difference between bipolar and mosfet

[Skeleton Man]

Just wondering could someone explain fairly simply what the difference is
between a bipolar and a fet ? Can I put a bipolar in place of a fet or vice
versa ?

Regards,
Chris

 

[Skeleton Man]

so if I'm to understand correctly.. a bi-polar will pass current between
collector and emitter when a voltage is applied to the base ? and a fet will do
a simmilar thing only doesn't require a current ? (at whichever terminal
corresponds to a base on a bipolar)

Regards,
Chris

 

[Kevin Aylward]

Robert Monsen wrote:

Kevin Aylward wrote:
Robert Monsen wrote:

Skeleton Man wrote:

Just wondering could someone explain fairly simply what the
difference is between a bipolar and a fet ? Can I put a bipolar in
place of a fet or vice versa ?

Regards,
Chris



The most basic difference is that a bipolar transistor requires
current at the control terminal (the base lead), whereas a mosfet
requires none. However, there are advantages to both in different
situations.
You generally cannot substitute a bipolar transistor for a fet,
because the circuit will not be designed to supply the required base
current.
MOSFETs have three leads, a source, a gate, and a drain. Bipolar
transistors also have three leads, but they are called emitter,
base, and collector. These leads roughly correspond to one another,
ie, the emitter is like the source, the base is like the gate, and
the collector is like the drain. Making the base (gate) more
positive (for NPN and N-MOSFETs) or negative (for PNP or P-MOSFETs)
with respect to the emitter (source) causes more current to flow
from collector (drain) to emitter (source).

This terminology is totally confusing,


Why?


I'm not saything they are wrong, I'm just saying that it's a confusing
blob of information until you memorize it. Once you figure it out, you
can convince yourself that it makes sense, just like any terminology.

The names have been specifically chosen to describe how the device
actually functions.

Charge carriers are sourced or emitted from the source/emitter. These
carriers are then drained off or collected by the drain/collector.
The gate or base *voltage* controls the flow of carriers. I will
give you that "base" is not on a par with "gate" in describing its
function.
and, sadly, you just have to
get used to it if you want to talk about these things.


Once one understands the names, one will understand how mosfet and
bipolar actually function. If you don't understand why the names are
as they are, you wont understand how the devices function.


Right, you have to understand how the device works in order to
understand the names of the terminals. Unfortunately, that is
confusing for beginners, who often want to simply build something
simple, and get confused by emitters, collectors, where which goes,
whether PNP or NPN should be used, etc.


MOSFETs are used to construct CMOS devices, and are thus the main
transistor component to microprocessors. They are also good for
constructing huge power transistors, which are easier to control due
to the lack of required gate current.

Bipolar transistors are generally more useful for analog design,
where the lower noise, more easily predicted voltage requirements,
and lower control voltages are useful.

For a FET, the electrostatic field of charges on the control
terminal (the gate) is used to moderate the output. MOSFETs have a
silicon oxide layer that insulates the gate from the charge. JFETs
use a reverse-biased PN junction's depletion region to isolate the
gate from the source and drain. For bipolar transistors, the
movement of charges across PN junctions controls the output.


Here we go again... for bipolar transistors, it is the application of
*voltage* to the base emitter PN junction that controls the output
current. The movement of charges is irrelevant.


Who cares? The point was that with a bipolar transistor, one needs
current into the base in order to pass current from collector to
emitter. It doesn't work without the current. This is a useful fact
which can often be exploited in circuits that simply want an on/off
switch.

I actually liked your description on this point. It stated the facts
without implying that base current controlled collector current. It was
only your later statement that I had the issue with.

Whether it's 'right' is yet another matter. Newtonian physics is
'wrong', and based on incorrect physics, but for most things, it's OK
to use.

I wouldnt say that Newtonian physics is 'wrong'. Its more of an
approximation. The essentials of Newtonian physics still as correct
today as ever.

This is also true of design using beta. Lighten up.

My point here is to avoid perpetuating common myths concerning the
bipolar transistor that invariable leads to much confusion. Its better
to nip some things in the bud.

