Op-amps and Transistors

[Bill Bowden]

"Ingvar Esk" <ingvar.esk@ericsson.com> wrote in message news:<chpgd7$64d$1@newstree.wise.edt.ericsson.se>...

"Kevin Aylward" <salesEXTRACT@anasoft.co.uk> wrote in message
news:5cX%c.157$F73.138@fe2.news.blueyonder.co.uk...
Ingvar Esk wrote:
Depends how to interpret your questions. Litterally or trying to
understand what you "really mean".

"zalzon" <zalzonishappy@zalll.com> wrote in message
news:5rrqj0lj4q5nvf9k269oh1pv1kmk7m0r2o@4ax.com...
Without long explainations, is it fair to say :

1) A transistor uses change in voltage to control a large flow of
current.

No.

No. You are wrong. The orginal claim was *100%* correct. Its not
debatable.

Well, as you state in your paper "Despite much literature that implies other
wise..." I think it must be debatable, otherwise I can't see that "much
literature" would imply so


A transistor amplifies current (hFE).

Indeed it does as a side effect, however the transistor is absolutely
and fundamentally a *voltage* controlled and operated device
(http://www.anasoft.co.uk/EE/bipolardesign1/bipolardesign1.html).

Applying a *voltage* to the base emitter injects carriers into the base
region. These carriers are then swept up by the collectors accelerating
voltage. The fact that a few leak away through the base is just a
nuisance. Ideally, there would be no base current at all.

From my point of view. If I feed 1mA into the base the Ic would be some
hundreds (hFE) mA (if available), even though I know that hFE is not very
accurate. On the other hand if I supplied 0.1V to the base, nothing much
will happen at the collector. I can accept that it can be seen as a Voltage
amplifier, but only around its working point (Vb ~0.6-0.7V).

Ingvar Esk

The transistor base voltage varies very little over the range
of collector current. Like you say, the base voltage relative to the
emitter will usually be 600 to 700 mV, or maybe a bit more.

A general rule of thumb to calculate voltage gain is
40 times the difference of supply voltage to collector voltage.
So, if the transistor were biased such that the collector is 3
volts and the supply is 6 volts, you get a voltage gain
around (6-3)*40 = 120. So, for every millivolt change on the
base, the collector should change about 120 millivolts.

Doesn't much matter what the hFE is, other than it effects
the input impedance.

-Bill

 

[Animesh Maurya]

Hi Kevin,

This misconception is quit a crucial one, I would love to see one such
tutorial on your site covering this issue.

Most of the leading texts including; Electronics Principles by Albert
Paul Malvino,
Electronic Devices & Circuit Theory by Robert Boylestad claims
transistor as current controlled voltage source. I don't really
understand why they pass on this vague concept, is this for the sake
of the readers or even they are also misleaded(shouldn't say like
this)?

Yes, I can remember you once already clarified my misconception, but
there are many more guys to be……

Best regards,

AM


"Kevin Aylward" <salesEXTRACT@anasoft.co.uk> wrote in message news:<kc_%c.402$F73.256@fe2.news.blueyonder.co.uk>...

Ingvar Esk wrote:
"Kevin Aylward" <salesEXTRACT@anasoft.co.uk> wrote in message
news:5cX%c.157$F73.138@fe2.news.blueyonder.co.uk...
Ingvar Esk wrote:
Depends how to interpret your questions. Litterally or trying to
understand what you "really mean".

"zalzon" <zalzonishappy@zalll.com> wrote in message
news:5rrqj0lj4q5nvf9k269oh1pv1kmk7m0r2o@4ax.com...
Without long explainations, is it fair to say :

1) A transistor uses change in voltage to control a large flow of
current.

No.

No. You are wrong. The orginal claim was *100%* correct. Its not
debatable.

Well, as you state in your paper "Despite much literature that
implies other wise..." I think it must be debatable, otherwise I
can't see that "much literature" would imply so

Its not debatable in the sense that it is accepted by any physicists
that understands how transistors actually work. Its simply impossible to
derive and sensible device equations on the assumption that base current
controls emitter current. The most simple derived equation is:

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

Its called the diode equation. It is derived/shown in *any* standard
text book devoted to the device physics of the transistor. Its based on
applying a potential to the junction. There is no iffs or butts about
it.

Unfortunately there are rather a lot of non or semi technical books that
use explanations equivalent to water down a pipe. The sooner novices get
to grips with the fact that the ic=ib.hfe model, is a *grossly*
simplified model, with limited use, the better. It causes never ending
confusion.



A transistor amplifies current (hFE).

