This is only a preview of the January 2024 issue of Silicon Chip. You can view 35 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. Items relevant to "Raspberry Pi Clock Radio, Pt1":
Items relevant to "WiFi Relay Modules":
Items relevant to "USB to PS/2 Keyboard Adaptors":
Items relevant to "Secure Remote Switch, Pt2":
Items relevant to "Multi-Channel Volume Control Part 2":
Purchase a printed copy of this issue for $12.50. |
SERVICEMAN’S LOG
Getting amped up
Dave Thompson
Dave gives a bit of background on the various guitar amplifiers he built or
bought, using the technology of the day. They evolved from valve-based
designs, initially made from modified radiograms, through to solid-state
hybrid and discrete amplifiers. All so that he could rock’n’roll!
Years ago, when I was a little ‘un (or wee tacker, as you
Australians might say), I got one of those 10-in-1 electronic
sets from Dad as a present for Christmas. I loved it, and it
became my favourite ‘toy’.
It only had one transistor, a germanium diode (yes, I’m
that old), a ferrite broadcast coil, a variable capacitor, a battery holder, a small speaker and a few resistors and capacitors. Still, I could eventually make more than the nominal
10 projects they published in the manual that came with it.
These days, you can buy the same sort of thing from the
local electronics shop, with 200 or more ‘projects’, but it
was pretty rare back in the late 1960s. I believe that Dad
purchased it in Australia, on one of his frequent trips to
Melbourne to see my Uncle Roger (not the infamous orangeshirt-wearing cook/comedian, before anyone asks!).
I loved that kit, but outgrew it relatively quickly. One of
my favourite projects (besides the crystal radio, obviously)
was the amplifier. It was nothing too fancy, but I could use
it to amplify the audio from another crystal set I’d made
the traditional (for that time) way. I used a multi-tapped
coil wound on a cardboard toilet roll centre, with a germanium diode as a detector.
The signal from that was puny, of course, although
Dad had sourced me a pair of high-impedance crystal
headphones (which I still have somewhere). That made
a huge difference over the standard crystal earbud of the
day. You know the earphone, cream-coloured case, clear
plastic earpiece and a twisted-pair cord. It was precisely
84
Silicon Chip
the type of thing Uncle Arthur had for his hearing aid.
Fun fact: those earphones are still available from your
local electronics store!
So, with my crude crystal set and a substantial long-wire
antenna that Dad helped me set up, I could get some quite
far-off stations, but the audio was weak. Using my 10-in-1
kit set up as an amplifier (connected up using wire into the
kit’s springs), I could now drive the small speaker in the
kit and free myself of the clunky (and, to be honest, quite
uncomfortable) Bakelite headphones.
I often listened into the night with that setup, although
I eventually built a much more advanced shortwave radio
from one magazine project or another. I did use better mid1970s headphones with that, so I could listen late at night
without disturbing anyone.
I would regularly ‘skip’ the likes of the BBC World Service and Radio Luxembourg, which at that time had some
excellent radio shows syndicated from around the world.
Good memories.
Learning to wield the axe
While this was all going on, I was learning to play the
guitar. I’d played the piano by this time for about six years,
but guitar was what I really wanted to learn. I bought a
rather dire electric example from a schoolmate and set
about teaching myself. The first thing I discovered was that
the sound it made was literally nothing, and I needed an
amplifier. Boy, did that open up a rabbit hole of discovery
and expense!
As anyone who has ever bought a guitar amplifier will
tell you, the choices are seemingly endless, and some manufacturers expected your pockets to be almost bottomless!
The irony is that if I’d bought one at the time, it would now
be ‘vintage’ and worth an absolute fortune. However, as a
budding serviceman, I had to fashion my own somehow.
I had seen a few related articles in some of the American
magazines I was buying at the time about how people were
modifying valve (vacuum tube) radiograms or lo-fi amplifiers to use as guitar amps. The main differences were the
input impedances of the preamp and overall gain of the
input stages; you needed enough gain to get some of that
famous valve ‘crunch’.
This was good news for a now-broke high school student. Home stereo systems were rapidly relegating radiograms and similar older console and mantle radios to the
scrap heap. I recall often seeing them sitting on the roadside, offered for free.
Australia's electronics magazine
siliconchip.com.au
Items Covered This Month
•
•
•
Home-made guitar amplifiers
Repairing an Icom 551-D transceiver
Fixing a car head unit
Dave Thompson runs PC Anytime in Christchurch, NZ.
