This is only a preview of the April 2026 issue of Practical Electronics. You can view 0 of the 80 pages in the full issue. Articles in this series:
Articles in this series:
Articles in this series:
Articles in this series:
Articles in this series:
|
Teach-In 2026
by Mike Tooley
World of Wireless – An
Introduction to Radio and Wireless Technology
Series 12, part 6: RF testing and measurement
I
n the last instalment, our focus
shifted to antennas, including
how several popular types are
built, how they work and how well
they perform. We investigated several basic antennas and explained
the importance of matching them
properly. Our Hands-On project
featured a 60-300MHz dipole reference antenna.
This month, we will be looking
at the principles and techniques
that underpin a range of useful RF
measurements, including voltage,
power and frequency. Our HandsOn project is a handy RF probe for
use with an ordinary DC meter for
RF measurements over a range extending from 100kHz to 100MHz.
Measuring RF voltages
If a signal is within its measurement range, the easiest method of
measuring RF signals is with the
aid of an oscilloscope and compensated probe. Reliable measurements of peak-to-peak voltage can
usually be made at frequencies up
to and beyond 30MHz. Even the
budget hand-held oscilloscope
shown in Fig.6.1 can produce accurate indications up to 10MHz.
If you are lucky enough to possess
a high-specification bench ‘scope,
you may be able to make measurements up to 100MHz or higher.
Where frequencies are beyond
the measurement range of a ‘scope,
a simple diode detector (like that
shown in Fig.6.2) can be used in
conjunction with a conventional
analog or digital meter display.
The detector responds to the peak
value of the RF input, but the meter
is often scaled in root-mean-square
(RMS) units (see later).
If you look carefully at Fig.6.2
you should be able to correlate
each of the surface-mounted components with those shown in the
circuit diagram, Fig.6.3 (there are
two diodes in one package).
The low-cost diode detector
shown in Fig.6.2 is specified for
Fig.6.1: using a
10MHz hand-held
‘scope and ×10 probe
to measure the RF
output of a low-power
7MHz transmitter.
4
operation from 100kHz to 3.2GHz
in a 50Ω system via the SMA input
connector.
To minimise any stray reactance
(that would otherwise degrade a
detector’s high-frequency performance), the RF voltmeter’s probe
tip and ground connection need
to be kept very short. Commercial
wideband RF voltmeters overcome this problem by using probe
mounts to extend the frequency
range well into the UHF/SHF range.
At signal levels below 1V, it’s
important to be aware that the DC
output voltage from a simple diode
detector responds to the peak value
of the applied RF voltage less the
diode’s forward voltage drop. This
results in nonlinearity at low input
levels, as illustrated in Fig.6.4.
Where a digital readout is required, a low-cost microcontroller
with integral analog-to-digital
converter (ADC) can be employed
along the lines shown in Fig.6.5.
Based on an Arduino Uno and a
16×2 I2C LCD screen, this delightfully simple arrangement only
needs a connection to one of the
Uno’s analog input ports from the
diode detector shown in Fig.6.6.
The code fragment shown in Listing 6.1 (later in the article) shows
how to cope with the RMS scaling
and conversion to dbV (more on
that later).
Fig.6.2: a low-cost
diode-based RF detector module.
Practical Electronics | April | 2026
C1
D2
+
RF input
D1
C2
R1
1.396 V
2.90 dBV
16×2 LCD
I2C display
Vout
Fig.6.3: the circuit of the low-cost diodebased detector module.
10.0
SDA
RF input
Vout
9.0
A4
A0
GND
A5
Diode detector
8.0
SCL
GND
Vcc
+5V
+5V
+5V
GND
0V
Arduino Uno
GND
GND
Indicated voltage (VDC)
7.0
Fig.6.5 an RMS-reading RF voltmeter based on an Arduino and 16×2 LCD screen.
6.0
5.0
4.0
3.0
2.0
Non-linear region
1.0
0
0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
Input voltage at 1MHz (VRMS)
Fig.6.4: a plot of output voltage vs RMS
input voltage for the Fig.6.3 detector.
Limitations
When interfacing with a microcontroller, it’s important to avoid
applying too much voltage to the
analog input ports. Fig.6.6 avoids
this problem with R1 and D3 protecting the input when more than
5V is applied to an analog input
port. Note also that the RF input
must be limited so that the maximum reverse voltage of the detector
diodes is not exceeded.
In Fig.6.6, each of the BAT42
diodes in the voltage doubler
arrangement has a maximum reverse voltage rating of 30V, so the
maximum input voltage that can be
safely applied to the voltage doubler
arrangement is about 20V RMS. For
greater input voltages, the BAT42
diodes should be replaced with
BAT41 diodes (100V maximum
reverse voltage).
