This is only a preview of the December 2021 issue of Silicon Chip. You can view 0 of the 112 pages in the full issue. For full access, purchase the issue for $10.00 or subscribe for access to the latest issues. Articles in this series:
Items relevant to "Hummingbird Audio Amplifier":
Items relevant to "SMD Trainer Board":
Articles in this series:
Items relevant to "Digital Lighting Controller Translator":
Items relevant to "Orrery (planetarium) using a Micromite LCD BackPack":
Items relevant to "Non-contact cloud-based temperature sensor with speech":
Items relevant to "USB Cable Tester – Part 2":
Purchase a printed copy of this issue for $11.50. |
MAILBAG
your feedback
Letters and emails should contain complete name, address and daytime phone number. Letters to the Editor are submitted on the condition that
Silicon Chip Publications Pty Ltd had the right to edit, reproduce in electronic form, and communicate these letters. This also applies to submissions to “Ask Silicon Chip”, “Circuit Notebook” and “Serviceman’s Log”.
Important information about the
Tele-com (OzPLAR)
In the Tele-com article on page 38
of the October issue, one of the alternative transformers for the ringer section in the parts list is shown as Triad
FS24-100-C2 (Mouser Cat 553-FS24100-C2). This should instead be Triad
FS24-100 (Mouser Cat 553-FS24-100).
Note that the Altronics M7024A is a
good performer, and about the same
price as the Triad FS24-100.
The FS24-100-C2 version (while
cheaper) is designed to be short circuit
proof, and as a result, it just doesn’t
perform well in the ringer circuit. I
purchased quite a few and was disappointed to discover this.
Another recommendation that isn’t
covered in the article is the type of
self-tapping screw used to mount the
board in the Pac-Tec LH96-200 case.
I initially used some 8mm-long 4G
self tappers and found that one of the
mounting stand-offs in the case split
down the side.
The PT (Plas-tite) screws WN1411KB30X6Z available from PSM Fasteners
in Marrickville, NSW can be used to
avoid this happening.
Editor’s note: we usually find
4G/3mm screws to be fine for this sort
of job, but perhaps 8mm is a bit long;
6mm screws are less likely to split
the posts.
Note that both feed bridge designs
were tested using a power supply similar to those in the parts list, and we
could not notice any audible switching artifacts in the telephones. However, some constructors may not use
the recommended PSU and instead
elect to use a small plugpack such as
the Altronics M8968B (superseded by
4
Silicon Chip
M8968C) because they are considerably cheaper.
Testing with the M8968B after the
article went to press showed that when
the LB1011AB feed bridge was used,
switching artefacts were audible in
the telephones as an annoying ‘digital
squeal’. It is not noticeable when the
M1000 inductor feed bridge is used
with the M8968B.
Fitting the M8968B with an 820μH
inline inductor using adhesive heatshrink and decoupling pin 8 of both
LB1011ABs with 4.7μF 50V capacitors
effectively eliminated the noise. Still,
the likelihood of constructors opting
for the LB1011AB feed bridge is minimal, although we can supply them
if needed.
The 820μH inductor tested was
Mouser Cat 815-AIAP03821K. I used
6mm adhesive heatshrink (Altronics
W0994A) to increase the diameter of
the cable on the M8968B and sheathed
the inductor with 19mm adhesive
heatshrink (shown below; Altronics
W0997A).
I haven’t checked the switching
frequency of the Altronics M8968B,
but I suspect it is around 500kHz. I
also tested a plug pack that operated
at 50kHz, and the modified inductor/
capacitor filter arrangement effectively
suppressed the noise from that as well.
Ross Herbert,
Carine, WA.
Silicon Chip magazines to give away
I have many Silicon Chip magazines that I wish to give away. If you
are interested, please e-mail silicon<at>
siliconchip.com.au and they will pass
your message on to me.
John Maarssen, Thornlie, WA.
Australia’s electronics magazine
SMD Test Tweezers – more than just
something to build
This is just a note to let you know
how much I enjoyed the October article by Tim Blythman on the SMD Test
Tweezers. I went right ahead and purchased the kit but never built it!
My interest started because I was
curious about how to use the OLED
display, how the sleep mode worked
and anything else I could learn from
it. It is great that you make the source
code available, so I immediately downloaded the C source code from your
website.
