Comments on Why Ib=const. for BJT output characteristics Ic=f(Vce)
Parent
Why Ib=const. for BJT output characteristics Ic=f(Vce)
Introduction: In some books and other technical papers (also from universities) it is - surprisingly - still claimed that the bipolar transistor (BJT) would be a current-controlled element. This is simply stated - without any explanation or proof (which I think is impossible).
But of course, the other representation (voltage-controlled) can also be found in many knowledge sources - just as it is considered in the SPICE models.
For me and also for many students this is an unsatisfactory situation. Therefore, I consider the distinction (current vs. voltage-control) to be very important to avoid contradictions between theory and practice. Because many circuits and observable effects can only be explained with voltage control Ic=f(Vbe).
As one argument pro current control often the output characteristic Ic=f(Vce) is referred to, where the base current Ib is considered as a fixed parameter (Ib=const). And - as a matter of fact: Although the collector current Ic is determined by the voltage Vbe, the characteristic curves Ic=f(Vce) are practically always given only for different (fixed) base currents Ib.
My Question: Why ?
EDIT: In response to Elliot Alderson`s comment I enclose a short contribution from the great Barrie Gilbert.
Post
My Answer:
There are only practical reasons for a representation with Ib=const. These characteristics are primarily used to determine the DC working point (Ic_o, Vce_o) - together with the working line (resistors Rc and Re).
Would it be helpful and useful to be able to read the corresponding voltage value Vbe_o? No - because due to the input characteristic curve (e-function) the curves for the most interesting values (0.65 V...0.7 V ) are very close together (and are subject to very large tolerances).
It is, therefore, common practice to use instead a standard value (0.7 Volt) for dimensioning, the exact value of which does not have a great influence on the collector current if the Re-stabilization is good. Furthermore, to calculate the base supply circuitry, the corresponding base current Ib_o must be found anyway. Therefore, for practical reasons, it makes sense to use from the very beginning the characteristics Ic=f(Vce) where the base current can be found immediately (parameter Ib=const).
But this design procedure cannot be used as a proof for a control of the collector current by the base current.
Question: Are there further arguments for publishing the output characteristics for Ib=const. and not for Vbe=const.
EDIT (response to coquelicot):
Perhaps you misunderstood my position.... certainly my fault.
I never have denied that Ib does exist. And I never have denied that there is a MODEL for current control. And this model works in some cases - however, there are many circuits (e.g. current mirror, RE-feedback...) and effects (tempco -2mV/K, EARLY effect,..) which cannot be explained with this model (because it does not reflect the physical reality).
I have many more examples: Try to explain the common-base stage or the cascode principle or the function od the long-tailed pair with the current-control feature.... Here, we need the transconductance model with gm=d(Ic)/d(Vbe)=Ic/VT.
This reality is given by the Ebers-Moll equations. It is quite simple: The emitter current is given by an exponential function Ie=f(Vbe) which then is split into two (exponential) currents Ic=f(Vbe) and Ib=f(Vbe). These three currents represent a cause-and-effect-relation. I am sure that you will agree up to this point.
Now comes the step which I consider as wrong:
Of course, we can divide the two equations for Ic and Ib (as you did) - and we get a relation between Ic and Ib - but this relation is a CORRELATION and must not be interpreted as a control function (Cause-and-effect in a physical sense). Of course, from the mathematical point of view, everything is correct. But the causality is lost because the voltage Vbe is the cause for BOTH currents. Do you think that 2 charged carriers injected into the base could release 500 carriers from the emitter (assuming B=250)?
Simple example: Voltage source V1 drives curents through two parallel resistors R1 (current I1) and R2 (current I2). From the current divider rule we get: I2=I1*(R1/R2).
From this relation, would you derive that the current I2 is "controlled/determined" by I1 ?
PS: It was not my intention to discuss again the question voltage vs. current control (this was discussed elsewhere several times). But your response gave me no other choice. The core of my contribution was only the question why the BJT output characteristics are given (in most books and data sheets) for Ib=const. and not for Vbe=const.. I have tried to find a possible answer and I am waiting for some comments to my answer.
1 comment thread