Alternatively, a compensated attenuating probe can be used.
Voltmeter calibration & sensitivity
Simple diode detectors measure
the peak RF voltage and display
RMS values assuming the signal
is a pure sinewave, making their
readings inaccurate for other waveform shapes. True RMS voltmeters
instead calculate the effective AC
value by determining the square
root of the average squared instantaneous values over a full cycle,
regardless of waveform shape.
Devices like the popular Analog
Devices AD8318 offer accurate true
Practical Electronics | April | 2026
RMS detection through dedicated
circuitry.
The sensitivity limitations inherent in a simple diode detector can be
addressed by employing a specialised integrated circuit such as the
AD8318. It is designed for precise
relative received signal strength
indication (RSSI) and automatic
power regulation.
The AD8318 utilises multiple
amplifier stages and detectors to
compress and sum the input signal,
delivering a DC output that reliably
represents the RMS value across an
extensive frequency range, from
50MHz to over 3.5GHz. The device
features nine cascaded amplifiers,
each with an individual detector
(see Fig.6.7), with outputs that are
collectively summed to generate the
final DC output voltage.
The AD8318 has a dynamic range
of 65dB but, because of the high
voltage gain, care must
be taken to ensure that
the input power is kept
below 16mW (corresponding to an input voltage of
0.89VRMS in a 50Ω system).
An interior view of the
C1
RF input 10nF
author’s home-built AD8318 RF
detector head is shown in Fig.6.8.
Using decibels for voltage
Decibels provide us with a means
of specifying quantities using a
logarithmic rather than linear scale.
This helps keep numerical values
within a reasonable range. Expressed in decibels, the ratio of two
voltages, V1 and V2, is calculated as:
n=20 log 10
(( ) )
(( ))
V
n=20 log V dBV=20 log 10 (V )dBV
n=20 log 1010 1 V dBrel
V ref
2
(V out ( pk− pk ))
P out =
(in watts)
V
n=20
log 108 R L dBV=20 log 10 (V )dBV
1V
( )
V 2out ( pk− pk ) 2 12.65 V 2 160 V 2
=
=0.4 W
P out =(V out ( pk− pk ))=
P out =
(in watts)
8 RL
8×50
Ω 400Ω
8 RL
( )
P2
2
n=10Vlog
V2
12.65 V 2 160
out (10
pk− pk ) dB
= an =0.4 W
P out =
P 1 = Fig.6.8:
8 R L RF detector
8×50 Ω head
400Ω
based on an AD8318 module.
P
n=10 log 10 P 2 dBrel
n=10R1
log 10 P refdB
4.7kΩ P1
(( ) )
P
Vout log ( P)dBW
n=10 log P
dBW=10
10
n=10 log 1010 1 W dBrel
P ref
( ))
(
D3
RV1
10kΩ
C2
100nF
V2
dB
V1
When used to specify a relative
V
voltage (rather
than
V 2 gain
n=20 log
dBrelor loss),
log 1010 Vvalue,
refdB
we use an=20
reference
Vref, as
V1
the denominator, ie:
D2
BAT42
D1
BAT42
( )
V2
1.3 V 2
P out =5V1RMS =P
=0.0025 W or 2.5 mW
n=10 logR10L
log 10 ( P)dBW
680dBW=10
Ω
1W
( )
GND
Fig.6.6: a diode detector circuit incorporating protection for a microcontroller.
V2
1.3 V 2
P out = RMS =
=0.0025 W or 2.5 mW
RL
680 Ω
Temperature
TEMP
Gain bias
Slope
VPSI
sensor
I
Ʃ
DM1
INHI
INLO
G1
DM2
G2
DM3
G3
DM4
DM9
I
V
V
VSET
VOUT
CLPF
C
G9
Nine stage demodulating logarithmic amplifier
Fig.6.7: the simplified internal arrangement of an AD8318 IC.
5
+40
+30
+20
dBV
+10
0
-10
-20
-30
-40
0.01
0.1
( )
1.0
Measured voltage (V)
10
100
V
n=20 log 10 2 dB
Fig.6.9:
V 1 a plot of dBV against RMS voltage.
1V is oftenV used as the reference
n=20 log
dBcase
value,
in10which
the relationrel
V ref
ship becomes:
load, producing a DC output voltage
proportional
V
to signal amn=20 log 10
dBV=20 log 10 (V )dBV
1V
plitude
To help you understand
this
rela•
applying
the
2
(V outFig.6.9
( pk− pk ) ) shows dBV plotted
tionship,
attenuated
RF Fig.6.10: a wideband AC voltmeter calibrated in RMS volts & dBV.
