G7: Practical Circuits

Your radio runs on DC (direct current, electricity that flows steadily in one direction, like a battery puts out). But the wall outlet gives you AC (alternating current, electricity that rapidly reverses back and forth). A power supply is the circuit that converts that wall AC into the clean DC the radio needs. It does the job in a few stages, and this group walks through each one.

The transformer: changing the voltage

The first stage is usually a transformer. A transformer is two coils of wire wound near each other; it takes AC in at one voltage and hands AC back out at a higher or lower voltage. Wall power is around 120 volts, but your radio's electronics may want something quite different, so the transformer steps the voltage up or down to a useful level. (No actual current passes between the two coils, they are linked only by a magnetic field, which is why a transformer can also keep your radio safely separated from the wall wiring.)

The rectifier: flipping AC into DC

AC still goes back and forth, so next we need to force it to flow only one way. The part that does that is a rectifier, which is built from diodes. A diode is a one-way valve for electricity: it lets current pass in one direction and blocks it in the other. There are two common rectifier designs, and the test asks you to tell them apart:

• A half-wave rectifier is the simple kind: only one diode is required. Because a single diode blocks one whole half of each back-and-forth cycle, it passes only the half that flows the "right" way. So a half-wave rectifier converts 180 degrees of the AC cycle to DC. (A full back-and-forth cycle is 360 degrees, so "half the cycle" is 180 degrees, hence the name "half-wave.")
• A full-wave rectifier is cleverer: it flips both halves of the cycle so they all flow the right way, converting the entire 360 degrees of the AC cycle to DC. One common way to build it uses two diodes and a center-tapped transformer. (A "center-tapped" transformer simply has an extra wire connected to the middle of its output coil, which is what lets just two diodes catch both halves of the wave.)

What does the output look like before any smoothing? An unfiltered full-wave rectifier feeding a plain resistor puts out a series of DC pulses at twice the frequency of the AC input. That "twice the frequency" detail makes sense: because the full-wave rectifier flips both halves up, you get two humps per original cycle instead of one. (A half-wave rectifier, by contrast, gives one pulse per cycle, at the same frequency as the input.)

The filter: smoothing the bumps

Those rectified pulses are technically DC (they only flow one way now), but they are lumpy, full of ripple, not the smooth, steady DC a radio likes. A filter network smooths them out. The parts used in a power-supply filter are capacitors and inductors. A capacitor stores up charge and releases it during the dips, like a water tank that fills during a surge and feeds out during a lull, while an inductor resists sudden changes in current. Together they iron the bumps flat into smooth DC.

The bleeder resistor: a safety drain

Here is an important safety part. Those filter capacitors can hold a dangerous charge even after you unplug the radio, ready to give you a nasty shock. So power supplies include a bleeder resistor. Its function is simple and lifesaving: it discharges the filter capacitors when power is removed. It quietly bleeds the stored charge away so the capacitors are safe to touch. (It is not a fuse and has nothing to do with ground loops, those are tempting wrong answers; its one job is draining the capacitors.)

Linear versus switchmode supplies

There are two broad styles of power supply. The older linear design uses a big, heavy transformer running at the wall's frequency. The newer switchmode design (also called "switching") chops the power into a very fast on-off stream first. The key advantage the test wants: in a switchmode supply, the high-frequency operation allows the use of smaller components. Running at high frequency lets the transformer and other parts shrink dramatically, which is why a tiny laptop charger can put out a lot of power without weighing a ton.

Figure G7-1: reading schematic symbols

A few questions point at figure G7-1, a sheet of standard part drawings called schematic symbols. A schematic is a wiring map, and each part is drawn as a small standardized icon. You just match the numbered symbol to the part name. Here is what to recognize:

Symbol in figure G7-1	What it is
Symbol 1	A field effect transistor (FET) — a transistor controlled by a voltage on its "gate"
Symbol 2	An NPN junction transistor — the common bipolar transistor with the arrow pointing outward
Symbol 5	A Zener diode — a special diode used for voltage regulation, drawn with a bent "Z-shaped" bar
Symbol 6	A solid core transformer — two coils with solid bars between them showing the iron core
Symbol 7	A tapped inductor — a coil with an extra connection partway along it

You do not have to memorize how to draw these from scratch. On the test you are shown the figure and just pick which numbered symbol matches the named part. (A "transistor" is a tiny part that can amplify or switch signals; a "diode" is the one-way valve from earlier; an "inductor" is a coil of wire.)