Indeed it does as a side effect, however the transistor is absolutely
and fundamentally a *voltage* controlled and operated device
(http://www.anasoft.co.uk/EE/bipolardesign1/bipolardesign1.html).

Applying a *voltage* to the base emitter injects carriers into the
base region. These carriers are then swept up by the collectors
accelerating voltage. The fact that a few leak away through the base
is just a nuisance. Ideally, there would be no base current at all.

From my point of view. If I feed 1mA into the base the Ic would be
some hundreds (hFE) mA (if available), even though I know that hFE is
not very accurate. On the other hand if I supplied 0.1V to the base,
nothing much will happen at the collector. I can accept that it can
be seen as a Voltage amplifier, but only around its working point (Vb
~0.6-0.7V).

You are looking at this from a way too naive point of view.

The transistor is a transconductance device. It outputs a current based
on its input voltage. It can be accurately exponential over 6 decades.
Sure, at low voltages, there is only a small current, but this doesn't
change how the transistor operates.

Its not a mater of what you want to accept. Its how it is. You can't
disagree with all the semiconductor physicista in the world, well not
unless your name is Einstein anyway-)

What I am explaining is *THE* *standard* *accepted* physics of the
situation. *Only* those *without* the academic background have
transistor operation mistaken. Go and have a read of *any* standard text
on semiconductor physics.

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.

 

[John Larkin]

On 10 Sep 2004 10:09:51 -0700, animesh_m@eudoramail.com (Animesh
Maurya) wrote:

Hi Kevin,

This misconception is quit a crucial one, I would love to see one such
tutorial on your site covering this issue.

Most of the leading texts including; Electronics Principles by Albert
Paul Malvino,
Electronic Devices & Circuit Theory by Robert Boylestad claims
transistor as current controlled voltage source.

The 'beta' model is a current-controlled current source. And it's all
you need for the majority of transistor applications. Hardly anybody
uses discrete transistor voltage amps any more, so transconductance
need not be considered in many cases, and then likely only in the
first stage.

John

 

[Ken]

On Tue, 07 Sep 2004 08:26:46 GMT, zalzon <zalzonishappy@zalll.com>
wrote:

1) A transistor uses change in voltage to control a large flow of
current.

Old "normal" transistors uses change in small current
to control a large flow of current.

There are many types of transistors today.
Some uses change in voltage to control a large flow of current.

 

[Kevin Aylward]

Ken wrote:

On Tue, 07 Sep 2004 08:26:46 GMT, zalzon <zalzonishappy@zalll.com
wrote:

1) A transistor uses change in voltage to control a large flow of
current.

Old "normal" transistors uses change in small current
to control a large flow of current.

Oh? and what transister would these have been? I know of no transisters
that are not voltage controlled.

There are many types of transistors today.
Some uses change in voltage to control a large flow of current.

They *all* use voltage to control current. Whether or not the current is
large is a matter of definitions. Incidentally, a flow of current is a
flow of flow of a substance, which is rather poor grammar. Current means
a flow of something already.

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.

 

[Ken]

On Mon, 13 Sep 2004 12:13:28 GMT, "Kevin Aylward"
<salesEXTRACT@anasoft.co.uk> wrote:

Old "normal" transistors uses change in small current
to control a large flow of current.

Oh? and what transister would these have been?
I know of no transisters that are not voltage controlled.

All normal transistors like 2N2222, BC547, BC650, BC141
and many, many more.
The big current flowing trough the transistor is controlled
by the current (not voltage). I know that because
I constructing many electronic circuits since the 60:s
and do so even today. Ther are curves in the data books
that tells the relationships, and that's I'm using.

 

[John S. Dyson]

In article <I9g1d.109$846.102@fe2.news.blueyonder.co.uk>,
"Kevin Aylward" <salesEXTRACT@anasoft.co.uk> writes:

Ken wrote:
On Tue, 07 Sep 2004 08:26:46 GMT, zalzon <zalzonishappy@zalll.com
wrote:

1) A transistor uses change in voltage to control a large flow of
current.

Old "normal" transistors uses change in small current
to control a large flow of current.


Oh? and what transister would these have been? I know of no transisters
that are not voltage controlled.

For the 'technologist' who wants a slightly better view of what transistors

do (or moreso, how they act), I suggest looking at Kevin's website. He
will probably reply more detailed URLs, but he has written a few 'tutorial'
documents that seem to be quite helpful.

IMO, it is VERY VERY important to avoid the Beta=Ic/Ib model as being
definitive. I do apparently disagree with Kevin that I believe that
the 'Beta' model can be useful for a few (VERY VERY few) purposes, while
Kevin seems to believe that it is useful for even fewer purposes .