P out =
(in watts)
8 R Lvoltage. Note that 1V =
against RMS
signal from a
0dBV, 10V = 20dBV, 100V = 40dBV
matched attenuator to an inte- auto-balancing bridge. The output of
2
2
and soVon.
Fig.6.10
a Vtypical
grated circuit that senses the true the bridge is a DC voltage that can be
12.65shows
V 2 160
out ( pk−
pk )
=
=0.4 W
P out =
=
meter
display
separate
scales
for
RMS value of input voltage from displayed using an analog meter or
8 R L with
8×50
Ω 400Ω
RMS voltage and dBV.
the attenuator
converted using an ADC (eg, within
The reason
• sampling the RF signal as a base- a microcontroller) for digital display.
P 2 that the formulae
n=10 log
dB
10 a scaling
above
use
factor of 20,
band AC signal, which can then
Due to the intermediate converP1
rather than 10 (as you might exbe directly digitised, processed sion of RF energy to heat, thermal
pect from the unit name decibel),
digitally and displayed
sensors provide a true RMS response
P
dB
isn=10
that log
decibels
are
• calculations of power based on over an exceptional range of frequen10
rel fundamentally
P
a measure of refpower ratio, not voltvoltage measurements made cies. Unfortunately, while they are
age ratio. When comparing voltages
with a wideband oscilloscope, excellent for measuring average
P impedance, power
across
the10 same
compensated probe and matched power, they can be too slow for some
n=10 log
dBW=10 log 10 ( P)dBW
1 W to the square of voltis proportional
resistive load
applications (eg, they will barely
age: P =2V 2 ÷ R.
Both thermal and detector-type respond to brief high-level bursts of
V RMSthe1.3
V2
ofWa squared
RF energy due to thermal inertia).
PTaking
= logarithm
=0.0025
or 2.5 mW measurements are often described
out =
680
Ω 10(V), and when
voltage Rgives
2log
as ‘direct sensing’ because the
Simple diode detectors provide
L
this is multiplied by 10 (the ‘deci’ conversion of an RF signal to a DC a fast response, but they can be inpart of decibel), the result is the output is performed by means of a sensitive without additional amplifamiliar 20log10(V) scaling.
probe or head mounted close to the fication. Due to the non-linearity of
signal source. More complex ‘indi- their characteristic and maximum
Measuring RF power
rect’ methods involve a multi-stage reverse voltage rating, diode detecSeveral different technologies conversion performed remotely, tors have a limited dynamic range.
are used for measuring RF power. where amplitude data is scaled to
The modern solution to the
They include:
provide a power reading.
measurement of RF power is with
• thermal methods, in which the
Thermal power sensors usually the aid of integrated circuits deheating effect of RF power in a employ thermistors attached to a signed specifically for true RMS
matched load is sensed using a matched load. The resistance of the voltage and power measurement.
thermistor (or ‘bolometer’)
thermistor sensor changes with load Supported by a matched attenu• using diode detectors to rectify temperature. This change is detected ating load and software running
the RF voltage across a matched using a temperature-compensated on a microcontroller or PC, these
devices provide a sensitive and
D1 BAT41
highly cost-effective solution to RF
power measurement at frequencies
SK1
up to several gigahertz.
( )
( )
( )
( )
( )
RF input
6
R1
50Ω
C1
10nF
+
V
-
Fig.6.11: the circuit
of a basic RF
power meter.
Using an oscilloscope
At lower RF frequencies, it is possible to use an oscilloscope to measure RF voltage and thus, provided
Practical Electronics | April | 2026
Power versus RMS voltage (50Ω load)
2.0
+20
+15
+10
1.5
dBW
Power (W)
+5
1.0
0
-5
n=20 log 10
-10
0.5
-20
0
2
4
6
8
10
RF voltage (VRMS)
V2
n=20
dBbetween power and voltage for Fig.6.11.
10
Fig.6.12:
the log
relationship
V1
( )
0.01
(( ) )
( )
( )
( ) ( ( ))
V
V 2 log
n=20
10 V10 dBV=20 log
dB
0.1 n=20 log 101.0
10010 (V )dBV
n=20
log
1
V dBrel
10
V1
V
Measured power (W) ref
2
Fig.6.14: dBW plotted against(VRMS
power.
pk ) )
V = out ( pk−
P
(in watts)
V
out
n=20
log
dB
10n=20 log
rel R
good-quality 50Ω load. Such instru- with voltage,
keeplog
nudBV=20
10 8helps
10 (V )dBV
V ref this
1 VL
that the load resistance is accurately
known, to determine
power from ments may have digital or analog
V
n=20 log 10
dBrel of a wavethe peak-to-peak
displays and are usually calibrated in
V refvalue
form observed on an oscilloscope watts (W) or milliwatts (mW).
