Max’s Cool Beans
By Max the Magnificent
Flashing LEDs and drooling engineers – Part 23
H
oly socks, Batman! I simply
cannot believe it’s Crimble
season again. I don’t know
whether I’m coming or going, I don’t
have a speech prepared, and I don’t
have a thing to wear. One problem is
that there are so many fun things to do,
but so little time to do them all. A related problem is that I’m doing so many
fun things that my poor old noggin is
jam-packed with ‘stuff’; for example...
Toroidal transformations
For reasons that will be made clear in
the fullness of time, I’ve been cogitating
and ruminating on the maths associated
with the topic of twisting a torus (plural
tori), a.k.a. a donut (and it’s not often you
expect to hear yourself say that).
Actually, full-blown tori are still sometime in our future. Before we get there,
let’s start by imagining that we cut a strip
of paper. Let’s call the corners at one end
of the strip A and C, while the corresponding corners at the other end are called A’
and C’. Now visualise bending our strip
around to form a circle (Fig.1a). If we join
the two ends ‘as-is’ without twisting the
strip so that A’ is connected to A and C’
is connected to C, we end up with a ring
that has two edges (top and bottom) and
two sides (outside and inside).
Now visualise what happens if we give
one end of the strip a half twist before
joining the two ends, such that A’ is connected to C, and C’ is connected to A. In
this case we have created a Möbius strip
(a.k.a. Möbius band a.k.a. Möbius loop)
that has only one edge and one side (A ->
A’ -> C -> C’ and back to A again).
I’m sure you’ve seen this before, but
on the off chance you aren’t convinced,
why not try making one yourself just for
giggles and grins. Finally, if we give the
end a full twist before connecting the two
ends, we return to having two sides and
two edges. And so it goes – twists of 0, 1,
2, 3... result in two sides and two edges,
while twists of 0, ½, 1½, 2½, 3½… result
in one side and one edge.
So far so good. Now, suppose we create
what we might call a ‘cross-strip’ (Fig.1b).
In this case, we might call the corners at
one end A, B, C, and D, while the corresponding corners at the other end are
44
called A’, B’, C’, and D’. Once again, let’s
start by visualising bending our strip
around to form a circle and joining the
two ends ‘as-is’ without twisting the strip
such that A’ is connected to A, B’ is connected to B, C’ is connected to C, and D’
is connected to D. In this case, we end
up with four edges (A -> A’, B -> B’, C
-> C’, and D -> D’) and four sides (both
sides of the plane delimited by A, C, C’,
A and both sides of the plane bounded
by B, D, D’, B).
Now for the questions... we’ll start with
the low-hanging fruit in the form of Q1:
How many edges and sides will we have
if we give one end of our cross-strip a full
twist before connecting the two ends (so
A’ is connected to A, B’ is connected to B,
C’ is connected to C, and D’ is connected
to D)? Q2: Only slightly harder, how many
edges and sides will we have if we give
one end a half twist before connecting
the two ends (so A’ is connected to C, B’
is connected to D, C’ is connected to A,
and D’ is connected to B)? Q3: This one
makes my head hurt; how many edges and
sides will we have if we give one end a
quarter twist before connecting the two
ends (so A’ is connected to B, B’ is connected to C, C’ is connected to D, and D’
is connected to A)? (The answers can be
found at the end of this column.)
Electronics that work
Gas-heated soldering iron
Have you ever heard of a gas-heated soldering iron? Note that I didn’t say ‘gas-powered,’ which might conjure up images of
a splendid steampunk contraption. What
I’m talking about is my very first soldering iron, which was literally a small cylinder of iron that was shaped into a point
at the business end. The other end was
connected via an iron rod to a wooden
handle, which (in my case) was held by
someone who didn’t have a clue what
he was doing.
This was circa 1969 and I was 12-years
old at the time. I have no idea where this
iron came from or who gave it to me. I think
it was originally intended for working
with metal, not electronics, but it was all
I had, and I was happy to have it. I would
heat my iron over the largest gas ring on
our cooking stove in the kitchen until it
glowed red-hot. Then I would race up the
stairs to my bedroom to solder as many
joints as I could before it cooled down too
much to continue. This provided great
exercise both physically (running up and
down the stairs) and mentally (trying to
work out why my circuit wasn’t working
due to dry/cold solder joints).
