Chapter 29: Measurements

Its coil’s magnetic field pushing against the magnet’s field

Current in the coil sets up a magnetic field that acts against the field of the U-shaped permanent magnet, so the coil pivots on its bearings. Heating drives other kinds of meter, not this one.

The moving-coil meter

A moving-coil meter reads current directly, more with a shunt, and voltage with a series multiplier. An oscilloscope shows voltage against time, but it is not the basic meter for current.

Its full-scale current and its internal resistance

RS = RM ÷ (n − 1) needs the movement’s resistance RM, and its full-scale current to find the scale factor n. The resistance of the circuit being measured does not enter the sum.

In parallel with the meter movement

The shunt goes across the movement, so most of the current flows past it and only a small part through it. A resistor across the circuit being tested would only add to the current being measured.

In series with the lamp, so that its current flows through the meter

An ammeter must carry the current it measures, so it goes in series with the lamp. Connected across the lamp, its very low resistance would short the lamp out and carry far too much current.

So that it barely changes the current it is measuring

An ammeter is in series, so any resistance it has adds to the circuit’s and lowers the current it measures. Drawing little current is the voltmeter’s reason for a high resistance.

30 Ω

n = 4 mA ÷ 1 mA = 4, so RS = RM ÷ (n − 1) = 90 ÷ 3 = 30 Ω. Dividing by n gives 22,5 Ω, which takes too much current, so the meter reads low.

4 mA

The shunt carries the current the movement cannot: 5 − 1 = 4 mA, while 1 mA flows in the movement. The whole 5 mA flows only in the leads to the meter.

100 Ω

n = 1 mA ÷ 200 µA = 5, so RS = 390 ÷ (5 − 1) = 97,5 Ω, nearest 100 Ω. 80 Ω is close to 390 ÷ 5 = 78 Ω: n used instead of n − 1.

Milliammeters and microammeters

Ammeters for small currents are milliammeters (thousandths of an ampere) and microammeters (millionths). A megohmmeter measures very high resistance, such as insulation, not current.

With a multiplier resistor in series with it

A multiplier in series drops the voltage the movement cannot take, so full-scale current flows at the chosen voltage. Connected straight across a circuit, the bare movement would carry far more than its full-scale current.

A series resistor of about 125 MΩ

R = V ÷ I = 2 500 V ÷ 20 µA = 125 MΩ in series to drop the voltage; the movement’s resistance is negligible. 125 kΩ takes µA for mA; a shunt only extends the current range.

Across the resistor, in parallel with it

A voltmeter reads the potential difference between two points, so it goes across the resistor, in parallel. In series its high resistance would be added to the circuit and starve it of current.

So that it takes very little current and barely upsets the circuit

In parallel, a voltmeter takes some current from the circuit; the higher its resistance, the less it loads the circuit and the truer the reading. Carrying the circuit’s current is an ammeter’s job.

It sees a much higher input resistance, so it is loaded less

A FET input draws almost no current, so the meter shows the circuit an input resistance of many megohms and barely loads it. A meter drawing more current would load it more, not less.

4 V

With the meter in parallel, the lower arm is 100 kΩ and 100 kΩ in parallel = 50 kΩ, so it reads 12 × 50 ÷ 150 = 4 V. 6 V is the voltage without the meter: loading ignored.

1,5 kΩ

The multiplier must drop the 2 − 0,5 = 1,5 V the movement cannot take, at the full-scale 1 mA: R = 1,5 V ÷ 1 mA = 1,5 kΩ. 2 kΩ forgets the movement’s own 0,5 V.

5 kΩ per volt

Each volt of range needs R = 1 V ÷ 200 µA = 5 kΩ, so the rating is 1 ÷ full-scale current = 5 kΩ per volt. 200 kΩ per volt would need a far more sensitive 5 µA movement.

It responds to the rectified average; its scale assumes a sine

Simple rectification makes the movement respond to the average, and the scale is marked so that a pure sine reads its RMS value; other waveforms have another average-to-RMS ratio. A true RMS response would suit any waveform.

It is a DC instrument: AC would push its pointer each way in turn

A moving-coil movement is designed for DC: its turning force follows the direction of the current, so on AC the pointer barely moves. AC flows through the coil easily; the pointer simply cannot follow it.

Voltage, current and resistance

A multimeter measures voltage, current and resistance, chosen with its mode switch; some models add capacitance and more. Capacitance is an extra, and current is never left out.

