Electronic – How is the Early voltage defined

transistors

The Early voltage is a definition rather than a real existing and measurable quantity. It can be found using the slope of the output characteristics \$I_c = f(V_{ce})\$.

However, there are two basic alternatives: constant \$I_b\$ or constant \$V_{be}\$.

Question 1: Which of these two alternatives can be used to find the EARLY voltage?

  • (a) Constant \$I_b\$,
  • (b) Constant \$V_{be}\$, or
  • (c) Both?

Background: Some books use definition (a), some others use (b) and some do even not specify if (a) or (b).

Question 2: If (c) is correct – how can we theoretically prove that both alternatives give us the same (correct) value for the Early voltage?

EDIT:
For my opinion, my problem is not yet solved; therefore some further remarks:

I am aware that my question is not too relevant for circuit designers – and that is the reason I was looking for some corresponding – theoretically oriented – contributions from Universities.

Here are the results:
One lecture note from Berkeley Univ. defines the Early voltage VEA for IB=const., but in another contribution VBE is held constant.

Papers from Stanford Univ. are defining VEA for VBE=const.

Lecture notes from the Columbia Univ. define VEA also for VBE=const.

Even more confusing is a contribution from the Georgia Inst. of Tech.. Here, the Early voltage VEA is definde for IB=const. However, in the same paper this parameter is used in the equation y22=IC/(VEA+VCE). And – as we know, y22 is the output conductance for VBE=const.
Can somebody explain these contradictions?

Here are corresponding links:

Best Answer

I think you'll find that there is a small difference due to voltage-dependent reverse leakage of the C-B junction, but it's not significant under normal conditions. To a first approximation the leakage is independent of voltage (for Vcb > a few Vt).

If the transistor is being operated at very low collector current (nA) and high collector voltage the constant base current method will give you a measurably lower value for Va.

I believe that constant Vbe is a better representation of behavior in real situations such as current mirrors.