Website: www.pcanytime.co.nz
Email: dave<at>pcanytime.co.nz
Cartoonist – Louis Decrevel
Website: loueee.com
Knowing what we know now, of course, it would be amazing to be able to pick up those antiques and use or restore
them – which, let’s be honest, has since become a huge
part of this hobby of ours. However, back then, they were
big, timber, cumbersome lumps and simply not wanted.
New-fangled solid-state hifi amplifiers and separate component systems soon superseded those in every household
I visited (including ours).
It was good that technology progressed so rapidly in
the 1970s; that meant I had my pick of old valve amplifiers. A friend gave me the amp from his parent’s radiogram
because he wanted to put a transistor amp into the case.
That suited me – I’d be horrified with myself now if I had
dumped some beautiful walnut cabinet. The antique radio
people would blacklist me!
My first guitar amp
But suddenly, I had this lump of an amp. My experience with valves was watching Dad maintaining the blackand-white TV he’d built us in the 1960s. I still remember
him showing me a nice fat arc when he held his Earthed
screwdriver close to the flyback output. I was mightily
impressed, I can tell you!
Of course, he warned me never to try it myself. I’ll leave
it to your imagination as to whether I followed that advice!
So, this dusty old valve amp had several inputs that were
no longer wired in, and I also made sure I got the speaker,
which had an output transformer mounted directly onto
the basket.
First I ensured it was dust-free, then plugged it in and
switched it on. This, of course, is against standard practice for valve amps that haven’t been used in a while. Still,
I knew no different then, and I watched all the filaments
glow, and a quiet hum came from the speaker. I did know
enough to be too scared to go anywhere near it while it
was running!
I shut it down and looked at the inputs. I can’t recall if
they were labelled or not, so I made up a lead with a guitar plug on one end and the older RCA-style connector this
amp used on the other. I plugged the lead into the guitar
and tried each of the inputs, but of course, it was either
weedy and thin or grossly distorted, and not in the nice
way we guitar players love.
At the time, I knew much less about amplifiers than I do
now; even the rare guides I discovered in magazines were
vague or purposely omitted values and figures. I could see
this wouldn’t fly, so I passed the whole thing on to a school
chum who thought he’d like to play around with it. That
left me with no amp (and, full disclosure, no real talent
either at that point!).
siliconchip.com.au
I started looking more seriously into buying a commercial amplifier, but that was simply out of the question. I
started asking around – these are the sorts of things people
buy and then give up on, so some could be quite cheap. A
neighbour down the street had just that: a Christchurch-
made Abby 30, a locally produced clone of the famous
Vox AC30.
The guy who made them eventually moved to Melbourne,
but he made a few of these and other styles of guitar amps
back in the day. I wish I’d kept it, as they are now regarded
as one of the best copies made, and likely worth a small
fortune due to their rarity. I used it for my formative years
though, so I got plenty of use out of it.
When it came to touring, however, those valve amps
became a real liability. The cabinet was solid timber with
two heavy-duty 12-inch (30cm) Celestion speakers sitting
in it and a solidly-made steel chassis sandwiched into the
top; that thing soon broke my rock and roll spirit, not to
mention my back!
Making it more luggable
What I needed was a solid-state ‘head unit’ and a single-
speaker cabinet for playing smaller clubs and bars. Something a lot easier to lug around, that didn’t take up so much
room in the small cars we had at the time (nothing like
today’s monster SUVs!). By then, I had done a lot more
research and read many more magazines, so I thought I
could easily make one.
The first one I made utilised a Sanken Hybrid SI-1050G
50W power module I had purchased a year previously.
I’d been intending to make a small foldback amp/cabinet, but decided instead to use it as the power section
of a guitar amp. The preamp I used was part of a project (if I recall correctly) in one of the English magazines
of the day, perhaps Practical Electronics or Everyday
Electronics.
The power supply was part of yet another guitar amplifier project that was similar to what the Sanken module
required to run it. I could have used either a ±33V split
rail or a 66V single supply; I chose the latter, mainly
because I had a nice beefy power transformer waiting for
such a project.
I had access to a transparency printer and made and
etched my own PCBs, so it all ended up quite good-looking,
and I was pleased with the result.
Australia's electronics magazine
January 2024 85
At 50W, it seemed underwhelming compared to other
50W guitar amps I had used, so I played around a little
with the gain of the preamp and got it running a bit hotter.
Still, it was not cutting through the chaff on stage. Overall,
a bit disappointing, then.
