This is only a preview of the September 2024 issue of Silicon Chip. You can view 42 of the 112 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 "Compact OLED Clock/Timer":
Items relevant to "Pico Mixed-Signal Analyser (PicoMSA)":
Items relevant to "IR Helper":
Videos relevant to "No-IC Colour Shifter":
Articles in this series:
Articles in this series:
Items relevant to "Discrete Ideal Bridge Rectifiers":
Items relevant to "Electric Guitar Pickguards":
Purchase a printed copy of this issue for $12.50. |
SERVICEMAN’S LOG
Turning to the dark side
Dave Thompson
Fear leads to anger; anger leads to hate; hate leads to... suffering
broken garden lights?
It is that time of year again. No, not the tech column
awards (I’ve never been invited!). I’m talking about it being
cold, damp, and dark.
Down here in the lower southern hemisphere, we are
somewhat used to the weather bombs that occasionally
circle up from the Antarctic and blanket this part of the
country with bitterly cold winds, snow to low levels and
lashings of ice thrown in to make things especially difficult.
Last year, we were clever and avoided much of this cold
and frosty weather by going to Europe, where there was a
smoking-hot summer. However, that gets quite expensive,
and it takes months of planning, so it is impractical to go
every year just to chase the summer sun.
On the darkest days here, during June, July and August,
even if we have sun, it is barely warm and hangs very low
in the sky. A 40W incandescent bulb would be warmer!
It is pitch dark from 5pm until 8am. If we get a cloudless
night during these months, the mercury drops like Wile
E. Coyote in a Roadrunner cartoon, and we get hard frosts
well into the negative digits.
Usually, when we have an ultra-crispy morning, we have
a reasonably nice day.
It’s a paltry silver
96
Silicon Chip
lining, but it’s better than rain. If it does cloud over during
the night, the temperature can drop to zero, and everything
will be cold, damp and miserable all day long.
This winter, and what remnants of winter we encountered on our return last year, have mostly been about cloud
cover and rain. It seems those halcyon days of clear and
frosty but dry winter days have gone, perhaps due to climate change or perhaps just because weather is notoriously
unpredictable and a pain in the bunions!
It also turns out my workshop is leaking, which makes
working in it a chore and a potential health hazard, as the
carpets are all damp and don’t get the chance to dry out.
To be honest, the whole garage/workshop needs bowling
and rebuilding. Oh, for a spare 60 grand!
Lighting my way
My point, as usual a long time coming, is that with all this
darkness about, outside lighting is really important. Falling
down the front steps would not only be embarrassing but,
at my age, potentially damaging. It is more than essential
that I have decent motion-activated security lighting and,
even better, outside pathway lighting.
Most of the old-style Par-38-style security lights were
mains powered, so one had to wire them in somehow.
They were great, don’t get me wrong. Still, where they are
mounted, under the barge boards, there was not always
handy mains wiring present, so we usually had to get
a friendly sparky to come in and wire the lights in
for us (I am legally obliged to say this).
My current house was once a single-storey bungalow until a previous owner added a second storey
and made it quite large. The problem is that to gain
access to areas where security or flood lighting needs
to be mounted, I’d have to be rake-thin and as agile as a
circus monkey.
These days, sadly, I am neither of those things! I once
used to crawl around wing tanks in airliners, wiring looms
and basically fighting my claustrophobia, but those days
are long gone. I was chosen for that task because I was
small and thin. All I could do these days would be to use
my body to plug a leak!
A few years ago, I decided to install some decent security
lighting on our driveway and along our pathway. The house
is on a back section, down a long shingle drive. The neighbour’s dog does a good job of letting us know if anyone is
walking up our driveway with a typical territorial protection
Australia's electronics magazine
siliconchip.com.au
Items Covered This Month
• The importance of home maintenance
• A modified Crosley radio
• Distorted and damaged PA speakers
• Repairing the solenoid in a washing machine
• Seismograph coil repair
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
bark. This, coupled with the scrunch of feet on shingles, is
usually enough to let us know if someone is visiting.
However, in the dark of night, and with my workshop
being just inside the gate at the end of the driveway, this
often isn’t enough to know if someone might be creeping up
there. Strategically placed security lights were the answer.
They worked perfectly for the first few years and, once
set up, operated with no problems at all. I have two solarcharged, battery-powered units now: one on a handy telegraph pole at the end of the driveway where it meets our
yard (a story in and of itself) and one mounted on the corner of the house.
