MosFet's from BS2e

Comments

• Philldapill Posts: 1,283

  2009-04-01 00:58 edited 2009-04-01 00:58

  I don't think you understand MOSFETs correctly. The are indeed a transistor, but they are very different than a bipolar junction transistor(BJT). A mosfet has a "gate" rather than a base. This gate controls the RESISTANCE between the "Drain"(like the Collector in an NPN transistor), and the "Source"(like the Emitter). The other difference, is that there is a tiny bit of capacitance between the Gate and Source. The voltage on this capacitance is what "turns on" the MOSFET. The other difference, is that there is no static voltage drop anywhere on the device, like there is with a BJT. The voltage drop between the Drain and Source, however, is directly due to the current flowing through it. This drop is equal to any resistive drop - I * R, where R is the resistance of the Drain-Source junction.
  
  To answer your questions:
  1) A 1K resistor would work fine, although you probably don't need it because once the gate is "charged", the mosfet remains on, and no more current flows into the gate - unlike a BJT where current MUST flow through the base to keep it on.
  
  2) You should probably use a resistor in that range(100k) from the GATE to SOURCE(probably ground). This is to ensure that the MOSFET stays off if the voltage is floating for some reason.
  
  3) Considering there is VERY little power dissipated in the resistor(if you are switching at fairly low frequencies - < 10kHz), the resistor sizes can be as small as you want.
  
  I hope that helps!
• Carl Hayes Posts: 841

  2009-04-01 01:39 edited 2009-04-01 01:39

  I noticed that, too.· It can't be an "NPN MOSfet", of course.· NPN implies a junction transistor, not a field-effect transistor.· It can be a P-channel·or an N-channel MOSfet, and can be depletion-mode (conducting with zero gate voltage, apply a voltage to shut it off) or enhancement-mode (not conducting with zero gate voltage, apply a voltage to turn it on).
  
  It's not quite correct to characterize a MOSfet as a resistance controlled by the gate voltage, though.· A resistor is perfectly linear, by definition, while a MOSfet is usually a little bit nonlinear.··At low source-drain currents it can be a lot nonlinear.· That's true with junction FETs, too, not just with MOSfets.· But as a design approximation, a resistance is often a good working assumption.
  
  I concur that a series resistor isn't needed -- except that it can protect the chip's logic output in the case of a malfunction such as a shorted gate, and also protect it from large transient gate-charging currents.· Some big MOSfets have a lot of gate capacitance.· But a series resistor slows the switching down, which can increase transient heating in the MOSfet as it switches.· Life is full of compromises.
  
  I'd use a resistor in series with the gate anyway, probably about 4.7K, though 100K would be OK too.· And I like the notion of a pulldown from gate to source for those times when the input (from a microprocessor) is in an input (floating)·state.· 100K may be OK, although it reduces the drive because of voltage division between the series and the pulldown resistors.· I'd probably use 4.7K and 470K respectively, so the drive reduction would be only 1%.· With 100K in both places, the drive would be reduced 50%, maybe too much.
  
  ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
  · -- Carl, nn5i@arrl.net
  
  Post Edited (Carl Hayes) : 4/1/2009 1:45:17 AM GMT
• hmlittle59 Posts: 404

  2009-04-01 02:18 edited 2009-04-01 02:18

  Thanks for the Info. , the parts (Transistor Polarity: P-Channel). I'll go back and study more on the difference.
  
  thanks again
  
  ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
  I HAVE LEARN SO MUCH...BUT STILL KNOW SO LITTLE!!!
  
  hmlittle59