The next build fared a little better. This time, I decided
to add 50 more watts, doubling the apparent power but not
the output sound level, giving little more than a nominal
3dB gain. However, it gave more punch and more headroom to play with.
This module was another locally-produced product, and
a kit to boot. It was a common-for-the-time push-pull power
amplifier design using the perennially popular 2N3055
NPN and 2N2955 PNP general-purpose power transistors.
Many well-liked guitar and hifi amplifiers utilised those
robust and easy-to-source components.
I used the same preamp, but as this module required a
split power supply, I had to redo my 66V single supply,
which was also burgled from the Sanken amp. Fortunately,
the case was large enough to accommodate the new module
and, with a bit of fettling, I soon had a dual-voltage power
supply running at around 32V per side.
That was just outside the module specs, but allowing for
sag and other factors, I figured it should be OK.
In use, this was a solid and reliable amplifier that really
had some punch, especially after I paired it up with a tone
booster, a gain/overdrive pedal and a better speaker (an
Eminence driver) in the single 12-inch cabinet. I used it
for many years on the circuit (pun intended!) before it was
fried one night when someone hit a power pole up the road
and killed the whole club’s power.
I knew I had blown something, and by that time, I was
in a better position to buy a commercial amplifier, which
is what I did.
Valve amp parts are still available
I still have a love for audio amps, though, and have serviced and repaired many vastly different varieties over
the years. In recent years, I’ve also re-embraced the valve
amplifier scene, learning about them by building and making my own.
The transformers and valves, which were always a headache for the average Joe like me to source, are now available,
86
Silicon Chip
both here and overseas. While not always cheap, at least
we can buy them.
I also had the good fortune to stumble upon a commercial transformer winding machine that had served a local
company here for decades. It sat in a friend’s garage for
many more years until he passed it on to me for a peppercorn fee, and I have since tidied it up and wound several
transformers with it.
The different-sized cores and wire are also available (if
not readily), so I’m lucky to be able to wind transformers
to my own requirements and specifications.
There are also many good physical schools and online
classes dealing with designing and building valve amps
(including an excellent one in Australia). If anyone is interested, many good books and tutorial videos are also available on the subject.
Like all amplifier theory and technologies, there is often
heated debate about what constitutes a good design, or
even a great design. The old timers could get it right, but
they also had access to high-quality, inexpensive valves
and very clever people who ate, lived and breathed valve
amplifier design.
These days, there are still a lot of clever people about,
and even those experimenting with operating valves in
low-voltage applications, down to 12V, which is fantastic
for the likes of stomp-boxes and portable guitar amplifiers.
Editor’s note: see our ‘Nutube’ Valve Preamplifier (January 2020; siliconchip.au/Article/12217) that runs from
just 9V DC!
While we don’t have unlimited access to those vast quantities of valves anymore, there are still a lot of NOS (new
old stock) valves available at ever-increasing prices. NOS
refers to parts that have been sitting around forever, but
that have never been used.
As those supplies dwindle, the prices will continue to
rise. While Russian and Chinese-made valves are still being
manufactured, modern valve aficionados claim the quality
of their valves is nowhere near as good as it was back in
the days of General Electric, Philips, Sylvania and others.
Sanctions are also causing supply problems for Russian-
made valves...
To take things even further, the advent of the computer
and amplifier ‘modelling’ technology means that just about
every ‘tone’ some legendary guitar player has come up with
is now available as a patch or preset in a hardware modelling amp, selectable at the touch of a button and able to be
used live at gigs, just like my own (rather crude) home-made
amps. Some modellers are even built into the guitar itself!
For the home recording artist, virtual instruments can
be loaded into a DAW (Digital Audio Workstation), and the
variety of sounds and tones available is almost limitless.
Anyone with a halfway decent computer and a set of studio monitor speakers or headphones can download a free
DAW and have a home studio almost more potent than
many I spent time in during the ‘80s and ‘90s.
It’s a whole different world in audio amplifier design
and implementation now. Yet the basics remain the same –
taking a signal and boosting it through several stages with
minimal unwanted distortion at the output. The distortion
figures on some of the amplifier designs in this magazine
would have been impossible 20 years ago.
I’d hate to think what that 10-in-1 amp’s figures were,
all those years ago!
Australia's electronics magazine
siliconchip.com.au
Icom 551-D transceiver repair
C. K., of Mooroolbark, Vic was asked if he could repair
a 50MHz all-mode transceiver, the Icom 551D, dating from
the 1980s.
