This is only a preview of the July 2025 issue of Silicon Chip. You can view 37 of the 104 pages in the full issue, including the advertisments. For full access, purchase the issue for $10.00 or subscribe for access to the latest issues. Articles in this series:
Articles in this series:
Items relevant to "The SmartProbe":
Articles in this series:
Items relevant to "Hot Water System Solar Diverter, part two":
Items relevant to "Low-cost electronic modules: 8×16 LED Matrix module":
Items relevant to "SSB Shortwave Receiver, part 2":
Purchase a printed copy of this issue for $13.00. |
SSB Shortwave
Receiver
Part 2 by Charles
Kosina, VK3BAR
Introduced last
month, this new Shortwave
Receiver covers the entire shortwave
band from 3MHz to 30MHz. It is digitally tuned and
has a host of useful features like squelch, USB/LSB support, good
sensitivity (a -107dBm signal gives 13dB SNR), fast or slow AGC, an RSSI display and
it runs from 12V DC. This month, we describe how to build, test and align it.
T
his is not an overly difficult
device to build, as it uses no
tiny components or fine-pitch
ICs. However, it has two boards that
are fairly packed with SMDs plus quite
a few though-hole components, so you
should ideally have decent soldering
skills if you’re going to attempt it.
You also need some test equipment
to calibrate the Radio. That includes
an accurate frequency counter up to
100MHz and a signal generator that
will work up to at least 30MHz that
can produce a signal down to 10µV
or less (or an attenuator to allow that).
An oscilloscope with 100MHz or more
bandwidth is also nice to have, but not
absolutely necessary.
Some sheet metalwork is needed.
74
Silicon Chip
I recommend having a stepped drill
bit or two (eg, 3-12mm & 12-24mm)
on hand. A drill press would be ideal,
but you can do it with a hand drill if
necessary.
There are many components overall, but the values are marked on the
circuit boards to ease construction. It
pays to be careful as you go through
the assembly process and make sure
each part goes where it’s supposed
to. Mixing up two visually identical
capacitors could be enough to prevent
the radio from working.
Construction
Virtually all the components mount
on two PCBs, the Control Board
(Fig.14, 152.5 × 81.5mm) and the RF
Australia's electronics magazine
board (Fig.15, 152.5 × 51mm). There
are some through-hole components
used, but the vast majority are SMDs,
mostly passives (resistors and capacitors) in M2012/0805 packages, which
measure 2.0 × 1.2mm.
These passives are on the small side
if you are used to through-hole components, but we still consider them to be
in the ‘easier to handle’ category compared to really small parts. So as long
as you have the right tools, a decent
amount of light and reasonable vision
(or magnification), you should not find
the assembly too difficult.
Similarly, the ICs are not in really
tiny packages; they are mostly SOIC
types with 1.27mm lead pitch, ie, half
that of a through-hole chip. Again,
siliconchip.com.au
Fig.14: the
Control Board
has parts on both
sides; fit all the
SMDs first, then
the throughhole parts on
the underside.
The ICs, diodes,
electrolytic
capacitors and
the Arduino
Nano module
must all be
installed with the
polarities shown
here for the
Radio to work.
these are not what we would consider
difficult-to-solder parts.
Control board
I recommend building the Control
Board first and testing it before you
move on to the RF Board. There are
components on both sides of the board,
but most of the parts, including all the
SMDs, are on the front. Start by soldering the two ICs first, making sure
their orientations are correct.
In each case, find the pin 1 marker (a
dot or divot on the top, or a chamfered
edge on the side) and make sure it’s
aligned as shown in Fig.14 and on the
PCB silkscreen. It’s possible to solder
the pins of these SOIC package devices
individually with a fine-tip soldering
siliconchip.com.au
iron. Add plenty of flux paste to make
soldering easier and reduce the possibility of bridging pins with solder.
If that happens, use copper braid
with a bit of extra flux paste to remove
the excess solder. In fact, we usually
don’t bother trying to avoid bridges
as it’s so easy to clean them up later;
we’re more focused on making sure the
solder flows onto each pin and pad, to
avoid high-resistance connections that
can be difficult to find later.