Kevin Aylward
salesEXTRACT@anasoft.co.uk
http://www.anasoft.co.uk
SuperSpice, a very affordable Mixed-Mode
Windows Simulator with Schematic Capture,
Waveform Display, FFT's and Filter Design.

 

[Kevin Aylward]

John Fields wrote:

On Sat, 08 Jan 2005 10:13:28 GMT, "Kevin Aylward"
salesEXTRACT@anasoft.co.uk> wrote:


Here we go again... for bipolar transistors, it is the application of
*voltage* to the base emitter PN junction that controls the output
current. The movement of charges is irrelevant.

---
The application of a forward voltage to the base-emitter junction of a
bipolar transistor will, of course, cause charge to flow between the
collector and emitter, but the movement of charges across the
base-emitter junction _is_ relevant,

No it isn't in the context of this question.

since without that movement there
can be no collector current.

It is not relevent in terms of *control* of the collector current. It is
an effect *caused* by Vbe. Whatever base current exist is besides the
point and is all in the wash. Hint:

Ie = Is.(exp(Vbe/Vt) - 1)

Where does base current current appear in this first order description?

If base current was relevant to *control* of the emitter/collector
current, it would surly appear in the first order description.

Kevin Aylward
salesEXTRACT@anasoft.co.uk
http://www.anasoft.co.uk
SuperSpice, a very affordable Mixed-Mode
Windows Simulator with Schematic Capture,
Waveform Display, FFT's and Filter Design.

 

[Miles Harris]

On Sun, 09 Jan 2005 00:24:03 -0800, Robert Monsen
<rcsurname@comcast.net> wrote:

More or less. That is why you can't just replace MOSFETs with bipolar
transistors. The bipolar transistor needs current into their base to
operate, and mosfet circuits will not be designed to supply it.

Yet another oversimplification. The bias requirements are *totally*
different and should not be studied by means of comparison with BJTs.
As Kevin Aylward said, it's better to nip these misconceptions in the
bud before they become entrenched views.

 

[Kevin Aylward]

Miles Harris wrote:

On Sun, 09 Jan 2005 01:31:39 -0600, John Fields
jfields@austininstruments.com> wrote:

and a fet will do
a simmilar thing only doesn't require a current ? (at whichever
terminal corresponds to a base on a bipolar)

---
Yes, but it still requires current to charge the gate capacitance.
However, once that capacitor is charged up, current can flow through
the drain-to-source channel with no further current required into the
gate.

Now the OP will be confused by another over-simplification. It depends
on whether the FET is of the enhancement or depletion mode type. Your
statement is correct for enhancement mode FETs, but wrong for
depletion mode ones. Depletion mode FETs are 'normally on' and will
conduct fully with *no* applied gate voltage. You have to apply a
*negative* voltage to the gate to moderate the drain current. Enough
negative voltage will cut-off the drain current altogether. No doubt
*you* know this, but it should be pointed out to the OP.

Well, its a bit more subtle

Depletion and enhancement both work exactly the same way, which is what
you actually said, but not so obviously. Increasing the gate voltage
will increase the current in both type of devices. The essential
difference is that at OV an enhancement device is off, where as a
depletion device needs a negative voltage to get it off. That is, its
*only* Vto that is different.

Kevin Aylward
salesEXTRACT@anasoft.co.uk
http://www.anasoft.co.uk
SuperSpice, a very affordable Mixed-Mode
Windows Simulator with Schematic Capture,
Waveform Display, FFT's and Filter Design.

 

[Miles Harris]

On Sun, 09 Jan 2005 14:54:03 -0600, John Fields
<jfields@austininstruments.com> wrote:

Why even bring that up?

In view of the fact that you brought up gate capacitance, it was
entirely apposite, IMV.

it's not germane to the discussion and it's
not true. Usable output may be obtained, depending on the
application, if the gate capacitance is only partially charged and
discharged.

The voltage gain of a FET is lousy compared to a BJT at the best of
times. If you're only going to paritally charge/discharge Cg., then
you'll only exacerbate that failing and be lucky to get any useful
gain at all.