(many oscilloscopes
can also perThe circuit of a basic RF power
V
n=20 log 10RMS measurements).
dBV=20 log 10 (V )dBVmeter is shown in Fig.6.11. This
form accurate
1V
V
The following
simple arrangement incorporates
n=20
log 10 2 dBrelationship is used:
V1
a suitably rated 50Ω resistive load
2
(V out ( pk− pk ))
across which the applied RF voltP out =
(in watts)
V 8 RL
age is measured. Fig.6.12 shows the
n=20 log 10
dB
V ref2 intorelcontext, the ‘scope
square law relationship between
Putting this
V out ( pk− pkearlier
160 V 2
12.65in
V 2 Fig.6.1
)
and applied voltage in
display
shown
=
=0.4power
W
P out =
=
8 RL
8×50 voltage
Ω 400Ωof
Fig.6.11. The resulting nonlinear
indicates aV peak-to-peak
n=20 log 10
dBV=20 log 10 (V )dBV
12.65V. This
devel- calibration of a typical RF power
1 V voltage is being
P 2 load consisting of
meter scale is shown in Fig.6.12.
oped across a 50Ω
n=10 log 102
dB
Fig.6.13 shows a typical example
two 100Ω
1W
resistors
connected
in
P
(V out ( pk− pk ))
1
P out =
(in watts)
of
power measurement at RF using
parallel.
The
output
power
can
thus
8 RL
a combined power and SWR meter
be calculated
as follows:
P
n=10 log 10
dB
and an external load. The meter’s
P ref V 2 rel160 V 2
V 2out ( pk− pk ) 12.65
digital display indicates an output
=
=0.4 W
P out =
=
8 RL
8×50 Ω 400Ω
power of exactly 2W.
P
n=10
log 10
log 10 ( P)dBW
If the
waveform
is dBW=10
not a sinewave
1W
P 2 of distortion)
(or has
a lot
and your Using decibels for power
n=10
log 10
dB
P 1 2has a built in RMS voltoscilloscope
Decibels are frequently used to
V
1.3 V 2 you can use
age measurement
specify power in RF systems. As
P out = RMS = feature,
=0.0025 W or 2.5 mW
680 ΩRMS measureL
that to get an
PRaccurate
n=10 log 10
dBrel
ment and then
convert
the voltage to a
P ref
power figure using the load resistance.
Some high-end
oscilloscopes
P
n=10
log 10 a mathematical
dBW=10 log 10 ( P)dBW
may have
1W
function to perform that
conversion
forVyou.
2
V 2RMS 1.3
P outIn
= that
=case, ensure
=0.0025 W or 2.5 mW
RL
680 Ω
you have
a sufficient
number of signal cycles
visible on the screen for
an accurate RMS measurement. If you can see, say, oneand-a-half cycles of the signal, the
calculation may be inaccurate due to
the imbalance that comes from including only half of the second cycle.
( )
V2
dB
V1
V
n=20 log V
dB
n=20 log 10 10 V2 refdB rel
V1
-15
0.0
( )
( ) ( )
merical values within a reasonable
range. Expressed
decibels,2 the
V 2out (in
160 V 2
pk− pk ) 2 12.65 V
)= and
(V
=
=0.4 W
PVoutpowers,
=
out ( pk− pk P
)log
n=20 of
log 10
dBV=20
(V
)dBV
P
,
is
ratio
two
10
1 (in
P out
=
watts)
8R
8×50
Ω2 400Ω
1V
calculated from: 8 R LL
( )
P out =
( )
2
P2
(V out ( pk− pk ))V 2
n=10 log
dB
12.65 V 2 160 V 2
out(in
( pk−
pk )
10 watts)
8PRoutL = 8 R P 1= 8×50 Ω = 400Ω =0.4 W
L
( )
When
used to specify a relative
2
2
V (rather
160
V2
12.65
power
we
out ( pk−
pk ) than
PV2 P= or
n=10
log 10gain
dBloss),
rel =0.4 W
P out =
=
n=10 log
P refdB400Ω
10
8R
8×50
use a reference
value,
Pref, as the
L
PΩ
1
denominator, as follows:
( )
P
P 2 log
n=10
dBW=10 log 10 ( P)dBW
n=10 log 10n=10
dB 10 P
P 1 log 10 P1 WdBrel
( )
( )
( )
( )
(( ) )
( ) (( ) )
ref
1W is often
the
V 2RMS as1.3
V 2reference
P =used
P
=
=0.0025 W or 2.5 mW
P
out
n=10 log
dB case680
value,
in10n=10
which
the
dBW=10
log 10 ( P)dBW
Ω relationP ref logR10relL 1 W
ship becomes:
( ) ( )
( )
P 2RMS 1.3 V 2
n=10 log 10P = VdBW=10
log 10 ( P)dBW
W or 2.5 mW
out
1 W R = 680 Ω =0.0025
L
This 2 relationship
is illustrated
V
1.3 V 2
graphically
P out = RMS = in Fig.6.14,
=0.0025 Wwhere
or 2.51W
mW
R L 10W
680 Ω= 10dBW, 100W =
= 0dBW,
20dBW and so on. Note that the
( )
( )
Standalone power meters
Fig.6.13: measuring the output power
from a hand-held digital transceiver.
