1
\$\begingroup\$

Some places mention an ideal transistor is unilateral. But is that the case in reality?

For example for a BJT we say that the base current or Vbe controls the output port such as emitter current or Vce not the otherway around. I guess this is what they call unilateral.

But in a common emitter amplifier if we change the load between Vcc and the collector wouldnt the base current change? Isnt that behaviour bilateral?

And is Early Effect is an example of bilateral behaviour?

\$\endgroup\$
  • \$\begingroup\$ Only an introductory text to BJTs would claim that, one example commonly used to describe a bjts is the en.wikipedia.org/wiki/… \$\endgroup\$ – sstobbe Oct 14 '18 at 16:20
  • \$\begingroup\$ In a common emitter amplifier changing the load connected to the collector changes the current through the emitter which changes the voltage at the emitter which changes the base-emitter voltage. \$\endgroup\$ – user253751 Oct 15 '18 at 4:36
  • \$\begingroup\$ Yes - the Early effect is an example for the described properties of a BJT. \$\endgroup\$ – LvW Oct 15 '18 at 6:45
  • \$\begingroup\$ @immibis Are you saying that the base current Ib is controlled by changing the load as well? Because it is related to Vbe. If so very strange I always thought Ib is fixed. Or is that because the change in Vbe is very small so we treat it like Ib is fixed? \$\endgroup\$ – cm64 Oct 15 '18 at 13:04
0
\$\begingroup\$

A real example of such non-ideal behavior is Miller effect. Changes in collector voltage are capacitively (via Ccb) coupled back to the base, reducing the base drive. By reducing the collector swing, the effect is reduced, however that’s not desirable. The cascode topology was developed to avoid that trade-off.

\$\endgroup\$
0
\$\begingroup\$

Older hybrid-PI models of bipolars, with 4 genuine parameters, included Hre, how the output change in voltage affected the input current.

This effect was on order of 0.0001, or 100 ppm.

As a very young engineer, with 8 or 10 years of teenage tinkering and FINALLY getting full (college) understanding of the HybridPI models, I realized Hre would be limiting my maximum gain, and I used cascodes to avoid that.

Modern datasheets no longer provide Hre.

Perhaps Early Voltage effects contain the Hre effect.

What "dimensions" are you using in your attempts to define "bi-lateral" parameters?

\$\endgroup\$
-1
\$\begingroup\$

A real transistor couples energy every which way. It generally only has power gain in the traditional "forward" directions, but it most certainly couples energy "backwards" -- this is a challenge for circuit designers who still work with individual transistors.

\$\endgroup\$
  • \$\begingroup\$ Can you elaborate on how this happens? \$\endgroup\$ – user253751 Oct 15 '18 at 4:37
  • \$\begingroup\$ Consider a bipolar device, with input signal on the base, and the output signal on the collector, just a few microns apart, with the base-collector junction being a reverse-biased junction. There will be some picoFarads of capacitance (for a nomal size, normal power (10mA, 20volt) bipolar transistor) between base and collector. Charge can and will flow both ways between base and collector. And MOSFETS have capacitance between the gate and the channel, that channel connecting to source and drain. Even if there is no channel formed (FET is biased OFF), there will still be gate-drain overlap C. \$\endgroup\$ – analogsystemsrf Oct 17 '18 at 5:21

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.