I also have a relatively expensive mains-powered unit
halfway between the two. I chose mains powered one for
this location because I could wire it into my garage without any major hassles. (Or at least, my sparky could!). It is
also dazzlingly bright compared to the other two, no doubt
due to it having access to mains power.
the ladder and climb up to the lights’ locations and have
to work on them!
I know what you are thinking: I should have carried out
preventative maintenance in good weather to ensure they
would work over the whole winter. You’d be correct, though
it might surprise you that I did just that.
When the days were getting shorter and the mercury was
dropping, I went around and cleaned the solar panels (on
those that used them) and cleared away the cobwebs of
the spiders that made their homes in the nooks and crannies of my lights.
I also ensured that the Fresnel lenses on the sensors were
not obscured by the usual guano, spider webs or anything
else that might prevent them from working. While they
looked a tiny bit crazed from a few years in the sun, they
seemed clear enough to allow things to work as expected.
The thermal sensor units are sealed, so there was no way
anything could get into them and obfuscate them. The only
way they could malfunction is electronically within the
electronic gubbins or if the lens was somehow obstructed.
Since the lenses seemed clear, I assumed the problem lay
inside the units.
Time to crack them open
The first thing I did was go up and really clean the solar
panels of the battery-powered lights. Admittedly, they
were a little dirty, but not so much that I thought the lights
would not charge.
Nothing is built to last any more
All these lights feature impressive LED arrays and,
when new, were very satisfactory for their roles. Now, not
so much. Over just a few years, the polycarbonate frontages have crazed (likely due to the ultra-high UV rays we
have beaming down on us here because of the ozone hole
– remember that crisis?). The plastic cases inevitably break
down and fall apart for the same reason.
I guess this is the problem with imported stuff that has
not been made to suit our environment. One could argue
that, as they are inexpensive, we can just replace them
every few years and we can all carry on with our lives, but
in my mind, that isn’t the point. If I spend hours installing and setting up something, I expect it to last more than
mere months.
Perhaps that’s just my naive expectations of how things
should be these days, but if I shell out good money (and
bigger money for a ‘proper’ security light rather than some
cheap rubbish from a big box store), I expect a reasonably
long time.
My parent’s security lights, installed by my dad at their
home, lasted for as long as I can remember. I helped him
change bulbs in the latter years, when he couldn’t, but the
unit itself worked for decades.
So, I have three such lights set up in different positions. As I am writing this, not one of them works. If I go
out at 6pm, I’d really like my lighting to come on so that I
don’t trip over something in the dark. Of course, when it
is freezing and raining and generally nasty weather is outside the window, that is not the ideal time to go and get
siliconchip.com.au
Australia's electronics magazine
September 2024 97
old friends, a bank of 18650 cells. A couple had vented
and made a mess of things, so that was likely why it wasn’t
working. No problem; a battery swap should see it going
for another two years – hopefully – before it completely
falls apart.
The second one, though, had six D-sized cells. Well,
they seemed not quite D-sized. They were odd, and like
much of the innards we find in this stuff, had no information on them at all. They would likely measure 1.3-1.5V
factory fresh; I’m reasonably sure of that. But now not one
measured more than 0.7V. No wonder it wasn’t working.
I know they build these things down to a price, but these
were more top-line than others, so I would have expected
a bit more life from the batteries, at least.
Not very repair-friendly
The solar panel assembly simply plugs into the unit’s
main body with a standard barrel connector, so it is easy
enough to unplug it and measure the juice coming from
it. Even in dim, grey weather, I was still getting a healthy
11V (plus change) from the panel, so obviously that wasn’t
the problem.
I was surprised to see so much output from the smallish
panels, even in low-light conditions. There was not much
else that could be at fault, so it was time to demount the
units, open them up and check the batteries.
Getting them down is the first problem. Weather, especially with the extremes we have here, does weird things
to screws and plastic, so taking all that off was a bit of a
mission. I think I need to invest in better-quality screws!
There’s not much I can do about the plastic breaking
down, but even the cadmium-plated big-box-store superscrews I’d used had corroded and most of them just sheared
off as I was trying to undo them. The ones screwed into
the telegraph pole all broke off at the holes, so perhaps
whatever they treated the pole with to stop it from rotting
had a detrimental effect on the screws; they seemed especially weak.
Once off, I got the units into my workshop. As you can
imagine, they are a little grubby. Birds tend to sit on them
and, well, you know. But at least the units came apart easily once I’d given them a wipe-down on the outside.
Mostly, they are pretty well made. I doubt they are
designed to withstand our summer sun, but the plastics
seem to have stood up well. All the screws holding things
together have little O-rings on them; another nice touch.
