8
\$\begingroup\$

Would anyone know of any merits (or demerits) to the common emitter amplifier represented in Figure 1 in comparison to one of a regular pattern (Figure 2)? Thanks.

(The 'DS548' transistor in Figure 1 is just a BC548.)

(The diode in Figure 1 is just a 1N4004 power diode that's there in case someone connects the power source the wrong way round.)

Source: Funway into Electronics Volume 1

(Source: S. Voron, R. Tester and M. Middleton, Funway into Electronics Volume 1. Chullora, NSW: Dick Smith Electronics, 2008)

enter image description here

Figure 2 (Source: Electronics Tutorials - Common Emitter Amplifier)

\$\endgroup\$
6
  • \$\begingroup\$ Upper schematic is depended on transistor parameters, which varies. \$\endgroup\$ – user263983 May 4 at 11:13
  • 2
    \$\begingroup\$ There is nothing "irregular" about the circuit in Fig1, it is quite a common design and there are plenty of merits for doing it that way.. Fig1 and Fig2 are the exact same with the only difference being that Fig1 includes a feedback resistor. If you make R1 and R2 in Fig2 very large and make Re very small, add a feedback resistor and you have Fig1. What exactly is it that you want to know?. \$\endgroup\$ – Vinzent May 4 at 11:23
  • \$\begingroup\$ @user263983 the same can be said for the lower schematic. Both circuits only work when the "puzzle" of transistor parameters and resistor values is solved well enough. \$\endgroup\$ – Bimpelrekkie May 4 at 12:15
  • 1
    \$\begingroup\$ It stabilises the operating point over temperature. Provided Beta * Rc >> Rf (feedback) then the circuit is actually insensitive to beta. The circuit also stabilises against beta (whatever it happens to be) variations (process variation) for relatively small process variations. It is a very common configuration in RF circuits. \$\endgroup\$ – Peter Smith May 4 at 12:19
  • 2
    \$\begingroup\$ @jasen your comment is false \$\endgroup\$ – Tony Stewart EE75 May 4 at 12:27
7
\$\begingroup\$

The "simple" two-resistor bias circuit might be used where every component added to the parts list has a cost, or space is strictly limited. Comparing the simple-bias circuit against the 4-resistor bias circuit:

Advantages of simple circuit:

  • SIMPLE DC bias with small parts-count
  • shunt feedback resistor improves linearity
  • less likely to oscillate if reactive source and load is present

Disadvantages of simple circuit:

  • less stable DC operating point for different transistors
  • DC bias is less temperature-stable
  • low input impedance due to shunt feedback
  • shunt feedback reduces maximum gain possible. (for RF, may be a good thing)
  • shunt feedback reduces output-to-input isolation
\$\endgroup\$
2
  • \$\begingroup\$ IMO the "simple" two-resistor bias circuit is really simple. Actually, it is a one-resistor circuit since the collector resistor does not belong to the feedback network. \$\endgroup\$ – Circuit fantasist May 4 at 17:02
  • 3
    \$\begingroup\$ @Circuitfantasist Seems that many folks belittle this arrangement. Even in mass-produced products, it is sometimes used. The OP's circuit seems to be a radio-frequency receiver preamp...in this case oscillation might be a problem. The somewhat-small-value feedback resistor likely aids stability. It is too simple here - any decent preamp should have some input selectivity to slash monster broadcast-band signals. \$\endgroup\$ – glen_geek May 4 at 17:16
6
\$\begingroup\$

Descriptors

  • Common Emitter NPN
  • AC-coupled input and output for RF amplification
  • self-biased for DC
  • Negative-feedback high gain (30dB) voltage amplifier

Characteristics

  • low input impedance to approach matching antenna impedance for max power transfer (for VHF?)