I sure wish that people would take the physical model (the diode equations
along with Ebbers-Moll or Gummel-Poon) more seriously... They'll have
LOTS less trouble for non-trivial practical circuits.

In the cases where the Beta model is sufficient, the transistor isn't really
doing anything very useful .

The real problem with accepting the 'Beta' model as SOMETIMES being useful
is that too many people will believe that it is more useful than it really
is. In a nuanced fashion, the Beta model can be helpful, but it is only
better than absolutely nothing.

John

 

[John S. Dyson]

In article <kf3ck0lqmh8t7tqars6p97q981lle3e8ci@4ax.com>,
Ken <___ken3@telia.com> writes:

On Mon, 13 Sep 2004 12:13:28 GMT, "Kevin Aylward"
salesEXTRACT@anasoft.co.uk> wrote:

Old "normal" transistors uses change in small current
to control a large flow of current.

Oh? and what transister would these have been?
I know of no transisters that are not voltage controlled.

All normal transistors like 2N2222, BC547, BC650, BC141
and many, many more.

Yep, Bipolars are physically voltage controlled.

The big current flowing trough the transistor is controlled
by the current (not voltage).

Ic = (approx) Is (exp(Vbe/Vt) - 1);

(I might be slightly wrong, but the form of the controlling equation
is that of an exponentional.)

Essentially, the associated Ib is a kind of leakage current, and only
by happenstance is approximately linearly related to Ic. However, that
equation (Ic=B*Ib), isn't really a very useful relation except in VERY
limited conditions. The 'Beta' model really dies a horrible death in
dynamic conditions, but is fairly bad for anything except for estimating
the required Ib given a collector current... The proper (voltage controlled)
model actually works really well for HF applications, while the Beta model
just doesn't work for its limited purposes except for simple DC.

I know that because
I constructing many electronic circuits since the 60:s
and do so even today. Ther are curves in the data books
that tells the relationships, and that's I'm using.

One problem with Beta is that it an important parameter, but confuses

the non-engineer 'technologist' about it's actual limited physical
applicability. Beta is helpful for some things, but so are Cbc, Is,
etc.

If you want to judge the importance of 'Beta', think about it as equvalent
to the 'Q' of an inductor. It measures a necessary evil, but isn't really
a defining attribute.

John

 

[Kevin Aylward]

John S. Dyson wrote:

In article <I9g1d.109$846.102@fe2.news.blueyonder.co.uk>,
"Kevin Aylward" <salesEXTRACT@anasoft.co.uk> writes:
Ken wrote:
On Tue, 07 Sep 2004 08:26:46 GMT, zalzon <zalzonishappy@zalll.com
wrote:

1) A transistor uses change in voltage to control a large flow of
current.

Old "normal" transistors uses change in small current
to control a large flow of current.


Oh? and what transister would these have been? I know of no
transisters that are not voltage controlled.

For the 'technologist' who wants a slightly better view of what
transistors do (or moreso, how they act), I suggest looking at
Kevin's website. He will probably reply more detailed URLs, but he
has written a few 'tutorial' documents that seem to be quite helpful.

IMO, it is VERY VERY important to avoid the Beta=Ic/Ib model as being
definitive. I do apparently disagree with Kevin that I believe that
the 'Beta' model can be useful for a few (VERY VERY few) purposes,
while Kevin seems to believe that it is useful for even fewer
purposes .

Well, I wouldn't go that far;-) I am really just trying to correct this
incredible widespread misconception that surrounds basic transistor
operation. Its potential differances that make charge move.

Base current is obviously something that needs to be accounted for.
Right off the bat, if your doing a power amp, you have to work out the
drive current from Iload/Hfemin.


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]

Ken wrote:

On Mon, 13 Sep 2004 12:13:28 GMT, "Kevin Aylward"
salesEXTRACT@anasoft.co.uk> wrote:

Old "normal" transistors uses change in small current
to control a large flow of current.

Oh? and what transister would these have been?
I know of no transisters that are not voltage controlled.

All normal transistors like 2N2222, BC547, BC650, BC141
and many, many more.
The big current flowing trough the transistor is controlled
by the current (not voltage). I know ...
{snip}

Ho hum...Dont try and each your granpa to suck eggs.

*All* transisters are voltage controled. End of story.
http://www.anasoft.co.uk/EE/index.html

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.