It’s an elaborate device and capable of more than 80W
output with SSB or FM and 40W with AM. I answered
rather foolishly, “Yes, I can repair anything”. Little did I
know what I was letting myself in for!
The owner had bought it on eBay in a non-functional
state. He did try to work on it, which resulted in smoke
coming out. Not a very promising start.
It looked like a fairly clean unit and powered up to show
a display, but nothing came out of the inbuilt speaker. Also,
while it showed a sensible frequency on the dial, the tuning knob did nothing.
Removing a few screws allowed me to take off the top
cover, revealing a large single-sided circuit board with
numerous components and a rats’ nest of wires heading
off to several connectors. Fortunately, I could download a
full maintenance manual. After getting familiar with the
various parts, I printed out the schematics on A3 paper so
they were readable.
All the schematics were hand-drawn; unfortunately, very
few components were labelled. The diagrams of the circuit
boards had all the components numbered, so together with
the parts list, I could eventually identify them.
I started by checking voltages. Three TO-220 NPN transistors on the main board: Q28, Q29 and Q30, in association with Q31 and Q32 plus discrete components, provide
regulated supply rails.
The collectors are fed by 4.7W 1/4W resistors from the
13.8V supply, which should give about 13.5V. Q28 and
Q30 gave correct readings but the collector of Q29 measured about 6V (see the diagram below).
On closer examination, the collector resistor looked very
burnt. That must have been where the smoke came from.
With the type of construction prevalent at the time, all
the resistors are standing up on the single-sided board, so
only one end is accessible. Replacing a resistor requires
taking the board out to get at the underside.
While the maintenance manual gave detailed instructions on removing the front panel, it gave no clues as to
how the main board should be removed. So it was up to
me to locate the many screws to be undone and all the connectors that had to be carefully unplugged.
Not only were screws holding down the PCB, but two
power transistors on one side were attached to heatsinks
bolted to the side of the case. All the associated screws
had to be removed, and I had to be careful not to damage the insulating washers between the transistors and
heatsinks.
Finally, after considerable time, I could ease the board
out and get to the underside. If the top was a rats’ nest, the
bottom was much worse! Look at all those extra components tacked on, including a 16-pin chip on a little subboard. I don’t know if these were some kind of modifications or were needed to fix design problems, but to my eye,
it looked like the epitome of bad design.
Editor’s note: they seem like the sort of ‘running changes’
made in a factory when they already have thousands of
boards made and find that a problem needs to be addressed
or an extra function included.
Having located the burnt 4.7W resistor and not having
that value, I replaced it with two 10W 1/4W resistors in
parallel.
Left: these transistors regulate the voltage rails
on the main board.
Below: the shaft encoder circuitry.
siliconchip.com.au
Australia's electronics magazine
January 2024 87
After careful reassembly, I turned on the power and
checked that the output on the emitter of Q29 was 9V.
When I turned the volume up, hiss came out of the loudspeaker. Attaching a signal generator gave a good signalto-noise ratio with an input level well below 1μV. However, the tuned frequency was well off the indicated frequency on the dial.
There is a calibration procedure in the manual that
should correct this. The phase-locked loop (PLL) module
has a 10.24MHz master crystal oscillator from which the
VFO frequencies are generated. A trimmer capacitor on
this oscillator sets the exact frequency.
However, on measuring this frequency at the test point
specified in the manual, I found it was too low and, even
with the trimmer capacitor at its minimum value, was a
long way off. The crystal had obviously aged and dropped
in frequency. I decided to ignore that for now, not having a
crystal with a suitable frequency in my collection.
The next major problem was the tuning knob. Accessing
that meant removing the front panel assembly by undoing
numerous screws and carefully unplugging the many connectors. Attached to the tuning shaft is a disc with slots
around the circumference. On one side are two phototransistors, and on the other side, mounted on a small subboard, are two LEDs.
These are an early SMD type of LED, TLR121, made by
Toshiba. They have a clear lens and provide a point source
at 700nm, a red wavelength. The LEDs and phototransistors
are positioned so that the outputs on the collectors are 90°
out of phase, meaning the rotation direction can be ascertained by the logic.
No light was coming out of either LED and, on removing
the sub-board, they both measured close to a dead short.
Not having any LEDs of the same size, I jury-rigged two
3mm red LEDs. It was an ugly workaround, but on reassembly, I was surprised to find that it worked; rotating the
knob changed the frequency smoothly in 100Hz and 1kHz
steps. Not being too happy with the long-term stability of
such an arrangement, I ordered some M3216/1206-size SMD
LEDs with clear lenses for a more permanent fix.