 

[Miles Harris]

On Sun, 09 Jan 2005 12:35:42 -0800, Robert Monsen
<rcsurname@comcast.net> wrote:

Also, the characteristics of these devices is completely different. A
BJT has a current gain which is exponential in voltage (or somewhat
linear in current), whereas current through a FET has a quadratic
relationship to gate voltage in the active region.

More like hyperbolic, actually.

 

[Miles Harris]

On Sun, 09 Jan 2005 15:18:24 -0600, John Fields
<jfields@austininstruments.com> wrote:

Well, Kevin seems to subscribe to the policy that one should learn to
run, then walk. That is, first throw Ebers-Moll at a newbie then,
later, beta. If you agree with that philosophy, then I'll have to
disagree with you both.

Then presumably you're the type of person who tells his kids Father
Christmas exists. All very nice and well-intentioned, but not fair on
the child when he finds out the real truth and starts to question
everything he's ever been told. I'd sooner be straight with people
right from the start.

 

[Robert Monsen]

Miles Harris wrote:

On Sun, 09 Jan 2005 12:35:42 -0800, Robert Monsen
rcsurname@comcast.net> wrote:


Also, the characteristics of these devices is completely different. A
BJT has a current gain which is exponential in voltage (or somewhat
linear in current), whereas current through a FET has a quadratic
relationship to gate voltage in the active region.


More like hyperbolic, actually.

Really? What formula do you use to predict Id? The one I use for the
active region is

Id = k * (Vgs - Vgs(th))^2

where

k = Ids(on) / (Vgs(on) - Vgs(th))^2

Since k and Vgs(th) are constant for a given FET, Id is quadratic in
Vgs, right?

--
Regards,
Robert Monsen

"Your Highness, I have no need of this hypothesis."
- Pierre Laplace (1749-1827), to Napoleon,
on why his works on celestial mechanics make no mention of God.

 

[John Fields]

On Mon, 10 Jan 2005 00:01:32 GMT, Miles Harris <mazzer@yahoo.com>
wrote:

On Sun, 09 Jan 2005 14:54:03 -0600, John Fields
jfields@austininstruments.com> wrote:

Why even bring that up?

In view of the fact that you brought up gate capacitance, it was
entirely apposite, IMV.

---
I brought up gate capacitance to illustrate that there's no ohmic
contact between the gate and the channel, not to change the direction
of the thread for the pupose of showing off.
---

it's not germane to the discussion and it's
not true. Usable output may be obtained, depending on the
application, if the gate capacitance is only partially charged and
discharged.

The voltage gain of a FET is lousy compared to a BJT at the best of
times. If you're only going to paritally charge/discharge Cg., then
you'll only exacerbate that failing and be lucky to get any useful
gain at all.

---
Well, that really depends on what's meant by 'partially' and what's
meant by 'useful', wouldn't you agree?

--
John Fields

 

[Kevin Aylward]

I dont see johns orginal post here

On Sun, 09 Jan 2005 15:18:24 -0600, John Fields
jfields@austininstruments.com> wrote:

Well, Kevin seems to subscribe to the policy that one should learn to
run, then walk.

I most certainly dont.

That is, first throw Ebers-Moll at a newbie then,
later, beta.

Not at all. One need only state that the collector current is a direct
function of base emitter voltage, and that when this voltage is applied,
there is some base current, which is typically much less than the
collector current.

This correct description is no more complicated that giving the *wrong*
base current controlled one.

If you agree with that philosophy, then I'll have to
disagree with you both.

I don't agree with a philosophy of giving false technical information,
if the correct information is just as easy to give.

Then presumably you're the type of person who tells his kids Father
Christmas exists. All very nice and well-intentioned, but not fair on
the child when he finds out the real truth and starts to question
everything he's ever been told. I'd sooner be straight with people
right from the start.

Yep. It is very rare that I think white lies are the way to go. In this
particular case, kids should informed from the outset that ideas such as
santa claus, god, elves, etc are simply made up fantasies.

Kevin Aylward
salesEXTRACT@anasoft.co.uk
http://www.anasoft.co.uk
SuperSpice, a very affordable Mixed-Mode
Windows Simulator with Schematic Capture,
Waveform Display, FFT's and Filter Design.