 

[Ken]

On Tue, 14 Sep 2004 09:51:27 GMT, "Kevin Aylward"
<salesEXTRACT@anasoft.co.uk> wrote:

*All* transisters are voltage controled. End of story.
http://www.anasoft.co.uk/EE/index.html

Not according to your page either (Ib).

 

[Kevin Aylward]

Ken wrote:

On Tue, 14 Sep 2004 09:51:27 GMT, "Kevin Aylward"
salesEXTRACT@anasoft.co.uk> wrote:


*All* transisters are voltage controled. End of story.
http://www.anasoft.co.uk/EE/index.html


Not according to your page either (Ib).

Ho hummm again. What part of "controlled" do you have trouble with?
Look, dude, the first order equation for transistor operation, derived
in any elementary text book on semiconductor physics, is:

Ie = Io(T).(exp(Vbe.q/KT) - 1)

Ic ~= Ie

Now, prey tell me, where is the base current in this equation?

The fact that some base current flows dose not mean that the base
current is a *controlling* factor. You need to appreciate the difference
between causal relations and correlated relations.

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.

 

[John S. Dyson]

In article <SFJ1d.13213$_66.12055@fe1.news.blueyonder.co.uk>,
"Kevin Aylward" <salesEXTRACT@anasoft.co.uk> writes:

Ken wrote:
On Tue, 14 Sep 2004 09:51:27 GMT, "Kevin Aylward"
salesEXTRACT@anasoft.co.uk> wrote:


*All* transisters are voltage controled. End of story.
http://www.anasoft.co.uk/EE/index.html


Not according to your page either (Ib).

Ho hummm again. What part of "controlled" do you have trouble with?
Look, dude, the first order equation for transistor operation, derived
in any elementary text book on semiconductor physics, is:

Ie = Io(T).(exp(Vbe.q/KT) - 1)

Ic ~= Ie

Now, prey tell me, where is the base current in this equation?

The fact that some base current flows dose not mean that the base
current is a *controlling* factor. You need to appreciate the difference
between causal relations and correlated relations.

My guess is that beginners tend to see the approximately constant Beta as

implying that 'Ic = Beta * Ib' is the controlling relationship.

I like to show the uselessness of Beta by trying to simulate a transient
circuit (even with the intrinsic capacitances included.) The Beta
model is just wrong. (Well, not quite WRONG, but so insufficient as
to being wrong ).

John

 

[Roy McCammon]

Ken wrote:
?

I know of no transisters that are not voltage controlled.


All normal transistors like 2N2222, BC547, BC650, BC141
and many, many more.
The big current flowing trough the transistor is controlled
by the current (not voltage). I know that because
I constructing many electronic circuits since the 60:s
and do so even today. Ther are curves in the data books
that tells the relationships, and that's I'm using.

The transistors do what they do regardless of how
we model them. Current control is algebraically
easy and a good enough approximation for many
purposes. Voltage control is very accurate but
mathematically more difficult.


--
local optimization seldom leads to global optimization

my e-mail address is: <my first name> <my last name> AT mmm DOT com

 

[Kevin Aylward]

Roy McCammon wrote:

Ken wrote:
?
I know of no transisters that are not voltage controlled.


All normal transistors like 2N2222, BC547, BC650, BC141
and many, many more.
The big current flowing trough the transistor is controlled
by the current (not voltage). I know that because
I constructing many electronic circuits since the 60:s
and do so even today. Ther are curves in the data books
that tells the relationships, and that's I'm using.

The transistors do what they do regardless of how
we model them.

In the bigger and grand schemes of things, there may well be many models
for the same phenomena, but this is in a different context then is being
used here.

Current control is algebraically
easy and a good enough approximation for many
purposes. Voltage control is very accurate but
mathematically more difficult.

You still miss the point. The transistor is a voltage controlled derive,
irrespective of the engineering model. This is getting exasperating. Its
applying a *voltage* to the base and emitter that injects carriers into
the base, that are then swept up by the collector. The base current is
simply irrelevant. That is, the flow of base current does not *cause*
transistor action. Therefore, the transistor is *not* "controlled" by
base current. Its that simple!!!

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.

 

[me]

You still miss the point. The transistor is a voltage controlled derive,
irrespective of the engineering model. This is getting exasperating. Its
applying a *voltage* to the base and emitter that injects carriers into
the base, that are then swept up by the collector. The base current is
simply irrelevant. That is, the flow of base current does not *cause*
transistor action. Therefore, the transistor is *not* "controlled" by
base current. Its that simple!!!

