🚀 go-pugleaf

In article <59q9yQA3Bdj2EwN3@jmwa.demon.co.uk>,
	John Woodgate <jmw@jmwa.demon.co.uk> writes:
> <76jqau$ltt$1@computer.my.domain>, Duncan Barclay
> <dmlb@ragnet.demon.co.uk> wrote:
>>This is NOT correct.
> 
> For certain far-out values of 'correct'.
>>
>>How does this explain the way a long-tailed pair works?
> 
> You took me to task for confusing device physics with circuit
> configuration. So you can't fairly use that argument now. You had some
> sort of a case if you confined the matter to device physics, but that
> was not appropriate at the level of the enquirer. It was a Musoft answer
> - correct but utterly unhelpful.

[Aside, what is Musoft? I've never heard of the term.]

>>A BJT allows you to control a current by varying the base emitter
>>voltage.
>>
> In an extremely non-linear manner. If one adopts this concept
> exclusively, it appears that it is not possible to make a linear (non-
> grossly distorting) amplifier.

Use feedback, or apply the small signal approach.  The classic four
resistor CE amplifier is linearised by the negative feedback generated
by the emitter resistance. Emitter followers are the same, so are common
base.

>>Certain circuit configurations make it appear that you can control the
>>collector current by changing the base current.
> 
> It doesn't just 'appear' - you can and do.
> 
>>  But as this ratio
>>varies with temperature, device, process, batch, which direction the
>>circuit board is facing, it's not very useful.
> 
> So why is it quoted in (almost?) every BJT data sheet?

Because it gives an indication of how much current the base voltage
biasing network will need to supply. And how good the process is.
 
> It's one thing to be formally and academically correct. It is quite
> another to insist that no other approach has any value at all. 

I get het up about this issue because viewing the BJT as
  Ic = beta Ib
is prone to problems in the long run. A circut that may work for a
device with beta of 100 may not work when you put a transistor in which
has a beta of 10 or 400. Being aware that beta is a poorly controlled
parameter makes for more robust design. BJTs are wonderful in that they
are guaranteed to have a gain goverened by
  gm = Ic/Vt = Ic/26mV.
All one has to do is to get Ic right. If you want temperature stable
gain then make Ic proportional to temperature. This equation holds for
about 6decades of current for most transistors. The absolute range
depends on the size of the device.

Barrie Gilbert has a rule of thumb for transistor level design. He calls
it the 70%-30% rule where bias circuits take 70% of the effort to
design, the signal chain takes 30%.

Duncan

-- 
________________________________________________________________________
Duncan Barclay          | God smiles upon the little children,
dmlb@ragnet.demon.co.uk | the alcoholics, and the permanently stoned.
________________________________________________________________________