually looking to control something big and nasty.. (-:

    Look for logic level FETs with a nice low Rds(on).

    Steve

    At 09:22 AM 6/12/2004, you wrote:

    >In a message dated 6/12/2004 12:12:08 PM Eastern Daylight Time,
    >smartdim@a... writes:
    >
    >
    > > Keep one thing in mind. If the current that runs through the mosfet ground
    > > path (mosfet) rises enough, there will be a voltage drop across the FET
    > and
    > > then
    > > the ground potential of the stamp is no longer at ground. A mosfet with
    > low
    > > on resistance will eliminate that potential.
    > >
    > > Example, a IRL520 on resistance is about .25 ohms and at .5 amps that will
    > > raise the ground potential .125 volts.
    > >
    >
    >I would think our max current will be about 70ma.
    >
    >Sid


    [noparse][[/noparse]Non-text portions of this message have been removed]
  • ArchiverArchiver Posts: 46,084
    edited 2004-06-12 19:29
    The IRL520 is a member of a family with varying Rds values. All are rated
    at 100V. Just select the device with the Rds you need:

    IRL520N: Rds = 0.18
    IRL530N: Rds = 0.10
    IRL540N: Rds = 0.044

    The above Rds values are for 10V on the gate. At 5V gate voltage the Rds is
    slightly higher (about 20%).

    -Randy
    www.glitchbuster.com



    > As a general rule, Rds(on) is related to the voltage rating of the
    > device. FETs with lower voltage ratings will likely have lower Rds(on)
    >
    > The IRL520 is rated (I think) at 100 volts and has a fairly high Rds(on)
    >
    > The 30N06L is rated for 30 volts and has a Rds(on) of less
    > than 0.05ohm. So at 1 amp you're only looking at a drop of 0.05 volts
    > across the FET.
    >
    > Sorry, I don't have any part #'s for TO92 sized FET's... when I need
    > a FET I'm usually looking to control something big and nasty.. (-:
    >
    > Look for logic level FETs with a nice low Rds(on).
    >
    > Steve
  • ArchiverArchiver Posts: 46,084
    edited 2004-06-12 20:21
    Question:
    For the same Vds, why wouldn't you always just select the device with the
    lowest Rds? What are the tradeoffs? Different max current? When would you
    want a high Rds? Or?
    Thanks!
    Steve
    //////////////////////////////////

    At 11:29 AM 6/12/04, you wrote:
    >The IRL520 is a member of a family with varying Rds values. All are rated
    >at 100V. Just select the device with the Rds you need:
    >
    >IRL520N: Rds = 0.18
    >IRL530N: Rds = 0.10
    >IRL540N: Rds = 0.044
    >
    >The above Rds values are for 10V on the gate. At 5V gate voltage the Rds is
    >slightly higher (about 20%).
    >
    >-Randy
    >www.glitchbuster.com
    >
    >
    >
    > > As a general rule, Rds(on) is related to the voltage rating of the
    > > device. FETs with lower voltage ratings will likely have lower Rds(on)
    > >
    > > The IRL520 is rated (I think) at 100 volts and has a fairly high Rds(on)
    > >
    > > The 30N06L is rated for 30 volts and has a Rds(on) of less
    > > than 0.05ohm. So at 1 amp you're only looking at a drop of 0.05 volts
    > > across the FET.
    > >
    > > Sorry, I don't have any part #'s for TO92 sized FET's... when I need
    > > a FET I'm usually looking to control something big and nasty.. (-:
    > >
    > > Look for logic level FETs with a nice low Rds(on).
    > >
    > > Steve
  • ArchiverArchiver Posts: 46,084
    edited 2004-06-13 01:09
    At 12:21 PM 6/12/2004, you wrote:

    >Question:
    >For the same Vds, why wouldn't you always just select the device with the
    >lowest Rds? What are the tradeoffs? Different max current? When would you
    >want a high Rds? Or?
    >Thanks!
    >Steve
    >//////////////////////////////////


    A lot of times the trade off is cost. Within the same family, the lower
    RDs(on) units are usually more expensive.
    No point in buying a $0.99 FET when a $0.89 one will do... :-)

    Steve


    [noparse][[/noparse]Non-text portions of this message have been removed]
  • ArchiverArchiver Posts: 46,084
    edited 2004-06-13 01:40
    The devices with lower Rds have higher max current rating, but also a higher
    cost. So cost is the tradeoff. For example:

    IRL520N: Rds = 0.18 -- $0.67
    IRL530N: Rds = 0.10 -- $0.87
    IRL540N: Rds = 0.044 -- $1.17

    For many Stamp applications even the IRL520N is overkill, but if you're
    looking for even lower Rds the IRL530N has just over 1/2 the IRL520N's Rds
    for only a slightly higher price.