The power meter indicates an output of
exactly 2W delivered to the attached 50Ω
load.
For regular measurements, a
dedicated power meter is a better
alternative to an oscilloscope, with
the added advantage that many standalone RF power meters incorporate a
Practical Electronics | April | 2026
7
Fig.6.16: a low-cost AD8362 true RMS power-sensing module.
True RF power
meters
A similar arrangement to
that shown earlier in Fig.6.11
can be used
to indicate RF
power, but in
this case, the
true RMS diode detector is
in conjunction
Fig.6.15: an RF power meter calibrated in RMS power and dBm. with a matched
resistive load
same relationship would apply if and the display scaled in milliwe had plotted dBm (dB relative watts or watts. The load needs to
to 1mW) against milliwatts (mW). be suitably rated in terms of power
Hence, 1mW = 0dBm, 10mW = dissipation and must be purely
10dBm, 100mW = 20dBW and so resistive.
True power meters must also
on.
Fig.6.15 shows a typical analog cope with the waveforms and crest
power meter display with separate factors (peak to average ratios)
scales for RMS power and dBm. With found in modern modulation sysits companion probe mount, this tems such as quadrature amplitude
modulation (QAM) and orthogonal
instrument is useful up to 40GHz.
Fig.6.17: using the AD8362
with an Arduino Uno and
digital display to measure the
power output from a hand-held
transceiver.
8
frequency division multiplexing
(OFDM). This places extra demands on the signal processing,
requiring a fast response coupled
with a wide measurement range,
typically 60dB or more.
Typical of the devices currently
available is the Analog Devices
AD8362. This chip is a true-RMSresponding power detector offering a 65dB measurement range,
designed for systems that demand
an accurate response to true signal
power. The chip operates from
50Hz to over 3.8GHz and accepts
inputs from -52dBm to +8dBm.
The AD8362 requires a single
+5V supply and consumes a mere
1.3mW when in the powered-down
standby state.
A typical low-cost power sensing module based on an AD8362
is shown in Fig.6.16. The analog
output of the module (0-5V) can
be applied to a DC meter or to the
analog port of a microcontroller
like that shown in Fig.6.17, using
code to display power in dBW or
dBm if required, as shown on the
LCD screen in that figure.
Note the use of a power attenuator in Fig.6.17 to reduce the signal
level into the AD8362. The first line
of the display in Fig.6.18 indicates
the power supplied to the sensing
module (which must be kept below
Fig.6.19: checking a dual-band FM
transceiver. The DFM on the right
indicates a frequency difference of 1kHz.
Practical Electronics | April | 2026
Fig.6.20: checking the calibration of an RF signal generator. The 8-digit Fig.6.21: a home-built low-cost DFM being used to check
DFM display shows 49467.584kHz with a 1Hz resolution.
the output frequency of the VFO described in Teach-In 12.3.
12dBm), while the second line
indicates the power dissipated in
the attached 50Ω load.
When using sensitive RF power
measuring devices for indicating high levels of RF power, it’s
essential to use a suitably rated
combined load and attenuator.
Measuring frequencies
Frequency is difficult to measure
accurately using a conventional oscilloscope. Modern digital storage
oscilloscopes (DSO) fare somewhat
better in this respect, as they can
often display frequency values
digitally. However, a dedicated
digital frequency meter (DFM) is
often a better solution.
The desirable characteristics of
such an instrument are high sensitivity and high upper frequency
limit as well as adequate accuracy
and resolution.
A sensitivity of 100mVRMS or
better is suitable for most applications, and an upper frequency
limit of 500MHz will be adequate
for most measurements at HF and
VHF. Most DFMs offer an accuracy
of ten parts per million (0.001%).
This is more than adequate for all
but the most critical applications.