Memory trick for the whole power supply: Transform, Rectify, Filter, Regulate. Change the voltage, flip it to one-way, smooth it out, and (often) hold it steady, with a bleeder resistor draining the capacitors for safety when you switch off.

This group covers three kinds of circuit families: digital logic (circuits that make simple yes/no decisions), amplifiers (circuits that make a signal stronger), and oscillators (circuits that generate a signal all by themselves).

Digital logic: gates, counters, and registers

Digital circuits work with just two states, usually called high and low (think "on/off" or "1/0"). The basic decision-making building block is a logic gate. The one the test names is the AND gate: a two-input AND gate gives a high output only when both inputs are high. Think of two switches wired in a line, the light only comes on if you flip both. (Other gates exist, like OR, which needs only one input high; but for the test, remember AND means "both.")

Stack gates together and you can count. A binary counter counts in the on/off number system. A handy fact: a counter with a certain number of "bits" (a bit is one on/off digit) can hold 2 raised to that many bits different values. So a 3-bit binary counter has 8 states (because 2 × 2 × 2 = 8). That doubling pattern, 1 bit gives 2, 2 bits give 4, 3 bits give 8, is worth remembering.

Another digital building block is the shift register. A shift register is a clocked array of circuits that passes data in steps along the array. Picture a bucket brigade: on each tick of the "clock" (a steady timing pulse), every stored value scoots one step down the line. It is how digital circuits move data along in an orderly, stepwise way.

Amplifiers: making signals stronger

An amplifier takes a weak signal and makes a stronger copy of it. Amplifiers are sorted into "classes" by how much of each cycle the amplifying device (the transistor or tube) is actually switched on, which trades off between faithfulness and efficiency.

• Class A is the most faithful: the device conducts 100% of the time (it stays on through the entire cycle). That makes it very clean and accurate, but the least efficient, because it is always drawing power.
• Class C is at the other extreme: the device is on for only a small slice of each cycle. That makes it the class with the highest efficiency, but it badly distorts the waveform, so it can only be used where distortion does not matter.

That brings up two related ideas:

• A linear amplifier is an amplifier in which the output preserves the input waveform, a faithful, undistorted copy, just bigger. You need linear amplification for modes like SSB voice, where the shape of the wave carries the information.
• So when is the distorting Class C acceptable? For FM. An FM (frequency modulation) signal keeps a constant strength and carries its information in frequency changes, not in the wave's shape, so the distortion from a Class C stage does not hurt it. That lets FM rigs enjoy Class C's high efficiency.

How do you measure how good an amplifier is at not wasting power? You measure its efficiency. For an RF (radio-frequency) power amplifier, efficiency is found by dividing the RF output power by the DC input power. In plain words: useful power out, divided by power fed in. A higher ratio means less energy wasted as heat.

One more amplifier word: neutralizing. Sometimes an amplifier misbehaves and starts generating its own signal instead of just amplifying, an unwanted howl called self-oscillation. The purpose of neutralizing an amplifier is to eliminate self-oscillations. It feeds back a small canceling signal to keep the amplifier from breaking into oscillation.

Oscillators: making a signal from nothing

An oscillator is a circuit that generates a steady signal all on its own, with no input signal to amplify. It is what produces the tone or carrier a radio needs. So what are the basic ingredients of a sine-wave oscillator? A filter and an amplifier operating in a feedback loop. The idea: the amplifier boosts a signal, a filter (which favors one frequency) sends part of the output back to the input in just the right way, and the loop sustains itself, generating a continuous wave. ("Feedback loop" means the output is routed back to the input.)

One very common type is the LC oscillator, which uses a "tank circuit" made of an inductor (L) and a capacitor (C). What sets the frequency it produces? The inductance and capacitance in the tank circuit. The values of that L and that C decide the pitch of the signal, the same way the size of a bell decides its note. (This is the same L-C resonance idea from G5; here it is being used on purpose to pick the oscillator's frequency.)

Memory tricks: AND means "both"; bits double the count (3 bits = 8); Class A is on "All" the time (100%) and clean; Class C is the efficiency Champ but distorts, so it is fine for FM; an oscillator is an amplifier plus a filter feeding itself.