Follow with the passives using a
similar technique. The resistors will
be marked with codes indicating their
values (eg, 10kW = 103 or 1002) while
the capacitors will not have any markings. In both cases, it’s best to unpack a
single value, then fit them all as shown
Australia's electronics magazine
on the overlay diagram so you can’t get
them mixed up. None of the passives
are polarised so they can be soldered
either way around.
Note that a few of these parts, like
the 68W resistor and 100μF capacitor,
are slightly larger than the others and
so have larger pads to suit. Also, two
of the 8.2kW resistors at centre left
are not fitted (marked R10 & R20 on
the PCB) as these are the I2C pull-ups
and the Si5351 module has onboard
pull-up resistors.
Follow by soldering the four identical Mosfets, which are all in three-pin
SOT-23 packages. The pins are small
but widely spaced, so this should not
be too difficult.
Don’t fit any of the through-hole
July 2025 75
components on the front side of the
PCB (where the SMDs have been soldered) yet. The Arduino Nano and
the Si5351 modules are on the back
of the board and can be plugged into
socket strips. This is important as if
either failed, replacing them would
be difficult.
15-pin socket strips are used for
the Nano and one seven-pin strip for
the Si5351. The only other parts on
the back of the board are the headers,
two electrolytic capacitors (which are
polarised, so make sure they’re fitted
the right way around) and the trimpot
for LCD contrast. All of those can be
mounted now.
As well as the speaker connector (CON4), there is another two-pin
header, CON3. This is connected in
parallel with the headphone jack
socket and may be wired to an RCA
socket on the back panel for an external powered loudspeaker.
Now go back to the front of the PCB
and fit the remaining through-hole
components. It’s important that the
switches, potentiometers and encoder
are square on to the board before soldering.
The way to ensure this is to attach
the black front panel with 16mm
tapped spacers to position these correctly; make sure that all controls turn
easily before soldering. For a better
appearance, rather than zinc-plated
screws, I used black 6mm machine
screws (which you can buy at Bunnings) to attach the front panel.
The jack socket is a unique part;
ensure it is pressed firmly on to the
board. Next, solder the 16-pin header
to the LCD module.
Don’t attach the LCD yet; clean the
board with de-fluxing solvent and
inspect all connections with a magnifier. Pay special attention to the solder joints on the socket strips for the
Nano, as they are not accessible once
the LCD is mounted. Before any modules are plugged in, use an ohmmeter
to ensure that there are no shorts from
the 12V or 5V supply lines to ground.
Finally, attach the LCD module using 5mm spacers and 12mm
machine screws and nuts. The Si5351a
module is held in place by 6mm M2
or M2.5 screws with 11mm threaded
spacers.
RF board assembly
If you want to take a break from
assembly now, you could skip down
to the “Programming the Nano”
sub-heading, complete initial testing
and calibration, then come back here
when you get to the part where you
need the RF board to be assembled.
The RF board overlay is shown in
Fig.15. Parts are only fitted to one side
of this board. As before, start with the
ICs (two NE612s, one LMC6482 and
the PCF8754) and ensure they are
all orientated correctly as you solder
them. All are in SOIC packages.
Then move on to transistors Q1-Q7;
Q7 is a Mosfet, while the others are
NPN RF transistors, but they all come
in the same packages, so don’t get them
mixed up.
Follow with Q8-Q10, which have
four pins since they are dual-gate
Mosfets. In all three cases, the wider
source lead goes towards lower right
with the PCB orientated as shown in
Fig.15. Fit diode D4 with its cathode
stripe as shown, then REG2 after first
applying a thin layer of flux paste to
its pads. That will assist in soldering
its tab properly. After that, solder the
SMD resistors & capacitors, noting that
the capacitors are again unpolarised,
and all the SMDs are on the board.
Mount the axial inductors next; they
have three different values, so make
sure the right ones go in each location.
They are not polarised, so you can fit
them in either orientation. Fit diodes
D2 & D3 next; they are polarised, so
ensure the cathode stripes both face
to the right. After that, solder RLY1 in
place with its pin 1 marking towards
the top.
