Secure Remote Mains Switch, part two (January 2025)
Techno Talk
Testing Times
When I was young, bright-eyed, and bushy-tailed, any test equipment
worth having was too expensive for me to have. Today, by comparison,
I’m amazed by the power, sophistication and affordability of the
offerings available to hobbyists.
T
he topic of testing and the
equipment we use to do so has
been much on my mind of late.
There are a couple of pieces of test gear
I personally could not do without. The
first is my trusty multimeter. Actually,
“multimeters” would be more apt, because I have them all over the place.
There’s one next to me on the desk
here in my office, another on my desk in
my study at home, one in my backpack
for when I’m on the road and others
lurking hither and thither. The amazing thing is how cheap they are these
days (I’m talking about multimeters
for hobbyists rather than professionalgrade devices). You can pick up a very
serviceable unit for less than £10.
Scope it out
Another piece of test equipment that
provides myriad test functions is an oscilloscope. Back in the 1950s and 1960s,
these devices were monstrous in terms
of size, weight and cost.
In addition to a cathode ray tube (CRT)
for display, their rugged metal cabinets
contained a cornucopia of valve-based
electronics. Those valves (also known
as vacuum tubes) used to perform tasks
like controlling the vertical and horizontal deflection of the electron beam(s),
amplifying signals and implementing
switching and timing circuits.
Oscilloscopes like that were still found
in the engineering department when I
started university in 1975. They were
so heavy that you could give yourself a
hernia lifting one, so we wheeled them
around on trolleys.
Things didn’t get much better in the
1980s, even though the valves were
swapped out for transistors and silicon
chips. Although these components were
smaller and lighter, the designers of the
scopes employed lots more of them to
squeeze in more features and functions.
It was like a ‘functionality arms race’
between the various vendors. Also, the
CRT remained in the heart of the beast.
A good professional-grade fourchannel oscilloscope could easily cost
£10,000 back in the day, so this wasn’t
something you would expect to find
on a hobbyist’s desk. How things have
changed. I have a four-channel Rigol
38
DS1054 rated at 50Mz and 1GSa/s with
a large 7-inch LCD screen sitting next
to me on my office desk.
It’s small at 13 × 7 × 5 inches (33 ×
18 × 13 cm), light (6.6 pounds or 3kg),
and affordably priced (currently £300
on amazon.co.uk). It also lets me take
screenshots directly to a USB memory
stick plugged into a port on the front; a
feature worth its weight in gold.
Take a guess
I was once offered a position based
on my stating that my modus operandi
for debugging things was to start with
a WAG (“wild-ass guess”).
“Perhaps you would care to elaborate a little”, said my interviewer. That
was his big mistake. Much like my dear
old mother, the real trick is getting us
to stop talking.
I explained that I would start with
the obvious options. If the system in
question was totally dead, for example,
I would first check to see that it was
plugged in (powered up). However,
if the system was up and running but
functioning erratically or incorrectly,
this was where I would take a WAG
based on how the system was behaving coupled with my past experiences.
Even if I hadn’t seen anything like the
current behaviour before, there may be
clues tickling my unconscious mind.
My justification was that, if it proved to
be correct, depending on the size, complexity, and cost of the system, taking a
WAG could save a lot of time, resources,
frustration, and money. Contrariwise, if
my WAG proved to be incorrect, little
effort would have been expended and
little time would have been lost. I could
then proceed to perform a regular indepth analysis and diagnosis.
I got the job (perhaps in a desperate
attempt to make me stop talking)!
Experience doesn’t hurt
If you look at my Weird and Wonderful
Arduino Projects column, which starts
on page 7, you’ll see that a member of
the PE community called Ian emailed
me with an interesting problem. In his
email, Ian said, “See the attached photo.
I have checked this a million times and
am convinced there are no errors”.
Max the Magnificent
I must admit that my first thought
on reading this was, “Oh Ian, why
dost thou tempt the fates so?” I quickly set up a Zoom call with my friend
Joe Farr because this sort of thing goes
much quicker when there are two of
you doing it.
Between us, Joe and I have a lot of
experience in debugging things. This
is largely due to the mistakes we made
when putting them together in the first
place. This explains why the first thing
we did was to check the power and
ground connections to the integrated
circuits (ICs) mounted on the breadboards in Ian’s photo.
At least, we checked the ground connections. We would have checked the
power connections… but there weren’t
any. I emailed Ian. He added the requisite wires and the circuit functioned as
expected. Problem solved.
Why did Joe and I check the power
and ground connections before all else?
We will never tell!
Trust? I think not!
I’m working on a new project that uses
a 9V DC external power source, which
I need to drop to 5V inside the device. I
can get by with 5V at 1A, but I’d prefer
to have 3A or more to play with.
I could use a regular, cheap-andcheerful linear regulator, but they tend
to run hot as the power rises.
I found some small and tasty buck
(step-down) regulator modules on
Amazon here in the USA. With their
9-36V input and 5V <at> 5A output, these
sounded too good to be true. They were
indeed (too good to be true).
Joe purchased some equivalents on
Amazon in the UK (https://pemag.au/
link/ac2m). His arrived before mine. Joe
tested them using a bench power supply
and an electronic load. Oh dear. The
output voltage started to droop at 1A,
it fell to 4.4V by 2A, and it collapsed
to 0V shortly thereafter.
Thank goodness for test equipment,
is all I can say!
Editor’s note: even some switch-mode
IC manufacturers rate their parts based
on the peak current they can deliver,
which can be considerably higher than
PE
the average!
Practical Electronics | January | 2025