When I eventually managed to save
enough of my pocket money to purchase
an electric iron, I opted for a cheap-andcheerful standalone device that came
equipped with a small metal stand to hold
the hot end away from the work surface
(the bedroom floor in my case).
Apart from a single foray into a fullblown soldering station, which ended
when I dropped it, thereby transmogrifying it into a paperweight, I’ve stuck with
standalone soldering irons until just a
couple of days ago. What happened was
that the nice folks at Tilswall ran across my
Apropos of nothing at all, earlier today
as I pen these words, my chum Steve
Leibson posted a review of a rather interesting book called Designing Electronics that Work by Hunter Scott – see:
https://bit.ly/3qk3qBq
I hold Steve in high regard, so if he says
a book is worth reading, I believe him. In
fact, Steve goes so far as to say: ‘Whether
you’re a maker, hobbyist, novice design
engineer, or one of
Twist= ?
Twist= ?
those seasoned grey
A
A’
A
A’
hairs who know how
B
B’
to hold their tongue at
D
D’
exactly the right angle
when making adjustC
C’
C
C’
ments, the PDF version
(a) Single strip
(b) Cross-strip
of this books is free, so
you should download
Fig.1. A simple strip of paper is more complicated than it looks!
a copy today.’
Practical Electronics | January | 2022
on a single Arduino Nano. If you search
on Github for the FastLED library, there
is an example called ‘Noise,’ which explains a lot better than I can how it works.
I copied the code and removed the bits
I didn’t need. There are a few variables
you can set for speed and level of depth
– I just used a simple pot on an analogue
input to alter the scale in real time.’
On occasion, I may have mentioned
my 12×12 array of ping-pong balls, where
each ball is equipped with a tricolour
LED. In fact, I vaguely recall dropping it
into the conversation in some of my Cool
Beans columns (PE, June 2020 to January
2021 and April 2021 to July 2021). The
point is, as soon as I get a free moment,
I’m going to download the example code
Paul talks about and tweak it to run on my
ping-pong ball display. I shall, of course,
report back in a future column.
Fig.2. Work starts on my new robot head with my new Tilswall soldering station (upper right).
Cool Beans Blog (https://bit.ly/2ZY4waZ)
and emailed me to ask if I’d like them to
send me one of their soldering iron stations (https://bit.ly/3BuhbiW) to use on
my hobby projects.
They caught me at just the right time,
because – as I discussed in my previous
column (PE, December 2021) – I was poised
to commence work on a new robot head
with servos and a corresponding control
system. A few days later, my new Tilswall soldering station arrived and I set
to work (Fig.2).
I’ve only been using this little beauty
for a couple of days, but I have to say that
I’m very impressed. This antistatic, ESDsafe station is a substantial unit that I believe offers amazing value for the price. It
also comes with a collection of five different bits and a small spool of lead-free
solder although – if the truth be told – I’m
afraid I’ll continue to use my trusty 63:37
Fig.3. The Fool’s Lantern (Image source:
Bad Dog Designs)
Practical Electronics | January | 2022
silver-lead solder (‘you can’t teach an old
Max new tricks,’ as they say).
Flashback to the 1970s
I’d like to say that I remember the 1970s
as though they were yesterday, but a lot
of Pooh Sticks have passed under the
bridge since then. What I do remember
is the abundance of groovy text fonts and
the proliferation of psychedelic colour
schemes. At that time, one of the big lighting effects that made its presence felt at
the discos I attended as a student was from
oil wheels. In fact, I just found a video
on the Funky Parrot website (https://bit.
ly/3qidtHe) showing a Selection of 1970s
Disco Lighting 6” Oil Wheels that took
me back in time (https://bit.ly/3qkNgYr).
The thing is – several times over the past
few years – I’ve seen swirling effects presented on tricolour light-emitting diodes
(LED) matrices that remind me of those
old oil wheel displays. And each time
I’ve seen something like this I’ve thought,
‘I’d like to do something like that,’ but I
haven’t known where to start.
Then, just when I was least expecting
it, my chum Paul Parry who owns Bad
Dog Designs (https://bit.ly/3BWMjb9) sent
me an email with a link to a YouTube
video of one of his latest creations that
he’s named the Fool’s Lantern (https://bit.
ly/3H6GMCv). This link was accompanied
by a message saying, ‘I made this earlier
on this week and thought you might like
it because it ticks lots of boxes.’ Basically, this is one of Paul’s famous Nixie tube
clocks presented in an antique wooden
cabinet and augmented by a 16×16 matrix
of WS2812B tricolour LEDs (Fig.3).
