BJT Biasing and Stability Questions and Answers

Answer: A

The base-emitter drop changes by about two millivolts per degree, and that change matters in proportion to how small the emitter drop is: two millivolts against one volt is a fifth of a per cent, but against two hundred millivolts it is one per cent. Divider current, dissipation and resistor tempco are all beside the point. Whenever a fixed error competes with a designed quantity, making the designed quantity larger is the direct remedy.

Answer: B

Every volt across the emitter is a volt unavailable to the collector, so the output swing shrinks as the stabilisation improves. Divider current, resistor dissipation and the bypass capacitor are all consequences that can be dealt with by choosing components, whereas the supply voltage cannot be argued with. Identifying which constraint in a trade is fundamental and which are merely inconvenient is what decides where the compromise actually lands.

Answer: C

Returning the emitter resistor to a negative supply rather than to ground lets it drop several volts while the emitter still sits near earth, so the whole positive supply remains available for the collector to swing into. A bypass capacitor addresses the gain, not the headroom; a stiffer divider addresses beta dependence; and reducing the collector resistor reduces the gain. Spending a second supply to escape a compromise is exactly the trade integrated circuits make, which is why dual rails are so common there.

Answer: D

Absolute resistor values are poorly controlled on a chip while adjacent devices match closely, so a current mirrored from a reference is far more predictable than one set by a resistor value. Area and gain are genuine secondary benefits, and resistors certainly can be made, just not accurately. That integrated design rests on matching rather than on absolute values is the single most useful thing to know about why its circuits look so different from discrete ones.

Answer: A

A current source has a very high impedance, so the load no longer dominates and the device's own output resistance sets the gain - the intrinsic limit of a few thousand. Its impedance is high rather than low, the collector circuit plainly matters, and the gain depends on the current only through the transconductance, which cancels. Removing whichever element dominates a limit is the way to reach the limit beneath it, and this is the standard application of that idea.

Answer: B

The drop and the ratio follow directly:

\[ \begin{aligned} V_{E} &= 2\mathrm{m} \times 1000 = 2 \ \mathrm{V} \\ \frac{V_{E}}{V_{CC}} &= \frac{2}{12} \approx 17\% \end{aligned} \]

so about 17 per cent, a little above the usual tenth but perfectly reasonable where stability matters more than swing. Expressing a design choice as a fraction of the supply makes it comparable across circuits running on different rails. Framing the question as which quantity the change acts upon makes the answer follow without calculation.

Answer: C

The emitter voltage halves and the junction drop is unchanged:

\[ \begin{aligned} V_{E} &= 1 \ \mathrm{V} \\ V_{B} &= 1 + 0.7 = 1.7 \ \mathrm{V} \end{aligned} \]

giving 1.7 volts, so the divider must be reproportioned. The current is unchanged but the stabilisation is halved, which is the cost of the extra volt of headroom gained. Every such adjustment moves one quantity and one specification together.

Answer: D

The drop moves only about sixty millivolts per decade of current, so a single figure is accurate enough for bias work across the whole useful range. It is certainly not exact, is unrelated to where beta peaks, and has nothing to do with runaway. An approximation is defensible when the quantity it replaces varies far less than the tolerances of everything else in the calculation, and this is a textbook case. The parameter that varies and the parameter that does not are worth naming before any substitution is attempted.

Answer: A

Bias work tolerates the approximation easily, but a circuit whose accuracy depends on two junctions agreeing to within millivolts - a current mirror, a differential pair, a bandgap reference - needs the exponential law and the actual currents. Ordinary changes in current, supply or resistance do not invalidate it. Knowing where an approximation stops working is more valuable than knowing the approximation. Checking the units of each quantity catches a wrong formula before the arithmetic begins.

Answer: B

The reason is true but the assertion is false: the improved stability is paid for in supply voltage and hence in output swing, and beyond a point the stage cannot deliver the required output at all. A rule of thumb of about a tenth of the supply reflects that balance. A statement containing the word always is worth testing against the cost side of the trade, because most component values have one.