The covers have a recessed O-ring as well, I guess just to
keep the worst of the weather out of them.
That’s a question for the techies: is a square seal still an
O-ring? The things we think about!
Once open, I could see that light number one hosts our
98
Silicon Chip
The problem I had now was how to replace them. As
usual, they are all spot welded together using nickel links
with perma-soldered connections to the PCB. Finding
replacement batteries shouldn’t be too hard, but finding
them with solderable tags on them is a whole other story.
I have one of those cheap spot-welders purchased from
the usual Chinese sites, but it seems to kill batteries; not the
batteries I’m trying to weld, but the model car/aeroplane-
type battery packs that power it. I’ve had two high-capacity
batteries for it now, and they’re not as cheap as they used
to be (nothing is!). Both failed internally after only a few
uses of this welder.
It must suck a tremendous amount of power out of the
battery, but either the batteries are just poorly made, or the
welder itself has some huge back-EMF that kills the cells or
fries some fuse. I’m not about to pour more money down
the drain buying expensive high-capacity batteries in an
effort to get that working.
There are far better options, but as I don’t do much of
this work, it would end up like the treadmill still taking up
space in my garage – used for a while, then forgotten about
or pulled out once in a blue moon to be utilised.
I did manage to find some 18650 cells with solder lugs;
they’d have to do, and that got light number one back up
and running. Light number two with the bigger batteries
was a little more work. While I could find cells, I couldn’t
find any readily available with solder lugs, so I had to use
my soldering station.
That is not ideal because adding that kind of heat is
detrimental to this type of battery. However, I rubbed the
contacts clean with a diamond file just before I soldered
them and used flux, so the solder flowed well onto the
joints. By spreading out the soldering process, I didn’t
get too much heat into them. That light is now working
again as well.
Light number three, the mains-powered one, the most
expensive and brightest by a wide margin, is 18 months
old now. Obviously, it is out of warranty (darn it), and
while it triggers, it is almost like a camera flash. It will
not lux adjust or stay on. I took it down and opened it
up, but the electronics are potted, and there is nothing to
see there, so for this one, the only option is replacement,
and that bites.
The LED arrays and the rest of it, while very well-built
and durable, are all just junk now because the brain is
dead. While it cost more than the others, it is still not
really worth digging into that much before it becomes one
Australia's electronics magazine
siliconchip.com.au
of those jobs that are just too complex to solve rather than
just buying another one. Yet another instance of built-in
obsolescence...
At least I got the other two back up to par (heh) and can
now see where I’m going at five o’clock of an evening. And,
as an added bonus, anyone coming up the driveway will
be flooded with light, which is good for good guys and bad
for bad guys (of which we know there are a few around).
Vintage radios and the modified Crosley set
I have held a ham license for around 65 years. When I
got my license, the exams were only held twice a year, in
February and August. They consisted of a 2½-hour paper
on theory, half an hour on regulations, and the demonstration of 10 words per minute Morse code transmission.
In many cases, the exams were held in the local post
office as the Post Master who oversaw the exams was usually quite proficient in Morse code since telegrams were
sent and received by Morse, and sometimes he had to fill
in for other staff off sick or on holidays.
Naturally, I grew up with all-valve equipment. The first
transistor I purchased was an OC70, a germanium transistor
that cost me nearly half my weekly wage as an apprentice:
4 pounds, 8 shillings and 6 pence, around $9.00.
I have been involved in repairing and modifying a large
number of old valve radios for many years. I lived for each
month when Radio & Hobbies, then Radio, Television &
Hobbies and finally Electronics Australia came out. Many
hours were spent poring over circuits for transmitters,
receivers, amplifiers etc.
I have built up an extensive collection of valves of all
types, as well as capacitors and resistors from the valve era.
I always try to make the repairs look as original as possible.
To that end, I found that I could carefully split old mica
capacitors in half using a small hand-held grinder. I can then
make a pocket inside the shells, insert a modern greencap
or polyester capacitor of the required value inside, then
cement the shell back together.
For the paper capacitors, I cut off the lead on one end and
carefully drill out the insides, replacing it with a modern
one and then closing the end with beeswax. This results
in a very original-looking unit.
Unfortunately, replacement power transformers for valve
radios are becoming very hard to obtain. I have endeavoured to use my metal lathe to wind new windings and
resurrect some, but it becomes impossible unless I know
the turn ratios.
One of my pet hates is the American transformerless
radio chassis. For some reason (probably cost-cutting),
many US-made radios do not have power transformers.