Input
Amplifier and pulse
shaper
Reference TCXO
The resolution of an instrument
depends on the number of digits
in the display. For example, if a
resolution of 100Hz is required
when measuring a signal at 50MHz,
a minimum of six display digits
will be required.
The block diagram of a basic digital frequency meter is illustrated in
Fig.6.18. The signal to be measured
is initially sent through an input
amplifier, followed by a Schmitt
trigger buffer, which converts and
shapes nearly all input signals
into a logic-compatible pulse
train. While the signal’s waveform
changes, the pulses still match the
original input frequency.
Having a logic-compatible signal
is important for connecting with
the signal gate and the circuits
that handle gating, counting, and
latching.
The reference oscillator provides
a very stable and accurate source
for the time base circuit, typically
using a temperature-controlled
crystal oscillator (TCXO). TCXOs
offer stability better than 1ppm,
compared with standard quartz
crystal oscillators, which usually
have an error of about 30ppm, taking into account both initial accu-
Gate
Pulse counter
racy, ageing, temperature drift etc.
The control pulses from the timebase have an accurate time duration
and are applied to the signal gate
which, in effect, opens and closes
to allow a train of signal frequency
pulses to pass through into the
counter over a very precisely controlled interval.
These pulses are then electronically counted, and the result is
passed to the display latches,
which retain the count during the
subsequent counting process.
To clarify this concept, consider
a scenario in which a 25MHz signal
is applied to the input of a five-digit
DFM with its timebase set to 1ms.
During this 1ms interval—when
the gate remains open—25,000
pulses are counted, and the latch
subsequently updates the display
to indicate “25000kHz”.
Conversely, if the time base is
set to 1µs rather than 1ms, only 25
pulses are registered, resulting in
a display of “00025MHz”.
Figs.6.19 through 6.22 show
examples of DFMs being used. In
Fig.6.21, the five-digit display indicates a frequency of 744.64kHz
with a resolution of 10Hz. Note
how in Fig.6.22, an inline 30dB
Display
Display latch and drivers
Time base dividers
Range
Fig.6.18: the simplified block schematic
of a digital frequency meter (DFM).
Practical Electronics | April | 2026
9
power attenuator is connected between the transmitter output and
DFM. This acts both as a matched
load for the transmitter and as an
attenuator to prevent overdriving
the input of the DFM.
Hands-On: Handy RF voltmeter probe
This month’s Hands-On project
is a voltmeter probe that’s ideal for
signal tracing in RF circuits. The
probe will allow you to quickly
and easily make in-circuit measurements of RF voltages using only a
low-cost digital meter for display.
The probe can be calibrated for
RMS (sinewave) voltages up to 20V
over a frequency range from 25kHz
to 95MHz.
The circuit of the RF voltmeter
probe is shown in Fig.6.23. The
probe uses two BAT42 Schottky
diodes, D1 and D2, connected in a
voltage-doubler arrangement like
that shown previously in Fig.6.6.
The DC output voltage (approximately twice the peak value of RF
voltage applied to the probe tip) is
developed across C2.
To overcome the non-linear
diode characteristics, a small
amount of bias voltage is applied
to the detectors. This voltage can
be adjusted by RV1 so that the
two diodes are just on the point of
conduction.
The bias supply is derived from
a single 3V coin cell, which is
mounted in a compact printed
circuit board holder.
Lithium coin cells are a severe
hazard to young children if they
swallow them, so this board with
the onboard cell holder must be
secured in a plastic case held together with screws. Luckily, there
is a neat case available that’s the
perfect size for the job (see the parts
list for details).
As with our previous Hands-On
projects, this one is constructed on
a small piece of perforated copper
stripboard measuring 25 × 64mm,
arranged as nine strips each of
24 rows. The component layout
(viewed from the top) is shown in
Fig.6.24, while the corresponding
track layout (viewed from below)
is given in Fig.6.25.
The required 16 track breaks can
be made using a spot face cutter
or small drill bit, and the six links
on the upper side of the board
can be made using short lengths
of tinned copper wire. Take care
to observe the correct polarity of
the two BAT42 diodes (the stripe
on the package is the cathode connection).
10
The completed stripboard is
mounted, together with S1, inside
an ABS probe or small project case.
Suitable enclosures are available
from several online suppliers at
a reasonable cost. However,
when purchasing, it’s important to check that
the dimensions of
the case are sufficient
to accommodate the
board (25 × 64mm). The
case used for the prototype
measured 120 × 40 × 25mm.
As explained above, the
board must be mounted in a case
to make the coin cell inaccessible.
Don’t skip that step! Make sure to
use a case that doesn’t come apart
without tools.