Next, fit VC1-VC3, which are polarised in a sense, because we want the
adjustment screws to be connected to
the ground pins in each case. So orientate them as we have shown.
For the varicap diode, VD1, you
may get it in a two-lead TO-92 package like we did, or in an axial package, similar to a regular diode, which
can be mounted vertically. Regardless,
ensure its cathode lead goes to the pad
marked K; with the axial package, this
will be the end with the stripe.
Bend REG1’s leads down and attach
it to the board using an M3 machine
screw and nut, ensuring its three leads
go into their pads. Solder and then trim
the leads. Don’t do this before tightening the screw or you could fracture
the leads. I used a 16mm-long screw to
attach the tab as it makes a convenient
ground point for testing later.
Next, install CON1 and CON2. That
just leaves the crystal filter module,
XF1.
The crystal filter comes with SMA
sockets attached, and at least the input
one has to be removed. As it’s supplied, only the top connections are
soldered; I used a hot air wand to carefully heat them and slide them off, but
a soldering iron with a large tip could
also be used instead. Take great care
that other nearby components don’t
get removed as well.
Attach the filter to the PCB using
four 10mm-long M2 or M2.5 screws,
nuts and 5mm spacers. Solder wires
to connect the input and output of
the filter to the circuit board, one for
signal and another for ground at each
end. In theory, XF1 is not polarised,
but it’s a good idea to mount it like
we did, with the angled capacitor on
Fig.15: all the parts
are on the top side
of this RF Board.
Polarised components
to pay attention to
include the ICs, dualgate Mosfets, diodes,
relay and variable
capacitors. It’s best
to remove the SMA
connectors from the
crystal filter module
(XF1) before mounting
it on this board.
76
Silicon Chip
Australia's electronics magazine
siliconchip.com.au
the right-hand side, since that’s how
we tested it.
Toroid winding
The 3-10MHz toroid (T1) needs 42
turns of 0.35mm diameter enamelled
copper wire (ECW) for its secondary
and four turns of the same wire for
the primary. The secondary will take
a little while to wind; do it first and
neatly, with the turns almost touching
each other. There is just enough room
for that number of turns with a small
gap in between the ends.
Attaching the transformers to the
PCB is one of the most fiddly parts of
the assembly process. T1’s primary is
soldered between the pads labelled A
& B on the PCBs, and is wound near
the ‘cold’ end of the secondary, while
the secondary is soldered between
pads C & D.
I found it easier to attach the transformers to the PCB first, solder the
secondary windings, then add the primary windings. It is a bit tricky but I
think it is the best approach!
Make sure you scrape off the enamel
from the wire ends before soldering
them to pads A-D; otherwise, you
won’t get a good electrical connection
and the radio won’t work. It helps to
tin the ECW ends after scraping them;
if the solder won’t stick evenly, that
means you need to scrape off more
enamel first.
There’s also room for a tinned wire
loop to help hold the toroidal core to
the PCB at upper right. We recommend
you add this to prevent solder joints
from fracturing due to movement over
time. This does not form a shorted turn
as the pads it’s soldered to aren’t connected to anything.
The second toroid (T2) has 15 turns
of 0.6mm diameter ECW for the secondary, which you should distribute evenly. Make sure the direction
of winding is such that the ends go
into holes G & H in the PCB correctly.
Then add the two-turn primary using
0.35mm diameter ECW, scrape and tin
the ends and solder it to pads E & F.
Again, solder the piece of tinned wire
to hold it to the PCB.
Once all components have been
installed, give the board a thorough
clean to get rid of surplus flux. Inspect
all soldered joints and check for any
shorts with an ohmmeter. Make up
the connecting 16-wire cable with
IDC sockets.
Use a small vice or crimping tool to
siliconchip.com.au
Wiring up the boards is pretty straightforward with a ribbon cable connecting
the control and RF boards. There’s room inside the enclosure to fit a three-cell
battery holder which can be used to power the Receiver.
evenly press the parts together; make
sure the cable is exactly square on to
the connector. Attach the loudspeaker
to the two wire connector and plug this
on the control board.