They arrived a week later. Barely large enough to straddle the hole in the PCB, it was a fiddly job to fit them and
I had to use solder blobs to affix them. Unfortunately, on
reassembly, the tuning knob was not working correctly. It
turned out that the brightness of the LEDs was insufficient
to saturate the phototransistors. Reducing the LED series
resistor from 560W to 330W fixed the problem.
Why both LEDs had failed in such a manner was a mystery. When I mentioned the problem to another contributor to the magazine, Andrew Woodfield, he knew about
this problem. Apparently, those LEDs were used in many
instruments and were notorious for their failure rate. Warranty failures due to those LEDs clogged the workshop.
As for the 10.24MHz crystal, local suppliers did not stock
such a value, but I could get them online from the likes of
AliExpress as long as I ordered a batch of ten and waited
for weeks. However, an associate of mine came to the rescue and gave me a suitable crystal. Having replaced that,
I could now tune it to the correct frequency.
What about the transmitter? My power supply can only
provide 3A at 13.8V. Also, my 50W dummy load is rated at
only 15W. I plugged in the microphone, pressed the switch
and whistled. This pinned the 3A supply meter, the dummy
load got warm, and a pickup loop on the scope showed a
clean sinewave.
This gave me confidence that the transmitter part was
working. Fortunately, the unit’s owner lent me a large power
supply and dummy load, enabling me to check that full
power was available.
Repairing a car head unit
These photographs show the Icom 551-D transceiver at
various stages of repair.
88
Silicon Chip
S. G., of Bracknell, Tasmania found that when you
can’t get replacements any more, you need to have a go
Australia's electronics magazine
siliconchip.com.au
at fixing the fault, even if it’s a bit outside of your comfort zone...
One of my radio club members was selling off one of his
amateur radios, a little Any Tone AT5888. This dual-band
radio covered the 2m and 70cm bands, with a power output of 50W on the 2m band and 40W on the 70cm band.
It was in very good nick. The price was right, so I soon
struck a deal to buy it.
This radio was to go into my car; one of the things I liked
about it is that it had a remote head. The whole front of
the radio can be removed, and one can use a Cat 5 cable to
interconnect the main part of the radio and the head unit.
This meant I could mount the radio in the boot and the
head unit under the dash of my Ford Falcon. The antenna
was mounted off the side of the boot.
The whole system worked well for around 12 months
until I had a car accident, and my car was written off (I was
OK; it was just my pride that was hurt). It took me around
a week to remove the radio; I had to go to the storage yard
to remove all my belongings.
I soon had a new car, a 2017 Hyundai Elantra. Mounting
the main radio was not difficult, but running the power,
Cat 5 cable and external speaker wiring was more challenging. I had to remove several plastic trim panels to run the
cable. I ended up not mounting the head unit, as the only
spot was just forward of the gear lever and under the air
conditioning controls.
I gave that job to the local car radio installation company as I did not want to break the plastic trim. This took
the company three hours, and I paid the going price at the
time, but I got what I wanted.
Several weeks later, I noticed that some of the digits in
the display were missing, making it hard to understand
the letters.
The only thing for it was to try to get a replacement head
unit. This was a total waste of time; after contacting several retailers in Australia and overseas, I came up blank.
It seemed like I would have to put up with the faulty LCD
screen.
A few months later, I thought to carefully push on the
display while the radio was working, and some of the letters
returned, only to go missing again once I took my finger off
it. I realised the LCD’s edge connector might have gone out
of alignment. I figured that taking it apart to check would
be unlikely to make the situation any worse.
I removed the head unit from the car mounting bracket
and took it to my workshop. I soon had the unit apart and
carefully removed the LCD glass and its rubber membrane.
I ended up cleaning the surface edge area of the membrane
with a little bit of contact cleaner, and did the same to the
contact edge of the LCD glass.
Putting it back together was not that hard; I just had to
take my time. The actual LCD glass sits in a tin bracket that
also holds in the rubber membrane that interconnects the
LCD glass to the circuit board. The tin bracket has four little tabs that allow the alignment between the display and
the circuit board to be adjusted, and it can also put a bit of
tension on the display.
Putting everything back together and reinstalling the
remote head unit to the dash bracket, it was working again.
The display had no more missing letters or numbers. That
was nearly nine months ago, and the display has not faltered yet.
SC
siliconchip.com.au
Australia's electronics magazine
January 2024 89
|