The video shows lots of different lighting
effects, but one that caught my eye was the
swirling oil wheel-type effect that starts
around 1:00 into the video. When I asked
Paul about this, he replied: ‘The code is
pretty simple and it runs quite happily
Teachable moments
If you are an old hand at constructing electronic doodads, you may want to jump
forward to the next topic. However, I do
receive a lot of emails from readers of
this column who are just starting out in
electronics and – while I was commencing work on my new robot head – I jotted
down a few notes that might be of interest to beginners.
Let’s start with the concept of static
electricity. When we walk around or rub
against something, electrons can be transferred from one object to another, leaving
one of the parties with an excess positive
charge and the other with an excess negative charge. We call this, ‘static electricity’.
If we accumulate such a charge and then
come close to a conductor with a path to
ground, the result will be an electrostatic discharge (ESD) event, which may include a spark that makes life exciting for
a while. If the conducting path happens
to pass through the pin of a semiconducting device like a diode, LED, transistor,
or integrated circuit (IC) (a.k.a. a silicon
chip), then – much like Monty Python’s
Dead Parrot sketch – we may well find
ourselves looking at an ex-semiconducting device with little tears running down
our cheeks.
The solution is to ensure that everything – our components, our tools (eg,
soldering iron), and ourselves – is at the
same electric potential before we touch
anything. As a reminder, you may recall
that I described the Tilswall soldering station we discussed earlier as being antistatic and ESD-safe. Now you know why
this is an important attribute to have in
your soldering station.
Take another look at Fig.2, which features a folding table I set up in the corner
of our family room to work on. My wife
(Gina the Gorgeous) is more than happy to
let me do this, just so long as everything
is cleared away by the end of the day (I
45
The solution is that
my chum Rick Curl
used to work for
a company where
they stripped this
braid off large
power cables and
– unbelievably –
threw it away because all they were
interested in was
the insulated core
cables. As a result,
several years ago,
Rick gifted me with
a lifetime supply
(Fig.4).
Once you’ve
wired your board,
Fig.4. Yours truly with his stash of copper braid that will protect a before you insert
plethora of projects from ESD.
any ICs and connect the power
supply, use your multimeter to check
so want to have my own workshop).
that everything is as it should be; ie,
There are several things to observe in this
that all of the things that are supposed
image, starting with the antistatic mat
to be connected to the ground terminal
covering the table. This mat is groundare indeed connected to the ground tered through the yellow banana plug to
minal, that all of the things that are supthe power strip. Similarly, the antistatposed to be connected to the power teric wristband (middle left-hand-side) is
minal are indeed connected to the power
grounded through the blue banana plug
terminal, and – most importantly – that
to the power strip.
the power and ground terminals aren’t
Now observe the small proto-board
connected to each other.
with the 18-pin IC socket and the three
Next, still without populating your
toggle switches sitting next to the blue
ICs, apply power to the board while
wire strippers. Using IC sockets is recombeing poised to immediately disconnect
mended for two reasons. First, it lets us
it if you hear any weird sounds, see any
easily swap out our silicon chips if they
strange smoke, or smell any unexpected
are damaged for any reason, like static
odors (all phenomena with which I have
electricity, for example. Second, they
had much acquaintance over the years).
let us easily confirm that the power and
This is a perfect time to recheck your
ground wires go to the right pins withIC data sheets to ensure that the power
out forcing us to discover they don’t by
and ground pins are where you expect
blowing up our chips.
them to be, and to use your multimeter
The point is that, even when we are
to ensure that these pins are connected
working on a board that is not yet poputo the power supply as they should be.
lated with ICs or other semiconducting
Only now should you power the board
components, it’s good practice to always
down, insert the ICs, power it up again,
work on an antistatic mat and wear an
and perform another check to confirm
antistatic wristband. If you always do
that everything is tickety-boo.
this so that it becomes ‘second nature,’ it
will save you a lot of grief down the road.
Trust me on this because I know whereof
Feel the joy!
