CRAC Class C (2025) — Questions 1061–1070

A 10 kHz–60 MHz signal spans a huge relative bandwidth (several decades), so a tuned network cannot restore missing half-cycles — the amplifier must be linear by itself. Only 甲类 (Class A) is linear in a single-transistor stage across such a bandwidth. Classes B/C/AB need push-pull or tuning; Class D is switching (nonlinear). So only Class A fits.

FM (frequency modulation) is constant-amplitude, so amplitude linearity is irrelevant and any class can amplify it: Classes A, B, C, and D all work (efficiency aside). Therefore all four classes qualify, which is option A.

With K=5 the CW necessary bandwidth is defined and constant-amplitude keying is assumed; the official key lists only Classes A and B as the valid choices here (option A). Note that in practice Class C and switching Class D are also commonly used for CW carriers, but per this question's answer key the selected pair is A, B.

A small-signal stage must stay linear, so it uses 甲类 (Class A), whose device never cuts off and reproduces the signal cleanly. Classes B/C/D are power/switching stages, not small-signal pre-amps in this context.

大信号放大器 (large-signal/power amplifier) deliberately swings the device outside the linear region. Classes B, C, and D all operate with the device cutting off or switching, so they are large-signal classes (option A = B, C, D). Class A is the small- signal/linear class here and is excluded.

Two complementary (or identical) output transistors each amplifying one half-cycle form a 推挽放大电路 (push-pull amplifier). It runs with very low quiescent current (Class AB/Class B) and high efficiency while staying linear — that is option A. B/C/D misdescribe it as merely "double current/power" or as an impedance-transform trick, which is not the defining purpose.

The 共栅极 (common-grid, grounded-grid) tube stage has a low input impedance that is easy to broadband-match to the driver (A correct), and grounding the grid breaks the plate–cathode capacitance feedback path, greatly improving stability against VHF/UHF self-oscillation (B correct). C is wrong (voltage gain is not maximized — common-cathode gives more gain); D is wrong (it does not minimize distortion).

Paralleling two power transistors shares the current, giving roughly twice the output current and thus twice the output power for a given load (A correct). Parallel is NOT push-pull (B wrong — that needs split phases); it does not double voltage rating (C wrong); and distortions do not cancel (D wrong).

负反馈 (negative feedback) takes a sample of the output and returns it to the input in opposite phase, partially canceling the input (A correct) — this reduces gain but improves linearity/stability. B describes positive feedback (which causes oscillation); C/D misdescribe the signal path.

Negative feedback gives lower distortion and better stability at the cost of gain (A correct), lets you set input/output impedance for easy matching (B correct), and widens bandwidth for broadband use (C correct). D is false — feedback reduces harmonics, it does not increase them.