    Another interesting thing to note is that the devices with the lower Rds
    will drop less voltage and dissipate less heat for any given load current.
    Using the IRL540N in place of one of the others with higher Rds could
    eliminate the need for a heat sink in some higher current applications, and
    end up costing less overall. In addition, more of the supply voltage gets
    to the load device instead of dropping in the transistor.

    But... for many applications the IRL520N without a heat sink is more than
    enough.

    -Randy
    www.glitchbuster.com



    Original Message
    From: "Steve Roberts" <sroberts@s...>
    To: <basicstamps@yahoogroups.com>
    Sent: Saturday, June 12, 2004 12:21 PM
    Subject: Re: [noparse][[/noparse]basicstamps] Mosfet circuit


    > Question:
    > For the same Vds, why wouldn't you always just select the device with the
    > lowest Rds? What are the tradeoffs? Different max current? When would you
    > want a high Rds? Or?
    > Thanks!
    > Steve
    > //////////////////////////////////
    >
    > At 11:29 AM 6/12/04, you wrote:
    > >The IRL520 is a member of a family with varying Rds values. All are
    rated
    > >at 100V. Just select the device with the Rds you need:
    > >
    > >IRL520N: Rds = 0.18
    > >IRL530N: Rds = 0.10
    > >IRL540N: Rds = 0.044
    > >
    > >The above Rds values are for 10V on the gate. At 5V gate voltage the Rds
    is
    > >slightly higher (about 20%).
    > >
    > >-Randy
    > >www.glitchbuster.com
  • ArchiverArchiver Posts: 46,084
    edited 2004-06-13 17:37
    --- In basicstamps@yahoogroups.com, laurasdog@w... wrote:
    > At 12:21 PM 6/12/2004, you wrote:
    >
    > >Question:
    > >For the same Vds, why wouldn't you always just select the device
    with the
    > >lowest Rds? What are the tradeoffs? Different max current? When
    would you
    > >want a high Rds? Or?
    > >Thanks!
    > >Steve
    > >//////////////////////////////////
    >
    >
    >
    > A lot of times the trade off is cost. Within the same family, the
    lower
    > RDs(on) units are usually more expensive.
    > No point in buying a $0.99 FET when a $0.89 one will do... :-)



    In reading the origional post, it appears that the situation is for a
    battery operated unit and that by turning the FET ON the device is
    powered.

    So, if you have a high resistance on the FET, the unit will stop
    working at a higher battery voltage. the difference can easily be a
    1/2 volt. And, if one is starting with 4 AA batteries, the dropout
    will have a pretty big difference for that 1/2 volt.


    I question I have is about the voltage and current needed on the gate
    at these low power/low voltage uses. Is there anything special
    needed ?

    Dave
  • ArchiverArchiver Posts: 46,084
    edited 2004-06-13 19:18
    >In reading the origional post, it appears that the situation is for a
    >battery operated unit and that by turning the FET ON the device is
    >powered.
    >
    >So, if you have a high resistance on the FET, the unit will stop
    >working at a higher battery voltage. the difference can easily be a
    >1/2 volt. And, if one is starting with 4 AA batteries, the dropout
    >will have a pretty big difference for that 1/2 volt.

    Dave, most low voltage FET's you would use in a battery operated
    circuit will have an Rds(on) of 0.2 ohms or less. There's *lots* of
    good cheap FETs with Rds(on) of less than 0.05 ohm.
    So even at 2 amps (which is a lot for 4 AA batteries) the voltage
    lost across the FET would likely be less than 0.1 volts. That's
    certainly better than using a bipolar device...

    >I question I have is about the voltage and current needed on the gate
    >at these low power/low voltage uses. Is there anything special
    >needed ?

    Not much required to drive a FET gate. The voltage on the gate
    (relative to the source) for logic level devices just needs to be a
    few volts to turn fully on, and usually can't exceed around 10 volts
    without possible gate damage. Check the data sheet for your
    device, and all that...

    The gate current requirement is essentially zero. Most MOSFETs
    have gate input impedances that exceed 10's of Megohms, and
    most exceed that by a great deal..