This group is about how the boxes of a real radio fit together, the special parts that create and recover an SSB signal, the filters that pick out the signal you want, and how modern software-defined radios do much of this with computer code instead of hardware.

Making and recovering SSB

SSB (single sideband) is the main voice mode on the shortwave bands, and a few special circuits create and recover it:

• A balanced modulator is the circuit that mixes your voice with a carrier frequency, and its output is double-sideband modulated RF. The clever part is that it cancels out (balances away) the carrier itself, leaving just the two sidebands, the two mirror-image copies of your voice above and below the carrier.
• But SSB only wants one sideband, not both. The circuit used to select one of the sidebands from a balanced modulator is a filter. The filter passes the sideband you want and blocks the other one, turning double-sideband into single-sideband.
• On the receiving end, how do you turn that SSB signal back into sound? With a product detector. A product detector is used in a single sideband receiver to extract the modulated signal, that is, to recover the original voice from the SSB. (Picture it as the "decoder box" near the end of the receive chain.)

Matching the transmitter to its feed line

At the output of a transmitter you sometimes find an impedance matching transformer. "Impedance" is a circuit's opposition to AC, measured in ohms, and a transmitter works best when it sees a particular impedance (usually 50 ohms). One reason to use an impedance matching transformer at a transmitter output is to present the desired impedance to the transmitter and feed line. It acts like an adapter so the transmitter and the antenna's feed line are properly matched and power flows efficiently.

Filters: passing some frequencies, blocking others

A filter is a circuit that lets some frequencies through and blocks others. Several test terms describe a filter's behavior, and they are easy once you picture a filter's "shape":

• The passband is the range of frequencies the filter lets through. The cutoff frequency of a low-pass filter is the frequency above which the output power is less than half the input power. In other words, it is the edge where the filter starts seriously knocking the signal down (the half-power point).
• Insertion loss is the term for a filter's attenuation inside its passband. ("Attenuation" means weakening.) Even in the range it is supposed to pass, a real filter swallows a little signal, that small unavoidable loss is the insertion loss.
• Ultimate rejection is the term for a filter's maximum ability to reject signals outside its passband. It tells you how strongly the filter can stomp on the frequencies you do not want, far from the passband.
• For a band-pass filter (one that passes a band of frequencies in the middle and blocks everything above and below), its bandwidth is measured between the upper and lower half-power frequencies. Those are the two points, one on each side, where the signal has dropped to half power; the span between them is the bandwidth.

Digital and software-defined radio

Modern radios increasingly do this work with digital math instead of bulky analog parts. A few key ideas:

• A direct digital synthesizer (DDS) generates signals digitally. Its standout characteristic: it offers variable output frequency with the stability of a crystal oscillator. You can tune it smoothly anywhere, yet it stays rock-steady because a precise crystal sets its timing reference. (A crystal is a tiny slice of quartz that vibrates at an extremely steady rate.)
• A DSP (digital signal processing) filter, a filter done in math, has a big advantage over an analog one: a wide range of filter bandwidths and shapes can be created. Because it is just software, you can reshape the filter on the fly instead of being stuck with fixed hardware.
• A software-defined radio (SDR) performs much of the radio's work in software. In an SDR, all of these functions, demodulation, filtering, frequency selection, and more, are performed by software. SDRs commonly use I-Q modulation, which uses two versions of a signal that are 90 degrees apart in phase (the "in-phase" I signal and the "quadrature" Q signal, 90 degrees being a quarter of a cycle). The advantage of I-Q is sweeping: all types of modulation can be created with appropriate processing. One radio, in software, can do AM, FM, SSB, and digital modes, just by changing the math.

Receiver sensitivity

Finally, what makes a receiver good at hearing faint signals, its sensitivity? Several things affect it at once, so the answer is "all these choices are correct." Sensitivity depends on factors like the noise the receiver itself generates and how much gain its front end has, no single item is the whole story.

Memory tricks: balanced modulator makes double sideband and a filter trims it to single; product detector pulls voice back out of SSB; cutoff is the half-power frequency; insertion loss is loss inside the passband while ultimate rejection is blocking outside it; I and Q are 90 degrees apart and let software make any mode.