Programming the Nano
The Nano should be programmed
before it is plugged in. You can use
the free programming software called
AVRDUDESS for Windows that you
can download from siliconchip.au/
link/aaxh or use the command-line
version, avrdude, if you’re running
Linux or macOS.
Australia's electronics magazine
Connect the Nano via a USB cable
and check what COM port it appears
as. Select a baud rate of either 57,600
or 115,200 depending on the version of
the bootloader in your Nano. Select the
programmer Arduino for bootloader
using STK500v1 protocol from the
drop-down list; it may be the default.
Press the Detect button and it should
recognise the chip.
It is important that the programming
is done in exactly the following way,
making sure that the settings in AVRDUDESS are correct. Otherwise, you
could end up with a bricked Nano.
July 2025 77
There are two files to be loaded: the
program file, “SSB RX xx.HEX”, where
xx is the version, and the “SSB RX
xx.EEP” file, which is a binary file
loaded into the chip’s EEPROM.
There are five boxes below Options.
Tick “Disable Flash Erase”. Under the
Flash box, “Format” should be “Auto
(writing only)”.
Locate the HEX file to be loaded
into Flash by searching for it in the
square to the right of the Flash window. Locate the EEP file and place
in the EEPROM window. Select Raw
Binary from the drop-down list.
Tick the Write circle under Flash
and press the Go box. This will result
in progress messages in the bottom
window.
Tick the Write circle under the
EEPROM window and press the Go
box. It will also have messages in the
progress window.
This completes the programming,
and the Nano may be disconnected
from your computer and plugged into
the Control Board. Make sure its orientation is correct.
Initial testing
Connect power to the board via
CON1. Make sure polarity is correct;
if not, nothing will happen as there is
a protection diode. The LCD backlight
should be on, but there may not be any
text visible. Adjust trimpot VR6 until
you see text on the screen.
With the Band potentiometer
(VR3) fully anti-clockwise, the top
line should have 3.600.000MHz
and the bottom line USB or LSB,
depending on the position of switch
S2. There should be a cursor visible
under one of the digits. When the
shaft encoder is rotated, this number
should change.
Depending on the particular encoder
used, it may operate backwards. In that
case, bridge the two pads marked DIR
above the LCD to reverse the direction.
Switch the power off and on; the
screen will show “SSB Receiver” on
the top line and version number on
bottom line for two seconds. Toggle
the USB/LSB switch and see that it
changes on the screen.
Press the switch on the shaft encoder
and check that each press moves the
cursor to another position under the
frequency. It should allow adjustments
in 10Hz, 100Hz, 1kHz, 10kHz, 100kHz
and 1MHz steps.
The Band potentiometer is a convenient way to cycle through the most
common amateur radio bands. It sets a
frequency partway through each band
starting with 3.6MHz, then 7.1MHz,
10.0MHz, 14.1MHz, 21.1MHz and
28.1MHz. 10MHz is not actually in
a ham band, but it has WWV transmitting time and accurate frequency
information.
Using an oscilloscope and a frequency counter, check the outputs of
the Si5351 module. CLK2 is the BFO
and that should read 8999.6kHz or
8996.6kHz depending on the position
of the USB/LSB switch.
CLK0 should have a frequency that
is the sum of the currently tuned frequency plus the BFO frequency. The
accuracy of these depends entirely
on the 25MHz crystal attached to the
Si5351 module, so you may get slightly
different values. At this point, it is
advisable to perform calibration.
Calibration
To calibrate the set, you need to
measure the actual frequency of the
25MHz crystal on the Si5351 module. This procedure will calibrate the
short-term accuracy to within less
than 5Hz. Switch off the receiver,
then rotate the Band potentiometer
fully clockwise.
Switch it on and the top line on the
LCD will show “Calibration”; the bottom line will show the nominal crystal frequency of 25,000,000Hz. In this
mode, the frequency on OUT0 is set to
exactly 10MHz.
Fig.16: the recommended hole locations and sizes for the rear panel. A stepped drill bit makes drilling these
straightforward. As this is at actual size, you could copy or download and print it and use it as a template.