I speak. There’s an old saying that goes
Returning once more to Fig.2, observe
something like, ‘Experience stops you
the grey wooden platform in the center
making mistakes, but it’s making mistakes
of the picture. The two black protuberthat gives you experience.’ The real trick
ances are JH-D400X-R4 10K 4-axis joyis to learn from someone else’s mistakes.
sticks, which you can find all over the
As an aside, given a choice, when I’m
place, including Amazon and eBay (Fig.5).
building one of my larger hobby projUnfortunately, they typically don’t
ects, like my Prognostication Engine
come with documentation, which can
(PE, March 2020), for example, I like to
make life interesting for beginners. Before
protect my cables by wrapping them in
you discard the box that the pot came in,
copper braid whose ends are connected
look for a small packet containing four
to ground (you thread a bunch of wires
tiny fixing screws that will be a pain in
through the hole running down the center
the nether regions to replace if you acof the braid, and then pull the ends of the
cidentally throw them away (once again,
braid, which causes it to shrink in diamI speak from experience).
eter and hug the wires). The problem is
The hard plastic ring with the four holes
that this braid is ferociously expensive.
can be lifted away. Less obvious is the
46
fact that the soft inner plastic cover with
the ripples can be slid up the joystick towards the knob (it may be a tight fit, but
it will slide). This exposes four tiny holes
in the corners of the case, which we will
use to attach the joystick to our console
(the grey wooden platform in my case).
What we do is drill a 1.5-inch-diameter
hole in the console surrounded by four
small holes for the screws (use the hard
plastic ring as a template). Push the knob
up through the hole from the bottom
while squeezing the soft inner plastic
cover so that it also comes through the
hole to be on top of the console. Keep on
raising the joystick so it presses against
the bottom of the console, slide the soft
plastic cover down until it rests on the
top of the console, return the hard plastic ring over the top of the knob, and use
the four tiny screws to secure the beast.
Now, although these purport to be 4-axis
joysticks, we should note that they are
really only 3-axis devices that consist of
three 10kΩ linear potentiometers (pots)
and a normally open (NO) pushbutton
switch on the top (the creators of these
devices are using literary license to persuade us that the pushbutton represents
a fourth axis of control).
Let’s remind ourselves that a pot has
three terminals, two of which are connected to either end of a resistive element (10kΩ in our case), while the third
is connected to a movable wiper (Fig.6a).
A pot can be used as a rheostat (variable resistor) or as a voltage divider. We
will be using our pots as voltage dividers.
This means that we will be connecting
power to terminal (1) and ground (0V)
to terminal (2), or vice versa – it doesn’t
really matter which way round these go
because we can swap the final readings
over in software. With respect to this
design, power will either be 3.3V or 5.0V
depending on what sort of MCU we are
using. If you are using a MCU like an
Arduino Uno, which is powered by 5V
and whose input pins support 0 to 5V,
then you will use a 5V supply for your
pots. In my case, I’m using a Teensy 3.6,
Fig.5. JH-D400X-R4 10K ‘4-axis’ joystick.
Practical Electronics | January | 2022
2
(b) Pot
3
3
2
2
2
Pot 0
Pot 1
Pot 2
Pot 3
Pot 4
Pot 5
In Fig.6b, I show the wiper terminal as
being the one in the middle. This is the
way things will be with more than 99%
of the pots with which you come into
contact, but it is possible to run across
other configurations.
Here’s the way I did things. For each
pot, using my trusty multimeter, I measured between terminals 1 and 2 to
ensure that the resistance value was
indeed 10kΩ, give or take (GoT). Remembering that, in the case of this type
of joystick, the stick automatically returns to its center position, for each pot
I checked that the resistance between
terminals 1 and 3 was 5kΩ GoT; similarly, between terminals 2 and 3. Finally, I checked that the resistance between terminals 1 and 3 varied between
0kΩ and 10kΩ, or vice versa, when the
pot was pushed or rotated to its limits;
similarly, between terminals 2 and 3.
You may feel that all this is overkill
but – once again – let me reiterate that
it’s a lot easier to observe and identify
problems when the pots are standalone
than it is when you’ve connected them
all together.
And, speaking of connecting them all
together, what we are going to do now
is to connect one side (which we call
terminal 1) of all of our pots together.
In the fullness of time, we will connect
this side to our power supply. Similarly,
we are going to connect the other side
(which we call terminal 2) of all of our
pots together. In the fullness of time, we
will connect this side to our ground (0V).
The result is that our six pots are wired
in parallel (Fig.7). Remember that I’m
using a Teensy 3.6, so I’ve shown the
power as being 3.3V, but it may be 5V
in your case. At this stage, however, this
is irrelevant because we haven’t actually
connected our pots to our power supply
yet. Also, we haven’t yet connected their
wipers to our MCU’s analogue input pins.