  • output impedance slightly lower (~ 80% of Rc ) due to excess gain NFB (neg. FB)

  • high gain ~ Av=160~200 no load with Antenna source impedance

  • low Av gain sensitivity for AC for high hFE with wide variation (3:1)

    • increases in hFE raises Ic , lowers Vce which lowers Ibe, thus reducing the swing of Vce from Rcb negative feedback.
  • fairly constant power consumption 40 ~ 60 mW ( due to Vbe feedback for hFE >70)

  • Vce tolerant to wide hFE variation with output swing much less than 9V

  • better linearity than H bias for THD or IMD

\$\endgroup\$
1
  • 1
    \$\begingroup\$ IMO the second circuit is also DC self-biased by the mechanism of the emitter degeneration... \$\endgroup\$ – Circuit fantasist May 4 at 17:09
2
\$\begingroup\$

The Common Idea

The two figures are dual; they present two possible implementations of the same idea - negative feedback. They are transistor analogs of the op-amp inverting (transimpedance) and non-inverting amplifier. Actually, both are inverting since the transistor is driven from the side of the base but the similarity with op-amp circuits is in the way negative feedback is applied.

Implementations

Inverting configuration

In this case (Figure 1), in both transistor and op-amp circuits, the input voltage and the output (collector) voltage are applied "in parallel" through two resistors (R1 and R2) to the transistor input (base-emitter voltage) or op-amp inverting input.

R1 is the internal resistance of the input voltage source or additional resistor in series; R2 is the resistor connected between the collector (op-amp output) and the base (op-amp inverting input). A university professor (but not me:) would instructively say it is a voltage shunt negative feedback.

Operation. In Figure 1, when the input voltage increases, the base voltage tries to increase as well. But the transistor "senses" this change and begins decreasing its collector voltage to restore the base voltage. As a result, a "virtual ground" appears at the base; so the circuit has low input resistance. This is the well-known Miller effect.

Because of the virtual ground at the base, like in the op-amp inverting configuration, the AC gain is determined by the ratio R2/R1. When there is no R1, there is no feedback and the gain is maximum (gm*Vin).

"Non-inverting" configuration

In this case (Figure 2), in both transistor and op-amp circuits, the input voltage and the output (emitter) voltage are applied "in series" to the transistor input (base-emitter junction) or op-amp differential input. The university professor would say it is a current series negative feedback.

Operation. In Figure 2, when the input voltage increases, the base voltage tries to increase as well. But the transistor "senses" this base-emittet change and begins increasing its emitter voltage to restore the base-emitter difference. As a result, the emitter voltage follows the base voltage and the difference between them is almost zero (VF). The base current (almost) does not change... and a "virtual high resistance" appears between the base and emitter; so the circuit has high input resistance. This is another (dual) version of the Miller efect that is known as "bootstrapping".

Output. Let's denote the emitter resistor by R1 and collector resistor by R2 to make a comparison with Figure 1. If we used the voltage VR1 = Vin across the emitter resistor as an output, this would be an emitter follower. But we have connected another resistor in the collector... and use the voltage drop VR2 across it as an output. The two voltage drops are connected, like by an "electrical transmission" I. So VR2/R2 = VR1/R1 = I, VR2/VR1 = R2/R1 and VR2 = VR1.R2/R1 = Vin.R2/R1. Thus the output voltage VR2 can be less, equal or higher than VR1 (Vin)... and we use the latter case to obtain an amplifier. More precisely speaking, we use its complement Vc to Vcc since it is grounded.

Gain. It is interesting to explain why the gain is only R2/R1 but not R2/R1 + 1. The reason is that, in contrast to the op-amp non-inverting configuration, the input voltage is not included in the output one.

"Inverting" voltage divider. So, this is a special case of a "R1/R2 voltage divider" where the input voltage is applied only across R1 and the output voltage is taken across R2 while, in the classic "R2/R1 + 1 voltage divider", the input voltage is applied across the whole resistance R1 + R2 and the output voltage is taken across R1. This trick is implemented by the help of the negative feedback.

\$\endgroup\$

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.