Too little deflection to read

1,5 V is only 0,3 % of a 500 V range, so the pointer barely moves and cannot be read. Overload damage is the risk of the opposite mistake, a range far too low.

The overload may damage the meter

50 V is twenty times the 2,5 V full scale, so far more current than the movement can take flows through it. Too little deflection is what a range far too high gives.

Test continuity without looking at the meter

The buzzer sounds when the probes find an unbroken path, so you can check continuity with your eyes on the circuit. It is not an overload alarm.

The full-scale range for each reading

Modern meters select the appropriate scale automatically, but the user still chooses the function with the mode switch. On older meters both are set by hand.

5,07 MHz

Frequency = count ÷ gate time = 50 700 ÷ 0,01 s = 5,07 MHz. Taking the count itself as the frequency, as if the gate were open for 1 s, gives 50,7 kHz, a hundred times too low.

100 Hz

One count more or less in 10 ms is a step of 1 ÷ 0,01 s = 100 Hz, so a longer gate gives finer resolution. 0,01 Hz mistakes the gate time itself for the step.

Its internal reference oscillator

The reference oscillator times the counting period, so any error in it appears in every reading, however many digits are shown. More digits give finer resolution, not more accuracy.

Several percent

An LC oscillator is far less stable than a crystal, so a counter timed by one may be several percent out; with a crystal it can be very accurate. Counting is exact only if the counting time is.

400 kHz

3 % of 14,2 MHz is 0,03 × 14 200 kHz = 426 kHz, nearest 400 kHz: enough to put you outside a band. 40 kHz slips a decimal place.

Checking the SWR on a feedline

An SWR meter is also called a reflectometer because it measures reflected power, which with the forward power gives the SWR. The manual prefers “SWR meter”, since the word also names tools in optics.

The forward power and the reflected power

An SWR meter measures the power going out towards the antenna and the power coming back, and works out the SWR from their ratio. The DC input power says nothing about the match.

90 W

The antenna takes what goes forward less what comes back: 100 − 10 = 90 W. Reading the forward figure alone, 100 W, ignores the power reflected at the mismatch.

Speech peaks sit far above its average, so the two readings differ

The human voice has peaks far above its average, and an SSB or AM signal follows it, so peak and average readings differ greatly. An average reading is lower than the peak, never higher.

FM

FM varies the frequency, not the amplitude, so the output stays constant whatever the modulating signal. AM, SSB and DSB output all rise and fall with the voice.

2:1

SWR is the ratio of the line impedance to the load, written as 1 or more: 50 ÷ 25 = 2:1. 0,5:1 has the ratio upside down; an SWR is never below 1:1.

It takes the strong signal picked up by the antenna as reflected power

The analyser’s own oscillator is very weak, so strong signals coming down the feedline look to it like reflected power; its frequency is not pulled. The manual suggests more power, or a selective voltmeter that rejects them.

Its built-in oscillator is so weak that this is allowed

The analyser’s internal oscillator puts out so little power that the manual calls its use legal outside the amateur allocations. Connected to an antenna, its weak signal is radiated, just very little of it.

500 Hz

Period = divisions × time per division = 4 × 0,5 ms = 2 ms, and f = 1 ÷ t = 1 ÷ 0,002 s = 500 Hz. 2 kHz leaves out the 4 divisions.

4,24 V

6 divisions × 2 V = 12 V peak-to-peak, so the peak is 6 V and VRMS = 6 ÷ 1,414 = 4,24 V. 8,49 V forgets to halve the peak-to-peak value first.

9 V

The peak is 2,6 × 5 = 13 V, and VRMS = 13 ÷ 1,414 = 9,19 V, nearest 9 V. Taking the 13 V peak as the RMS value is the usual slip.

Sweeps the dot across steadily, then returns it very quickly

The time-base drives the X plates with a smoothly rising voltage, sweeping the dot from left to right at the chosen rate before it returns quickly. Up and down movement comes from the input, through the Y amplifier.

Its voltage

The Y plates are driven by the input voltage, through the Y amplifier, so the height of the trace is voltage, calibrated in V/div. A current shows only as the voltage it drops across a resistor.

The oscilloscope

An oscilloscope has a horizontal (X) channel driven by the time-base and a vertical (Y) amplifier for the input signal. A multimeter’s pointer moves one way only and needs no such channels.