Fig.6.26 shows the internal wiring of the RF voltmeter probe. The
probe tip is connected to P1-2 using
a short length of insulated hookup
wire, and a 100mm length of insulated stranded wire (terminated
with a crocodile clip) is connected
to P1-1. This wire should be kept
reasonably short to ensure good
high-frequency performance.
The DPDT slide switch, S1, is
connected to P2-2 and P2-3 using
Fig.6.23: the circuit of the RF voltmeter
probe.
Probe tip
Ground
clip
C1
10nF
D2
BAT42
D1
BAT42
C2
10nF
P1-2
P1-1
C3
10nF
Fig.6.22: using a low-cost DFM to
measure the output frequency of a 1W
7MHz transmitter.
short lengths of insulated hook-up
wire. 500mm lengths of black and
red insulated multi-core flexible
wire are used to link the DC output
from the probe to the voltmeter.
These leads should be fitted with
4mm banana plugs and the positive
(red) and negative (black) connections terminated at P2-1 and P1-1,
respectively.
P2-2
P2-3
+ B1
3V
S1
On/off
R1
1kΩ
RV2
10kΩ
P2-1
RV1
500Ω
C4
100nF
+ (red)
Output to DC
voltmeter
- (black)
Figs.6.24 & 25: the component & track layouts for the RF voltmeter probe circuit board.
Practical Electronics | April | 2026
2.0
Indication (V)
1.5
Probe ON
1.8
1.0
Indicated voltage (VDC)
1.6
1.4
0.5
Probe OFF
1.2
0.0
10k
1.0
100k
1M
10M
100M
Frequency (Hz)
Fig.6.28: the frequency response of the RF voltmeter probe.
0.8
0.6
0.4
0.2
0
0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
Applied voltage at 1MHz (VRMS)
Fig.6.27: indicated voltage (VDC) plotted against input voltage
(VRMS) with the probe switched on and off.
Adjustment and calibration
The RF voltmeter probe requires
initial adjustment and calibration
using an RF signal source. If you
don’t have a suitable RF signal
generator, you can use the VFO
(‘variable frequency oscillator’) or
reference source module that we
described in Part 3 of Teach-In 12
(in the January 2026 issue). The
following steps are required:
1. Switch the probe off and set both
RV1 and RV2 on the probe to
their fully anticlockwise positions. Insert the CR2032 3V coin
cell into the holder with the positive (+) side up.
2. Leaving the probe tip and ground
lead shorted together, connect
the red and black output leads
from the RF voltmeter probe to a
digital voltmeter. Select the 20V
DC range on the meter and keep
it on this range for all subsequent
steps.
3. Ensure that the probe tip remains unconnected and switch
the RF voltmeter probe on. Ignore any reading on the meter
at this stage.
4. Advance the setting on RV1 on
the probe (turning it slowly clockwise) and adjust it for a reading
Practical Electronics | April | 2026
of between
0.02V and
0.03V.
5. Connect the
probe tip to Fig.6.29: using the RF voltmeter probe to investigate voltages in
SK1 on the an RF power amplifier.
VFO (or an
alternative signal source of
against a known instrument and
about 2V peak-to-peak at around
RV1 re-adjusted accordingly.
1MHz). Clip the ground connecFig.6.27 shows the response of
tion to the 0V common rail.
the RF voltmeter probe to voltages
6. Adjust the VFO to produce an below 2V. The result of applying
output at about 800kHz and set a bias is evident, as is the error at
RV1 on the VFO to mid-position. applied voltages of less than 0.8V.
At this point, the meter should For most purposes (and when carbe indicating a reading of around rying out relative adjustments), this
2V.
small error can be ignored. Note the
7. Note the reading on the meter difference with the probe switched
and then adjust RV2 on the probe on and off.
so that the meter reading falls to
The frequency response of the
exactly 0.4 times (40% of) the RF voltmeter probe is shown in
initial reading indicated. For ex- Fig.6.28. The response is substanample, if the initial reading was tially flat from 100kHz to 10MHz,
2.1V, RV2 should be adjusted so with a small resonant peak due to
that the meter reads (2.1V × 0.4) stray reactance at about 50MHz.
= 0.84V (the RMS equivalent).
Thereafter, the probe is useful to
8. The RF probe should now indi- about 80MHz, above which the
cate the approximate RMS value output falls rapidly.
of the sinewave. If possible, the
calibration should be checked Using the RF voltmeter probe
Fig.6.29 shows a typical application for the RF voltmeter probe
when fault-finding on an HF power
amplifier.