    Steve


    [noparse][[/noparse]Non-text portions of this message have been removed]
  • ArchiverArchiver Posts: 46,084
    edited 2004-06-14 04:01
    > Dave, most low voltage FET's you would use in a battery operated
    > circuit will have an Rds(on) of 0.2 ohms or less. There's *lots* of
    > good cheap FETs with Rds(on) of less than 0.05 ohm.


    My first selection was a 2N7000 MOSFET commonly used in place of a
    2N2222 darlington.

    the spec sheet from Fairchild lists it as 5 ohms.

    I was surprized at the high resistance, thus the cause fro my post.

    On the surface, it seems the TO-220 case may offer a larger area for
    the junction and therefor much lower resistance.



    Dave






    > So even at 2 amps (which is a lot for 4 AA batteries) the voltage
    > lost across the FET would likely be less than 0.1 volts. That's
    > certainly better than using a bipolar device...
    >
    > >I question I have is about the voltage and current needed on the
    gate
    > >at these low power/low voltage uses. Is there anything special
    > >needed ?
    >
    > Not much required to drive a FET gate. The voltage on the gate
    > (relative to the source) for logic level devices just needs to be a
    > few volts to turn fully on, and usually can't exceed around 10 volts
    > without possible gate damage. Check the data sheet for your
    > device, and all that...
    >
    > The gate current requirement is essentially zero. Most MOSFETs
    > have gate input impedances that exceed 10's of Megohms, and
    > most exceed that by a great deal..
    >
    > Steve
    >
    >
    > [noparse][[/noparse]Non-text portions of this message have been removed]
  • ArchiverArchiver Posts: 46,084
    edited 2004-06-14 05:48
    > > >For the same Vds, why wouldn't you always just select the device
    >with the
    >> >lowest Rds? What are the tradeoffs? Different max current? When
    >would you
    >> >want a high Rds? Or?
    >> >Thanks!
    > > >Steve

    Another option for your power-hold circuit might be one of the
    photomos relays, example Aromat AQV212 in 6 pin DIP, rated 400 mA.
    These have an LED input, and a mosfet output. The pushbutton via a
    resistor could turn on the led, which powers up the Stamp, which
    continues to power the led. You only need a couple of resistors in
    addition to the AQV212. It would be more expensive than the circuit
    using the plain mosfet, though.

    I'm not sure if anyone mentioned this, but another factor in the
    choice of a mosfet is switching speed. The higher current devices
    often have very large junction area, and that translates into a large
    input capacitance. For example, the capacitance of the IRL520 is
    about 500 pf, while the capacitance of the higher current IRL540 is
    1800 pf. The circuit that drives the gate has to charge up that
    capacitance or discharge it every time it changes state. That is not
    a limitation for the DC circuit you are talking about, but it could
    be a factor in a PWM circuit or a switching power supply.

    There are many applications for mosfets. For example, the mosfet
    chosen for a high speed sample and hold circuit might have an ON
    resistance of 5 ohms, but an input capacitance of 10 or 20 pf. A low
    gate to channel capacitance tranlates to rapid switching and also low
    "charge injection" from the digital to the analog circuitry.


    -- Tracy
  • ArchiverArchiver Posts: 46,084
    edited 2004-06-14 07:02
    > I'm not sure if anyone mentioned this, but another factor in the
    > choice of a mosfet is switching speed. The higher current devices
    > often have very large junction area, and that translates into a
    large
    > input capacitance. For example, the capacitance of the IRL520 is
    > about 500 pf, while the capacitance of the higher current IRL540 is
    > 1800 pf. The circuit that drives the gate has to charge up that
    > capacitance or discharge it every time it changes state. That is
    not
    > a limitation for the DC circuit you are talking about, but it could
    > be a factor in a PWM circuit or a switching power supply.

    I had mentioned the high capacatance of the MOSFET a few months ago,
    but it was about the need to switch the MOSFET fast. if one tries to
    trickle the unit to turn on, the current passing the junction will
    destroy the chip as the small sections start to pass the current, the
    current will overheat the sections and burn it up.

    That is one of the reasons for using mosfet driver chips with high
    amp ratings in really tiny packages.

    Very interesting discussion this is becomming.

    Dave





    >
    > There are many applications for mosfets. For example, the mosfet
    > chosen for a high speed sample and hold circuit might have an ON
    > resistance of 5 ohms, but an input capacitance of 10 or 20 pf. A
    low
    > gate to channel capacitance tranlates to rapid switching and also
    low
    > "charge injection" from the digital to the analog circuitry.
    >
    >
    > -- Tracy
Sign In or Register to comment.