So, if you were to set your multimeter
to measure resistance and connect one
probe to the wire connected to all of the
Practical Electronics | January | 2022
3
2
which is also powered by 5V but whose
inputs will only tolerate 0 to 3.3V, so I’m
using a 3.3V supply for my pots (this
supply is generated by a regulator on
the Teensy 3.6).
Returning to Fig.5, the two pots we
can see on two of the sides respond to
forward/backward and left/right motions
of the joystick. The third pot, which is
hidden inside, responds to our twisting
the stick clockwise or anticlockwise.
When we take our hand off the stick, it
returns to its center position with respect
to all three pots.
It’s up to us to add wires to the forward/backward and left/right pots. When
it comes to the third (internal) pot, observe the five wires coming out of the
base. Two of these wires will be the
same colour; these are the ones that are
attached to the pushbutton switch. In
my case, these two wires are yellow,
but they are blue on the equivalent pots
purchased by my co-conspirator Steve
Manley in the UK. All we have to do
is connect one of these wires to ground
and the other to the input of our microcontroller (MCU). Actually, in my case,
I’ll be using a special switch debounce
IC, but I’ll be talking about this more in
a future column. If you do connect the
switch wire directly to the input of your
MCU, then add an external 10kΩ pull-up
resistor or declare this input as being of
type INPUT_PULLUP in your code (assuming you are using the Arduino integrated development environment (IDE)).
The remaining three wires are connected to the internal pot. Once again,
their colours may vary depending on the
origin of the pots. In my case, these wires
were white, black and red. My knee-jerk
reaction was to assume that the white
wire was the signal wire that was connected to the wiper, but it turned out
that it was the black wire, so I’m glad I
checked. And, speaking of checking...
Let’s remind ourselves that we currently
have six pots that are each supposed to be
10kΩ. If you adopt a mantra of ‘trust nothing; verify everything,’ then the radiance of
my smile will lighten your life. The main
point to note here is that it’s a lot easier to
test your pots in isolation than it is when
you’ve connected them all together.
3
2
Fig.7. Six 10kΩ pots wired in parallel.
3. 3 V
1
2
Fig.6. (a) Pot symbol; (b) backside view.
Test, test, and test again
1
10 k
3
1
10 k
1
10 k
1
3
10 k
3
(a) Sy mbol
1
2
10 k
3
1
10 k
1
Not y et
connected
0V
terminal 2s and the other probe to any
of the wipers (this is something anyone
might do), what resistance would you
expect to see?
Would you be surprised if I told you
that you should be seeing something in
the region of 2.92kΩ. Also, that if you
drive the pot whose wiper you are measuring to its full extreme one way, you
will see a reading of 0kΩ, while driving
it to its full extent the other way will
result in a reading of 1.67kΩ.
If you say, ‘Well, duh!’ then you need
read no further. Alternatively, if you
are scratching your head and your face
is scrunched up into a frown (which is
not a becoming look on you, I’m afraid to
say), then I would be happy to expound,
explicate, and elucidate further. Unfortunately, our illustrious editor says we
don’t have the time or space to do this
here, so I’ve written it up as a Cool Beans
Blog (https://bit.ly/3omhAzC).
Next time
In my next column I hope to be able
to show you the code that makes oil
wheel-type swirly effects on my ping
pong ball display. Also, we will be experimenting with more effects on our
SMAD boards. And, of course, we will
continue to work on adding motion to
my robot head, with the ultimate goal of
having it be able to turn to track me as
I move around the room or to respond
to any unexpected sounds. Until then,
as always, I welcome your comments,
questions, and suggestions.
Toroidal transformations answers
A1: There are still four sides and four edges,
where the paths of the edges can be denoted as A -> A’, B -> B’. C -> C’, and D -> D’.
A2: There are two sides and two edges,
where the paths of the edges can be denoted as A -> A’ -> C -> C’ and B -> B’ -> D -> D’.
A3: There is only one side and one edge,
where the path of the edge can be denoted
as A -> A’ -> B -> B’ -> C -> C’ -> D -> D’.
Cool bean Max Maxfield (Hawaiian shirt, on the right) is emperor
of all he surveys at CliveMaxfield.com – the go-to site for the
latest and greatest in technological geekdom.
Comments or questions? Email Max at: max<at>CliveMaxfield.com
47