 

[Miles Harris]

On Sun, 09 Jan 2005 19:13:40 -0600, John Fields
<jfields@austininstruments.com> wrote:

I brought up gate capacitance to illustrate that there's no ohmic
contact between the gate and the channel, not to change the direction
of the thread for the pupose of showing off.

It can't by any stretch of the imagination be described as "showing
off" since this is all pretty fundamental stuff - and fundamentals are
terribly important. Would you happily build a house on a flawed
foundation?

Well, that really depends on what's meant by 'partially' and what's
meant by 'useful', wouldn't you agree?

Nitpicking isn't going to help the OP.

 

[Miles Harris]

On Sun, 09 Jan 2005 17:05:58 -0800, Robert Monsen
<rcsurname@comcast.net> wrote:

If you actually measure, in the real world, a real FET's Vgs against
Id at several points up to Idss., you'll find the curve you get is
closer to hyperbolic than to quadratic. Your method may well be
arithmetically correct, but like so many things involving maths in
electronics, it's simply an approximation with an inevitable degree of
inaccuracy.

 

[Miles Harris]

On Sun, 09 Jan 2005 19:06:19 -0600, John Fields
<jfields@austininstruments.com> wrote:

Beta doesn't exist? That's news to me!

The beta model isn't even suitable for under-tens to study. Sooner or
later the OP will discover it falls apart in certain circumstances.
Then some kind soul will introduce him to Ebers-Moll and the
transconductance model and he will damn your hide for spinning him
such snake-oil early in his studies.

 

[John Fields]

On Mon, 10 Jan 2005 08:19:50 GMT, "Kevin Aylward"
<salesEXTRACT@anasoft.co.uk> wrote:

John Fields wrote:
On Sun, 09 Jan 2005 14:31:59 GMT, Miles Harris <mazzer@yahoo.com
wrote:

On Sat, 08 Jan 2005 11:11:22 -0600, John Fields
jfields@austininstruments.com> wrote:

The application of a forward voltage to the base-emitter junction
of a bipolar transistor will, of course, cause charge to flow
between the collector and emitter, but the movement of charges
across the base-emitter junction _is_ relevant, since without that
movement there can be no collector current.

Yes, but that's misleading. It's essential to concentrate on the
relationship between the applied voltage to the base/emitter junction
and the resultant collector current. The BJT is a transconductance
device and should be viewed as such.

---
It's only a transconductance device because of the voltage required to
force charge through the base-to-emitter diode,

No. They are a transconductance device because applying a voltage across
the base emitter junction injects carriers from the emitter to the base
*region*. This charge essentially *all* flows out of the collecter, not
the base terminal.

---
Not no. From:

http://searchsmallbizit.techtarget.com/sDefinition/0,,sid44_gci214200,00.html

"Transconductance is an expression of the performance of a bipolar
transistor or field-effect transistor (FET). In general, the larger
the transconductance figure for a device, the greater the gain
(amplification) it is capable of delivering, when all other factors
are held constant.

Formally, for a bipolar device, transconductance is defined as the
ratio of the change in collector current to the change in base voltage
over a defined, arbitrarily small interval on the
collector-current-versus-base-voltage curve. For an FET,
transconductance is the ratio of the change in drain current to the
change in gate voltage over a defined, arbitrarily small interval on
the drain-current-versus-gate-voltage curve.

The symbol for transconductance is gm. The unit is the siemens, the
same unit that is used for direct-current (DC) conductance.

If dI represents a change in collector or drain current caused by a
small change in base or gate voltage dE, then the transconductance is
approximately:

gm = dI / dE

As the size of the interval approaches zero -- that is, the change in
base or gate voltage becomes smaller and smaller -- the value of dI /
dE approaches the slope of a line tangent to the curve at a specific
point. The slope of this line represents the theoretical
transconductance of a bipolar transistor for a given base voltage and
collector current, or the theoretical transconductance of an FET for a
given gate voltage and drain current."

---

that charge changing
the electrical properties of the base material to more closely
approximate those of the collector and emitter. That is, when charge
is injected into the base-to-emitter diode of a PNP transistor, the
"N" type base material becomes more and more "P" like as more and more
current is forced through it, with the result that the transistor
starts looking more and more like a single piece of low-resistance "P"
type material as more and more current flows through the
base-to-emitter junction.