NO, the base emitter voltage turns on the diode they form, the amount of
current that flows through the junction determines the collector current.
Because we are dealing with a diode junction we can use the relationship
between voltage and current in a diode to model with transconductance
equations. We also use transconductance model with voltage controlled
current devices like mosfets. BUT, mosfets have no beta/hfe because they
are voltage controlled devices...


-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----== Over 100,000 Newsgroups - 19 Different Servers! =-----

 

[Kevin Aylward]

me wrote:

You still miss the point. The transistor is a voltage controlled
derive, irrespective of the engineering model. This is getting
exasperating. Its applying a *voltage* to the base and emitter that
injects carriers into the base, that are then swept up by the
collector. The base current is simply irrelevant. That is, the flow
of base current does not *cause* transistor action. Therefore, the
transistor is *not* "controlled" by base current. Its that simple!!!


NO,

No, its yes. The transistor is voltage controlled. Period. End of story.

the base emitter voltage turns on the diode they form,

Yes. That is what I stated.

the amount
of current that flows through the junction determines the collector
current.

Yes. That is what I stated.

Because we are dealing with a diode junction we can use the
relationship between voltage and current in a diode to model with
transconductance equations.

Yes. That is what I stated.

We also use transconductance model with
voltage controlled current devices like mosfets. BUT, mosfets have
no beta/hfe because they are voltage controlled devices...

Yes. So whats your problem?

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.

 

[Roy McCammon]

Kevin Aylward wrote:

irrespective of the engineering model. This is getting exasperating. Its

It must be your destiny to feel that way.


--
local optimization seldom leads to global optimization

my e-mail address is: <my first name> <my last name> AT mmm DOT com

 

[Captain]

"Kevin Aylward" <salesEXTRACT@anasoft.co.uk> wrote in message
news:Rl_1d.1690$U04.558@fe1.news.blueyonder.co.uk...

Roy McCammon wrote:
Ken wrote:
?
I know of no transisters that are not voltage controlled.


All normal transistors like 2N2222, BC547, BC650, BC141
and many, many more.
The big current flowing trough the transistor is controlled
by the current (not voltage). I know that because
I constructing many electronic circuits since the 60:s
and do so even today. Ther are curves in the data books
that tells the relationships, and that's I'm using.

The transistors do what they do regardless of how
we model them.

In the bigger and grand schemes of things, there may well be many models
for the same phenomena, but this is in a different context then is being
used here.

Current control is algebraically
easy and a good enough approximation for many
purposes. Voltage control is very accurate but
mathematically more difficult.

You still miss the point. The transistor is a voltage controlled derive,
irrespective of the engineering model. This is getting exasperating. Its
applying a *voltage* to the base and emitter that injects carriers into
the base, that are then swept up by the collector. The base current is
simply irrelevant. That is, the flow of base current does not *cause*
transistor action. Therefore, the transistor is *not* "controlled" by
base current. Its that simple!!!

When you come right down to it, all currents are controlled, and generated,
by voltage. I think a certain Herr Ohm said something about this.

cap

 

[Kevin Aylward]

Captain wrote:

"Kevin Aylward" <salesEXTRACT@anasoft.co.uk> wrote in message
news:Rl_1d.1690$U04.558@fe1.news.blueyonder.co.uk...
Roy McCammon wrote:
Ken wrote:
?
I know of no transisters that are not voltage controlled.


All normal transistors like 2N2222, BC547, BC650, BC141
and many, many more.
The big current flowing trough the transistor is controlled
by the current (not voltage). I know that because
I constructing many electronic circuits since the 60:s
and do so even today. Ther are curves in the data books
that tells the relationships, and that's I'm using.

The transistors do what they do regardless of how
we model them.

In the bigger and grand schemes of things, there may well be many
models for the same phenomena, but this is in a different context
then is being used here.

Current control is algebraically
easy and a good enough approximation for many
purposes. Voltage control is very accurate but
mathematically more difficult.

You still miss the point. The transistor is a voltage controlled
derive, irrespective of the engineering model. This is getting
exasperating. Its applying a *voltage* to the base and emitter that
injects carriers into the base, that are then swept up by the
collector. The base current is simply irrelevant. That is, the flow
of base current does not *cause* transistor action. Therefore, the
transistor is *not* "controlled" by base current. Its that simple!!!

When you come right down to it, all currents are controlled, and
generated, by voltage.

Exactly. Fundermenatlly, there is a charge that produces a voltage
differance. This voltaege is usually referred to as an accelerating
voltage as it is this that accelerates other charges. The flow of charge
is an effect. However..., there is a somewhat subtle complication in
that all effects are due to an exchange of motion at a deeper level


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.