78
Silicon Chip
Australia's electronics magazine
siliconchip.com.au
Measure the frequency on OUT0
with an accurate frequency counter
and rotate the tuning knob, which
increments or decrements by 10Hz,
until the output is as close as possible to 10MHz. It’s possible to get to
within 1Hz. It may take many turns,
as the crystal could be out by more
than 10kHz.
Turn the Band potentiometer anti-
clockwise to leave calibration mode.
The new value for the 25MHz crystal overwrites the original value in
EEPROM loaded by the EEP file, and
will be read every time the power is
switched on.
This calibration needs to be done at
regular intervals as the crystal may age
and can also drift with temperature.
Case preparation
The modifications to the case
involve drilling four 3mm holes in
the base to attach the RF board, plus
numerous holes in the back panel for
the power socket, antenna connectors
and the loudspeaker.
I used the Jaycar AS3025 rectangular speaker, but just about any small
8W speaker will be suitable. You will
need to adjust the mounting hole
positions if using a different speaking, though.
Fig.16 shows the suggested hole pattern. I used a stepped drill bit, as they
make clean, circular holes. If you have
a drill press, that would be ideal, but
you can hand-drill these holes neatly
if you’re careful.
A word of warning! The drill can
grab the plate and spin it around, possibly injuring you, so for safety, always
make sure the plate is clamped firmly
while drilling the large holes.
Alignment
This should be done with the two
circuit boards not yet assembled into
the case to allow easier access to test
points. The only adjustments on the RF
board are the three variable capacitors,
and they should be peaked at 9MHz.
The way you do this depends on what
equipment you have.
Connect the control board to the RF
board with the flat cable and carefully
check that the two pin 1s are connected
together, ideally with the red striped
side of the cable to those pins (check
for continuity between the pin 1 pads
on the two PCBs). If you connect the
cable backwards at one end, you could
do damage!
Switch on the power and set the
frequency to, say, 7.1MHz or some
other convenient frequency. Use an
oscilloscope probe to check that you
have the VFO signal at about 16MHz
on TP3 and the 9MHz BFO on TP5.
Connect a signal generator to CON1
with the output set to about 100µV.
This is way above the lowest level,
but is useful for the initial setup procedure.
Tune the signal generator for a
whistle from the loudspeaker, which
should be very loud. Reduce the signal
generator output until you get some
background hiss from the speaker.
Adjust the antenna tuning potentiometer (VR1) for maximum signal, measure the DC voltage on TP6 and tweak
the three trimmers (VC1-VC3) for maximum output.
Fig.17: the front panel PCB overlay for the SSB Receiver. It is shown here at 70%
of actual size.
siliconchip.com.au
Australia's electronics magazine
That’s all there is to it; the receiver
should work across the whole range.
You will find as you tune across
10MHz, the relay will click to switch
between using T1 & T2.
Final assembly
The two boards and front panel can
now be assembled into the case. The
RF board uses M3 × 10mm threaded
spacers to attach to the bottom of the
case.
The antenna input connects via a
short ready-made cable between the
SMA connector to a BNC connector on
the back panel. Power is from a 2.1mm
or 2.5mm ID (inner diameter) DC jack
on the back panel to the two pin connector, CON1, on the Control Board.
The front panel attaches to the case
by four screws on the corners. Instead
of using the zinc-plated screws that
came with the case, I found some black
screws that look better.
All the rotary controls have 6mm
diameter shafts. It is preferable that
these have fluted shafts, as knobs for
these are more common. Still, there
are a few sellers on AliExpress that
have knobs with grub screws (they
are listed in the parts list last month).
Those were ideal for D-shaped shafts
but can also be used on fluted shafts.
Conclusion
Your Radio should now be operational and you can start scanning
the bands for signals! You can even
use it on the go, powered from a 12V
battery.
Note that while you could run this
radio from a 12V vehicle battery, you
must not do so (or connect it) while
the engine is running as it doesn’t
have protection from voltage spikes.
Unless you add a suitable filter, it’s
far safer to run it from its own interSC
nal battery.
When first running the Receiver
you must calibrate it by following
the text under the cross-heading
“Calibration”. This ensures accurate
timing.
July 2025 79
|