This is not an accurate description of the bipolar transistor. This
description is more relevant to operation of the mosfet. The npn
junction simply does not act like a slap of N type. If it did, base
current would be huge.

---
yes, were it not for the current limiting resistance external to the
base the base current could become huge. After all, the base-emitter
diode is just that, a forward biased diode operating on the far side
of the VI knee.

The intent, in both devices, is the same. That is to cause a
non-conductive region in a semiconductor to become conductive. In a
MOSFET it's accomplished by treating the channel like the plate of a
capacitor and making it _seem_ like it's composed of the same material
as the drain and the source by influencing the charge distribution in
it using the gate metalization as the other plate of the capacitor,
while in a BJT it's accomplished by forcing dynamic charge into the
base ["base region" if you like ] and using that charge flow to make
it seem like the base region material is becoming more and more like
the emitter and collector material as the base current increases.
---

That being the case, collector current will
flow when base current does, and will increase with increasing base
current until the transistor goes into saturation. Of course it's the
base-to-emitter voltage which makes the whole thing happen,

Indeed it is.

but what
_I_ think is misleading is to burden an inquirer with too much detail
too soon.

This is not too much detail at all. Its can't get any simpler. vbe
controls the collector/emitter current. End of story.

---
Hardly. Here this newbie asks "What makes a BJT different from a
FET?" and you reply "If you put a voltage across the base and emitter
terminals of a BJT current will flow between the collector and
emitter, while if you put a voltage across the gate and source
terminals of a FET current will flow between the drain and the
source." So, while your description may be true, its utter simplicity
leads the newb to think they're the same same thing with differently
named terminals.

Here is my original exchange with Skeleton Man:

<QUOTE>
On Sun, 09 Jan 2005 06:41:55 GMT, "Skeleton Man"
<invalid@guestwho.com> wrote:

so if I'm to understand correctly.. a bi-polar will pass current between
collector and emitter when a voltage is applied to the base ?

---
Essentially, yes. But, the voltage applied to the base must force
charge through the base-emitter junction before collector current can
flow.
---

and a fet will do
a simmilar thing only doesn't require a current ? (at whichever terminal
corresponds to a base on a bipolar)

---
Yes, but it still requires current to charge the gate capacitance.
However, once that capacitor is charged up, current can flow through
the drain-to-source channel with no further current required into the
gate.
<END QUOTE>


Do you have a problem with that?
---

Hence, initially describing the BJT in terms of beta and
leaving out the transconductance part alleviates the confusion which
will

No. No. No. It most certainly doesn't.

---
Yes. you're right. That was poorly stated. See my original reply,
above, to the OP for clarification.
---

Referring to the bipolar as "a current controlled device" causes never
ending confusion that is a bloody nightmare to correct. This is a case
in point. You yourself are trying to put forward the idea that that idea
has merit. It doesn't.

---
The problem which arises here, I think, is that the change in base
voltage required to affect a change in collector current is so tiny
that it becomes easier to consider what happens on the other side of
the change in base voltage. That is, the collector-to emitter current
change due to the base-to-emitter current change.
---

inevitably arise if the BJT and the FET are both described in
terms of transconductance.

Since this is the actual truth to the matter, this is what should be
said. Lying doesn't help one iota.

---
I agree. See my original reply, above, to the newb.
---

. After all, the question wasn't "How are
the BJT and the FET alike?" it was "How are they different?".

They are different, in part, in that the bipolar requires base current,
but that this base current is simply a nuisance.

---
Hardly a _nuisance_; it won't work without it.

Kind of like that we are different from corpses, in part, because we
are required to breathe, but that that breathing is simply a nuisance.

--
John Fields

 

[John Fields]

On Mon, 10 Jan 2005 16:30:18 GMT, Miles Harris <mazzer@yahoo.com>
wrote:

On Sun, 09 Jan 2005 19:13:40 -0600, John Fields
jfields@austininstruments.com> wrote:

I brought up gate capacitance to illustrate that there's no ohmic
contact between the gate and the channel, not to change the direction
of the thread for the pupose of showing off.

