🚀 go-pugleaf

Roy McCammon wrote:
> 
> Duncan Barclay wrote:
> >
> > In article <76ja03$gud$1@news-2.news.gte.net>,
> >         Doug <do@not.spam> writes:
> > > Duncan Barclay wrote:
> > >>
> > >> This thread started with someone (Nightjar) asking how they worked.
> > >> Out of four replies I followed up on one thread was not asking how
> > >
> > > Getting back to NightJar ...
> > >
> > > BJTs can be thought of a CURRENT CONTROLLING DEVICES
> > > and valves can be thought of VOLTAGE CONTROLLING DEVICES.
> > >
> > > BJTs offer variations in the emitter-base CURRENT controlling
> > > variations in emitter-collector CURRENT.
> 
> > This is NOT correct.
> >
> > How does this explain the way a long-tailed pair works?
> >
> > A BJT allows you to control a current by varying the base emitter
> > voltage.
> >
> > Certain circuit configurations make it appear that you can control the
> > collector current by changing the base current.  But as this ratio
> > varies with temperature, device, process, batch, which direction the
> > circuit board is facing, it's not very useful.
> 
> I think Doug said "current controlling", not "current
> controlled"
> which I take to mean that the output of a bjt is modeled
> as a current output rather than a voltage output.  That
> would
> apply whether you viewed the bjt as a current (beta)
> amplifier
> or a transconductance (gm) amp.
> 
> I agree that when I am analyzing a circuit for gain, margin,
> or stability, I think of the bjt's as transconductance amps,
> but when I am just trying to figure out how the circuit
> works,
> I just consider bjt's as current amps (beta).

It ends up being whatever works for you. 

If you have anything that can be modeled as a 2 port 
device, you have several options to choose: 

1) VCVS = hybrid parameter g  
2) CCVS = transfer impedence parameter rm
3) VCCS = transfer admittance parameter gm 
4) CCCS = hybrid parameter h 

So whatever works for you, use it. 
If not, then obviously don't. 

So your question about admittance gm of a tube is 
right on the penny.  My fault, since what I suggested 
is merely *my way* of looking at them.  And I admit 
that I haven't worked with them in a while. 

I never thought that by suggesting a simple model for 
someone asking for help with the operation of valves 
or transistors that it would turn into what's been going 
on.  I'm sure after this, someone will be extremely 
hesitant to even suggest to a beginner the analogy of 
electric curent to water flow.  And I'm sure at this 
point, there's some beginners who might have a lot 
of either serious doubts or questions as a result. 
We've probably done more to muddy up the waters than 
clear it. 

But for what it's worth, I *think* I understand what's 
possibly the basis for all the misunderstanding.  

Simply my opinion, now, I'm not stating it as fact ... 

I was first taught the diode curve as a function of 
the *Activation Energy* of the material be it Si or 
Ge in terms of electron volts rather than an applied 
bias voltage.  At least back then, it was easier to 
understand some of the actual mechanics of what was 
happening.  It followed right in line with someone 
here suggesting the actual mechanism of a BJT is based 
on the work function of the material.  And that's 
good enough for me in terms or *understanding* BJTs. 
*Explaining* BJTs might require yet another model 
as has been suggested.  Take your pick. 

It would be clumsy to think of BJT circuit design 
completely in terms of electron volts (eVs), though.  
So, from that activation energy based equation, I 
think I remember we derived the bias voltage version 
of the equation.  The bias form is simply more convenient.  

It is not obvious to me that what I was taught (the electron 
volt version) as a way of teaching the fundementals of a BJT 
is really any better than any other way.  Maybe in the end 
that too is simply *another way* of explaining it, such as 
a solid stater would. I don't know, I'm not a solid state 
physicist and don't intend to impress people as such. 

It is extremely important to note that I am not suggesting 
any one form is any more *correct* than any other. 

I'll leave that to someone else. 

---------------------------------------- 
Return Address is dmckean'at'gte'dot'net 
----------------------------------------