A trigger circuit starts each trace at the same point of the wave

The trigger circuit starts each sweep when the input reaches a set voltage, so every cycle is drawn over the last and the picture appears still. The glow of the screen fades within moments.

A monitor oscilloscope

A monitor oscilloscope shows the transmitted waveform itself, so poor keying or flat-topping can be seen; a field strength meter shows only that RF is present. A spectrum analyser, not offered here, does better still.

The final amplifier’s output, by a small pick-up loop

The envelope exists only once the RF is modulated, so a small loop held near the final tuned circuit feeds a sample to the vertical plates. The driver’s RF carries no modulation yet.

A pointer meter is far too slow to follow such rapid changes

A meter’s pointer has too much inertia to keep up with fast-changing signals, while an oscilloscope draws the waveform itself against time. RF does not burn out a meter; the pointer simply cannot follow it.

A microprocessor digitises the signal for display

A digital oscilloscope samples the input with a microprocessor and shows it on a screen, or on a computer or phone, with controls much like the old type. It shows any waveform, not only digital ones.

A crystal oscillator whose harmonics fall at known, even steps

A marker generator is a crystal oscillator designed to produce harmonics, giving “markers” at fixed steps such as every 100 kHz across HF. A swept oscillator has no fixed, accurately known frequencies to mark anything by.

So that its markers fall on accurately known frequencies

A crystal holds its frequency very accurately, so every harmonic lands on a known frequency and can be used to check a receiver’s dial. A crystal oscillator is fixed in frequency; it cannot be tuned across a band.

230 kHz

Two 100 kHz markers make 200 kHz and three 10 kHz markers 30 kHz more: 230 kHz above 14 MHz, at 14,23 MHz. Swapping the two counts gives 320 kHz.

7,100 MHz

Markers fall on every multiple of 100 kHz: 7 100 kHz is the 71st harmonic. 7,150 MHz lies halfway between two markers and would need markers every 50 kHz.

Modern transceivers almost all have accurate digital frequency readouts

Almost all modern transceivers show their frequency accurately on a digital readout, and frequency counters are freely available, so markers are seldom needed. Their signals are as easy to hear as ever, and no law bans them.

A tunable RF oscillator: a small signal generator

A dip meter is a variable-frequency LC oscillator with its coil outside the case, so it is a small RF signal generator. It sends out RF; it does not receive like a radio.

Valve versions showed a dip in their grid current

In old valve dip meters the grid current fell when a nearby tuned circuit drew energy from the oscillator, hence the name. The anode current does the opposite: it rises at the dip.

The frequency at which a tuned circuit or antenna resonates

A dip meter finds the frequency at which a tuned circuit or antenna resonates: its reading dips where the circuit takes energy from its oscillator. SWR at the antenna is a job for an antenna analyser.

7 MHz

f = 1 ÷ (2π√LC) = 1 ÷ (2π × √(2,2 × 10⁻⁶ × 220 × 10⁻¹²)) = 1 ÷ (2π × 2,2 × 10⁻⁸) = 7,23 MHz, nearest 7 MHz. Leaving out the 2π gives 45 MHz.

It is not wired in at all: its coil is held close to the circuit’s coil

The dip meter’s plug-in coil sits outside its case so that it can be brought near the circuit, and energy couples across magnetically. Wiring it in would load and detune its oscillator.

A resistor that absorbs a transmitter’s output without radiating it

A dummy load is a non-inductive resistor, normally of 50 Ω, that turns the transmitter’s whole output into heat and radiates nothing. Cutting the power on its way to an antenna would be an attenuator’s job.

A dummy load

A dummy load takes the antenna’s place and turns the transmitter’s power into heat, so nothing is radiated. An attenuator only reduces the power and still needs a load after it.

So that it can be adjusted without radiating a signal

A dummy load lets you test and adjust a transmitter off the air: radiating needlessly wastes spectrum and is poor operating practice. The good match it gives is a side benefit, not the reason.

A non-inductive resistor

A dummy load must look like a pure resistance at RF, so a non-inductive resistor is used. Most resistors made for high power are wirewound, and their inductance makes them unsuitable.

Its coiled wire has too much inductance at RF

The coiled resistance wire gives the resistor considerable inductance, so at RF it no longer looks like a plain resistance. Such resistors handle plenty of power, which is why they tempt builders.