It can't by any stretch of the imagination be described as "showing
off" since this is all pretty fundamental stuff

---
Precisely! ;^)
---

- and fundamentals are
terribly important. Would you happily build a house on a flawed
foundation?

---
If I didn't know the foundation was flawed, of course.
---

Well, that really depends on what's meant by 'partially' and what's
meant by 'useful', wouldn't you agree?

Nitpicking isn't going to help the OP.

---
Never mind the newbie, the question was directed to you.
Do you agree or not?

--
John Fields

 

[John Fields]

On Mon, 10 Jan 2005 16:30:19 GMT, Miles Harris <mazzer@yahoo.com>
wrote:

On Sun, 09 Jan 2005 19:06:19 -0600, John Fields
jfields@austininstruments.com> wrote:

Beta doesn't exist? That's news to me!

The beta model isn't even suitable for under-tens to study. Sooner or
later the OP will discover it falls apart in certain circumstances.

---
And in others is eminently useful. In the real world, for instance,
what's important when driving, say, a relay is the beta available and
forcing that beta to a value which will never change regardless of
tghe environment into which the circuitry is placed.
---

Then some kind soul will introduce him to Ebers-Moll and the
transconductance model and he will damn your hide for spinning him
such snake-oil early in his studies.

---
You obviously have a reading comprehension problem if you _assume
that's what I did. Here is the exchange to which you seem to have
taken exception:


<QUOTE>
On Sun, 09 Jan 2005 06:41:55 GMT, "Skeleton Man"
<invalid@guestwho.com> wrote:

so if I'm to understand correctly.. a bi-polar will pass current between
collector and emitter when a voltage is applied to the base ?

---
Essentially, yes. But, the voltage applied to the base must force
charge through the base-emitter junction before collector current can
flow.
---

and a fet will do
a simmilar thing only doesn't require a current ? (at whichever terminal
corresponds to a base on a bipolar)

---
Yes, but it still requires current to charge the gate capacitance.
However, once that capacitor is charged up, current can flow through
the drain-to-source channel with no further current required into the
gate.
<END QUOTE>


Can you point to where I advocated the beta model for him to study?

--
John Fields

 

[Miles Harris]

On Mon, 10 Jan 2005 12:47:04 -0600, John Fields
<jfields@austininstruments.com> wrote:

If I didn't know the foundation was flawed, of course.

But you *do* know it's flawed, don't you - unlike the OP. You've
admitted as much to Kevin Aylward. And yet you're still prepared to
sell this flawed foundation to the OP!

Well, that really depends on what's meant by 'partially' and what's
meant by 'useful', wouldn't you agree?

Nitpicking isn't going to help the OP.

---
Never mind the newbie, the question was directed to you.
Do you agree or not?

It's a silly and pointless question so I'll decline, thanks.

 

[Miles Harris]

On Mon, 10 Jan 2005 12:39:48 -0600, John Fields
<jfields@austininstruments.com> wrote:

Not no. From:

http://searchsmallbizit.techtarget.com/sDefinition/0,,sid44_gci214200,00.html

"Transconductance is an expression of the performance of a bipolar
transistor or field-effect transistor (FET). In general, the larger
the transconductance figure for a device, the greater the gain
(amplification) it is capable of delivering, when all other factors
are held constant.

Formally, for a bipolar device, transconductance is defined as the
ratio of the change in collector current to the change in base voltage
over a defined, arbitrarily small interval on the
collector-current-versus-base-voltage curve. For an FET,
transconductance is the ratio of the change in drain current to the
change in gate voltage over a defined, arbitrarily small interval on
the drain-current-versus-gate-voltage curve.

The symbol for transconductance is gm. The unit is the siemens, the
same unit that is used for direct-current (DC) conductance.

If dI represents a change in collector or drain current caused by a
small change in base or gate voltage dE, then the transconductance is
approximately:

[snip]

What a lousy definition. Gate voltage and base voltage as referred to
above are totally misleading. The relevant values to concentrate on
are Vbe and Vgs; the potential difference applied directly across the
base/emitter junction (in the case of the BJT) and the PD applied
directly across the gate/source junction in the case of a FET.