The probe tip is taken along the
signal path to the points under
investigation, and the ground clip
left attached to a suitable ground
point. In this case, the grounded
metal enclosure has been used, but
a nearby ground point on the PCB
Fig.6.26: the internal wiring of the RF
would also be suitable.
voltmeter probe.
11
n=20 log 10
( VV )dB
rel
ref
Fig.6.30 Vshows how the RF probe
n=20be
logused
dBV=20
logoperation
10
10 (V )dBV of
can
to check
the
1V
the amplitude modulator described
2
in Teach-In
4 (February 2026
(V out ( pk− pkPart
))
issue).
the ground clip
P out = Here again,
(in watts)
8 RL
is attached
to a convenient point on
the common
0V rail 2and the2 probe
V 2out ( pk− pkto
160 V
12.65 V
) different
tip
moved
= test points.
=0.4 W
P out =
=
8 RL
8×50
Ω 400Ω
For comparison,
we’ve
shown the
DC and RF test voltages at various
P circuit.
points in the
n=10 log 10 2 dB
The signal
P 1 path can be clearly
seen. It’s also possible to determine
the output Ppower from the modulan=10appearing
log 10
dBrel
tor,
the 680Ω load
P ref across
resistor, R4. The 1.3VRMS measurement at this
point (dropped across
P
n=10can
log 10 be used
dBW=10
log 10 ( P)dBW
R4)
to determine
the
1W
power dissipated in R4, as follows:
( )
( )
( )
Coming up
In next month’s instalment, we
will be investigating low-cost spectrum and virtual network analysis.
We will also have two Hands-On
projects. One takes the form of a
combined load and power meter
that makes adjusting low-power
HF transmitters a breeze; the other
is a separate 25W load suitable for
PE
operation up to 450MHz.
9.5V
RF
0V
R5 560Ω
C5
47µF
L1
100µH
DC
4.4V
RF
0.7V
R3
100kΩ
TR2
2N2222
C1 10nF
SK1
RF input
1.3V
RF
0V
1.3V
e
DC
3.3V
c
RF
0.7V
TR1
2N2222
e
R1
100kΩ
P1-2
Fig.6.30: DC
and RF voltage
measurements made
on the amplitude
modulator circuit
from Teach-In Part 4.
DC
0V
RF
1.3V
Mod. RF output
C4
1nF
b
Com.
Listing 6.1:
sample code
for RMS scaling
and dBV
conversion.
RF
RV1
500Ω
P2-3
C6
100nF
+
SK2
+
DC
9.5V
+12V
c
R2
100kΩ
AF input
P1-1
DC
b
C2 10µF
( )
V2
1.3 V 2
P out = RMS =
=0.0025 W or 2.5 mW
RL
680 Ω
DC
+
C3
10µF
R4
680Ω
DC
0.7V
RF
0V
P2-1
Com.
P2-2
void loop() {
int rawValue = analogRead(inputPin);
float voltage = (rawValue / ADC_STEPS) * V_REF;
float voltRMS = voltage / 1.414;
lcd.setCursor(0, 0);
lcd.print(voltage/1.414, 3);
lcd.print(" V");
double dBV = 20.0 * log10(voltRMS);
lcd.setCursor(0, 1);
lcd.print(dBV, 2);
lcd.print(" dBV");
delay(1000);
lcd.clear();
}
Parts List – RF voltmeter probe
1 25 × 64mm piece of stripboard (9 × 24 holes)
1 ABS probe case (129 × 40 × 25mm) [eBay 291019910762]
1 panel-mount probe (eg, machine screw sharpened to a
point and matching nut)
1 2-pin male 0.1in/2.54mm header (P1)
1 3-pin male 0.1in/2.54mm header (P2)
1 DPDT miniature slide switch (S1)
1 PCB-mounting CR2032 coin cell holder (B1)
1 CR2032 3V lithium coin cell (B1)
4 brass or nylon hexagonal M3 × 10mm stand-off/spacer
4 M3 panhead machine screws
4 M3 countersunk head machined screws
1 red 4mm banana plug
1 black 4mm banana plug
1 insulated crocodile/alligator clip
1 50cm length of red PVC flexible test lead
1 50cm length of black PVC flexible test lead
1 10cm length of black PVC-insulated flexible wire
Semiconductors
2 BAT42 Schottky diodes (D1, D2)
Capacitors
1 100nF 50V ceramic (C4)
3 10nF 50V ceramic (C1-C3)
Resistors
1 1kΩ ¼W axial, 5% or better (R1)
1 500Ω top-adjust miniature preset variable/trimpot (RV1)
1 10kΩ top-adjust miniature preset variable/trimpot (RV2)
12
Practical Electronics | April | 2026
|