To carry off the heat they produce while in use

All of the transmitter’s output turns into heat in the resistor, so big loads need fins, fans, oil or a mix of these. A plain resistor needs no tuning for each band, and it stores nothing.

A short antenna, a diode to detect RF, and a sensitive meter

A simple field strength meter has only a short antenna, a detecting diode and a sensitive microammeter. With a tuned circuit added it becomes frequency selective: an absorption wavemeter.

Comparing how an antenna radiates in each direction

Kept at one spot while the antenna turns, a field strength meter shows how strongly the antenna radiates each way: its directivity and rough gain. Its readings are only relative, so it cannot give power in watts.

It responds to RF across a very wide frequency range

A simple field strength meter has no tuned circuit, so it is not frequency selective and reacts to RF across a wide range. A tuned version, the absorption wavemeter, can give a rough frequency.

They are only relative: good for comparisons, not true field strengths

A simple meter’s uncalibrated antenna and diode give readings to compare, such as one direction against another. True field strength needs a calibrated receiver and antenna, normally in a laboratory.

Such meters are insensitive and need intense fields

Simple field strength meters are not very sensitive, so they need a strong field and are used close to the antenna. A diode detector responds to an FM carrier as readily as to any other.

It adds a tuned circuit, so it responds to one band at a time

An absorption wavemeter is in effect a field strength meter with a tuned circuit added to pick the frequency. Its tuning is broad, so it gives only an approximate frequency, never an exact one.

Its tuned circuit is not very selective

The wavemeter’s tuned circuit is broad, so the meter peaks over a spread of dial settings and points only to the rough frequency. The meter’s accuracy affects the size of the reading, not the frequency.

The third harmonic of the transmitter

21,15 ÷ 7,05 = 3, so the energy is at three times the operating frequency: the third harmonic. The second harmonic would be near 2 × 7,05 = 14,1 MHz.

Checking whether a transmitter radiates harmonics

Tuned to two or three times the operating frequency, a wavemeter shows whether a transmitter radiates harmonics; energy near 7 MHz from an 80 m transmitter is the manual’s example. It gives no accurate power reading.

With a calibrated receiver and antenna, in a laboratory

Meaningful figures need a calibrated measuring receiver and a calibrated antenna, normally in a laboratory such as an anechoic chamber. A simple meter, however carefully placed, gives only relative readings.

Two tones of equal amplitude, not harmonically related

A two-tone generator gives two audio tones, equal in amplitude and not harmonically related, for the microphone input. A tone and its exact second harmonic are harmonically related.

Its peak envelope power output

With its fixed 2:1 peak-to-average ratio, a two-tone signal lets a wattmeter reading true average power, with an oscilloscope, give the PEP; the pattern also shows linearity. Deviation and modulation index belong to FM.

120 W

A two-tone signal’s peak-to-average power ratio is exactly 2:1, so PEP = 2 × 60 = 120 W. Multiplying by 1,414 confuses a voltage ratio with a power ratio.

240 W

With two tones PEP is twice the average: 500 ÷ 2 = 250 W, so 240 W is nearest and keeps the PEP under 500 W. Dividing by 1,414 gives 354 W and would overdrive it.

Its peak-to-average ratio varies with the voice and the moment

The ratio of peak to average power differs greatly between voices and from one moment to the next, so no fixed factor turns an average reading into PEP. Brief peaks still set the PEP.

The transmitter is being overdriven

Flat-topping, with the rounded tops and bottoms of the envelope chopped off, means the transmitter is overdriven; turn the drive down. Wrong bias shows instead as gaps where the loops should meet.

The amplifier is incorrectly biased

Gaps where the loops should meet mean the amplifier is incorrectly biased, so it does not conduct properly near zero. Overdriving shows instead as flattened tops.

Intermodulation distortion, which must be cured before going on air

A non-linear SSB transmitter produces intermodulation distortion, products that spread close around the signal, so the fault must be fixed before going on air. A low-pass filter cannot remove products that close.

A spectrum analyser

A spectrum analyser separates the signal’s frequencies and shows the strength of each, so distortion products show up directly. A peak-reading wattmeter gives the PEP, not the linearity.

Smooth, rounded loops that meet at zero between them

A linear transmitter turns two equal tones into an envelope of smooth, rounded loops that touch zero between them, as in the figure on p282. A steady band is what a single tone gives.