They use valves with the heaters in series, quite often with
17V, 25V, or 50V heaters to make the heater chain add up to
the 110V AC mains. These valves also frequently have low
plate voltages, although some radios use voltage-doubling
circuits to get a higher plate voltage.
These radios are deadly. They rely on the operator inserting the power plug into the GPO the correct way around.
However, because many power leads only have two pins
on the plug, it is very easy to make the chassis live. Most
of these radios don’t have an Earth wire because it would
blow a fuse if plugged in the wrong way around!
siliconchip.com.au
Servicing Stories Wanted
Do you have any good servicing stories that you would like
to share in The Serviceman column in SILICON CHIP? If so,
why not send those stories in to us? It doesn’t matter what
the story is about as long as it’s in some way related to the
electronics or electrical industries, to computers or even to
cars and similar.
We pay for all contributions published but please note that
your material must be original. Send your contribution by
email to: editor<at>siliconchip.com.au
Please be sure to include your full name and address details.
Besides using a variac when servicing these radios, I
always use an isolating transformer to ensure my safety.
There were some really well-made US radios, usually of
the TRF type, before superhet radios became common.
I had a lady bring in a Crosley radio one morning complaining that it had “blown up”. It was a five-valve transformerless superhet with several strange valve numbers
because of the series heater chain. Further questioning
resulted in her telling me that they had owned the radio for
many years and it was a wedding gift from her husband’s
father when they lived in the USA.
It apparently had been “modified” by a radio tech in the
USA before they came to Australia to suit the higher Australian mains power voltage.
I put the radio on the workbench and tried to find the
so-called modification she had claimed was done. Everything looked OK, but further checks showed that most of
the valve heaters were open-circuit.
She said that the radio had worked for years in their old
house, but when they moved into the retirement village,
the power lead was far too long, so her husband cut about
two metres off it and
reinstalled the
plug.
Australia's electronics magazine
September 2024 99
I must be getting old because it took me fully five minutes to realise that the power lead he had cut was a ‘resistance’ lead that dropped the 230-240V AC in Australia
to the 110-120V AC that’s common in the USA! Unfortunately, the radio was beyond economical repair due to the
high applied voltage.
I am currently working on a timber-cased STC 528 that has
seen better days. It is working again, but it is still very deaf.
Still, it keeps me out of the pub and busy at 80 years of age.
J. A., Narangba, Qld.
hearing was like that caused by a rubbing voice coil in
the woofer. It was slightly gritty, but the speaker was still
capable of going loud. I took the grille off the woofer so I
could push the cone to see if it was rubbing. It felt fine, so
I proceeded to remove the amplifier module.
Probing the woofer output with an oscilloscope while
playing music didn’t immediately reveal anything. The
waveforms looked musical and were swinging nicely in
both directions. However, when I fed in a sinewave, I could
see the waveform wasn’t quite as smooth as the input.
I disconnected the speakers so I didn’t have to listen to
the tone and proceeded to trace the signal through the circuit. The output of the preamp looked fine. From there, it
went into a voltage-controlled op amp used as a limiter
(to protect the speaker drivers). The output of this stage is
where the distortion appeared.
This part of the circuit has only the op amp and four
resistors; the control voltage comes from another op amp
that rectifies the audio signals from both the woofer and the
tweeter. The rectifier stage is fed with different amounts of
signal from each driver so it can limit the signal at different levels, depending on whether the overload is HF or LF.
I could see with the ‘scope that the rectifier stage was
doing what it should.
All this suggested the limiter op amp (BA6110) IC was
faulty, but I didn’t have any on hand, and they are now
obsolete. To prove my theory, I removed the op amp and
linked between its input and output to see what happened.
The result was nice clean audio, just a bit low in level.
Then I remembered I had one of these amp modules in
the ‘graveyard’.
Ten minutes later, I had the BA6110 out of the donor
and into the customer’s amplifier, but the fault was still
there! A closer look at the circuit diagram revealed a 47kW
resistor from the +15V rail to a pin on the BA6110 labelled
“bias”. It measured as an open circuit. A replacement resistor restored proper operation.
Several weeks later, the customer delivered another
identical speaker, this time with no HF output. Some HF
output was apparent when I tested it, but not much. Testing the amplifier module indicated that all was well, so I
removed the horn driver for inspection. This revealed that
the diaphragm had shattered!
The voice coil was intact, but not much was left of the
diaphragm. A new horn driver had it sounding good again.
P. M., Christchurch, New Zealand.
Another tale of two speakers
Simpson washing machine solenoid repair
A customer dropped off a powered PA speaker for repair,
saying it sounded distorted. I played some music through
it, and indeed, it did sound distorted.
Problems like this present a quandary to me, as I am
unsure whether to inspect the amplifier module or to start
with the speaker drivers. Many modern, powered speakers have Class-D amplifiers that can deliver hundreds of
watts. It is not uncommon to see labels on the speaker grille
claiming 2000W, which is usually (!) a peak value.
However, if the RMS value is only a quarter of that, it
will still need very substantial drivers to handle the power.
[I think you can drop a zero to get closer to the RMS power
rating from these inflated figures – Editor]
This particular speaker is an older model with conventional amplifiers and solid drivers. The distortion I was
I refurbished a Simpson Contessa washing machine
about two years ago and wrote it up for the October 2022
Serviceman’s Log (page 80). It had been working well until
recently, when my wife found that it was not spin-drying
the clothes but just bunching them up in one place and
then going out of balance...
I suspected the spin solenoid was at fault. If that was
the case, it could be a problem, as I don’t have any spares
left now, having used several over time for repairs to various washing machines. This particular component has a
higher failure rate than others for some reason.
I started by removing the machine’s lid to access the lid
switch so I could hold it in while I turned the machine on
in the spin cycle. Sure enough, the agitator started turning, and the familiar clunk of the solenoid was not present.
The horn driver from a PA
speaker with a shattered
diaphragm.
The Simpson
washing
machine
solenoid
had one of
its terminals
break off.
100
Silicon Chip
Australia's electronics magazine
siliconchip.com.au
I disconnected the machine and pulled it out so that I
could access the back panel and remove the seven screws
that hold it on. I then turned the machine on its side to
access the spin solenoid underneath it.
One of the terminals had broken off, but the solenoid’s
core was still moving freely, and there was no sign of overheating. That was a good sign, as it meant I could probably repair it.
I removed the two #3 Philips screws holding the solenoid
on and turned the solenoid on the elliptical pin to remove
it from the machine. Next, I got my multimeter and checked
that the wiring was still intact; it was. Another good sign.
I plugged in my 20W soldering iron and, while it was
heating up, got a pair of long-nosed pliers and a scraper
to clean the broken terminal where I would be soldering
it back together.
I tinned both parts of the broken terminal and, holding the
loose piece with the long-nosed pliers, I applied heat and
soldered the piece back onto the solenoid. After cleaning
the terminals, I checked the solenoid again with my multimeter, and it was all good. The repair might not look as
good as new, but it would get the machine working again.
I refitted the solenoid in the machine, reattached the back
panel and stood the machine back up again. I refitted the
lid, grabbed some wet washing, loaded the machine, set it
on the spin cycle and pulled up the timer knob.
The familiar clunk was present and, after the pump ran
for a short time, the machine started spinning and ran for
a few minutes before stopping. I checked the clothes, and
they were as dry as usual, so the machine was back in action.
My wife was happy to have the machine working again,
and the repair cost nothing but a bit of time. These solenoids
are not readily available and cost around $50 or more, so
repairing the old solenoid at no cost was a win. It’s very
handy being able to do our own repairs; it saves a fortune
in call-out fees, and the old Simpson lives on.
B. P., Dundathu, Qld
Seismograph coil repair
My seismograph stopped responding some time ago and
I finally got around to fixing it. The detector coil is 50,000
turns of 0.1mm diameter enamelled wire, about 50mm in
diameter. It had gone open circuit, not at the connections
but internally. Still, it lasted about 40 years!
I remembered the story of a bloke who had an open-
circuit coil in a radio IF stage. He connected a 500V bridge
megger to the ends, wound the crank fairly briskly, and it
reconnected the coil. How? Electrostatic attraction? Punching through a corroded spot? Who knows, but it worked.
The sensing coil carries almost no current, less than
microamps, and is in a strong magnetic field, so how long
would such a repair last? I don’t know, but for some time
now, I’ve been thinking of upgrading the detector to a lightbased one that will also provide a DC resting graph.
The magnetic one only responds to definite movement, whereas the commercial ones respond to very low-
frequency, almost DC movement.
Anyway, after doing that, it is working again, ready to
detect quakes anywhere in the world. It is really that sensitive. I feed its output and that of the electrometer into
two channels of a four-channel data logger connected to a
small notebook PC.
P. L., Tabulam, NSW.
SC
siliconchip.com.au
Australia's electronics magazine
September 2024 101
|