I’m by no means a ‘crack’ C programmer. Still, on opening the code
in MPLABX, it just looked very neat
and concise with its separate includes
for the I2C communications, the OLED
operation and convenient utility routines such as getDigit for displaying
on the OLED. I mapped the font.h file
onto a spreadsheet with “x”s to see
how the fonts were made.
It was a real learning experience.
For example, I discovered the term
“include guard” when exploring the
#defines. I always had a vague idea
what that was about, but there it is,
nice and simple. It inspires one to try
to write neat code.
I’ve decided to get familiar with the
PIC16F1459, so as an exercise, I set
the circuit up on a breadboard, using
different pins (because the processor
is different). I created a new MPLAB
X project with your code and began
porting it across. All this meant getting familiar with MPLAB X again,
fiddling around with the PICkit programmer and using different pins for
wake up on interrupt.
In the end, I now have a nice little
siliconchip.com.au
test jig (which behaves as a pair of
test tweezers) for practising with I2C
driven OLED displays and a whole
lot of new ideas as to how to format
the code.
The reason I’ve written is to thank
you for your efforts and to let you know
a different angle from which some of
your readers might derive enjoyment
from such projects.
Dave McIntosh,
Eastwood, NSW.
Backwards compatibility nightmares
My reading list is far and wide, and
I always read (or at least flip through)
your magazine with interest because
I never know what I will find. April’s
edition was no exception, and I wanted
to add to your editorial comments
(“Adobe making our lives difficult”),
which are not exclusive to Adobe.
I published some books on lighting
design, for which a few of the fonts
have since been ‘updated’. This means
that I need to change the whole contents because the new fonts are not
only ghastly and inappropriate, but
they completely scramble all of my
work, including paragraph endings,
tables and the like.
To overcome this problem without
re-doing the layouts, I check with the
printers to ensure that the fonts I’ve
typed are still in their system. That
is the best I can do under the circumstances, though I now have three publications requiring the same method
for printing.
But it doesn’t stop there! When I was
forced to upgrade to a new computer
and Windows 10, I was assured that
every piece of technology would work
on the new system. However, my brilliant Epson 1260 scanner, which also
does 35mm colour slides and negatives, would not work.
After wasting so much time, I contacted Epson directly only to be told
that the scanner was too old and I
needed to purchase another item
which is superior (probably because
it was also a printer – whoopty-doo).
This is what it’s all about – getting us
to upgrade all of our equipment whenever a newer model is produced.
Thank goodness the medical professional doesn’t value our lives in the
same manner!
Then a few months ago, I thought
I’d watch a quick DVD on Windows
10, only to discover that I now need
to purchase and download an app
6
Silicon Chip
that will allow my DVD player to run.
This is absolute nonsense in its highest form. The new regime has allowed
third parties to piggyback and make
more money by stealth, not only in
the computer industry but with most
bookings or purchases done online.
As a result of all this, I have had to
go back to using Windows Vista on my
laptop so that I can use my preferred
software. But I have to keep it offline
to ensure that there are no ‘updates’
that break its operation. Some of us try
desperately to uncomplicate our lives.
I look forward to your next edition
of Silicon Chip.
Karen Wardell,
Nelson, New Zealand.
Comment: perhaps the most frustrating aspect of this culture of providing
updates that break backwards compatibility is how little these companies, to
whom we pay a considerable amount
of money to use their software, seem
to value our time or effort. They are
quite happy to cause us hundreds of
hours of work and frustration, then act
as though they are doing us a favour.
More feedback and a suggestion
A quick note to let you know I really
enjoyed the October issue; it is packed
with many interesting projects and
reviews!
I was also wondering if you’d consider publishing an updated electronic
load circuit. This is a very handy
device on the bench, and the last Silicon Chip design is from 2006.
Olivier Aubertin,
Singapore.
Comment: we have a contributor working on an electronic load design, and
it sounds like it will be finished soon.
Getting competitive about vintage gear
I expect another older reader will
top this, but I can beat Greig Sheridan’s
venerable EA power supply (mentioned in his letter you published in
the October issue) by some decades.
That magazine’s predecessor, Radio,
Television & Hobbies, published an
FM tuner circuit in January 1957, presented by its avuncular editor, John
Moyle.
The one I built was in almost everyday use until the advent of stereo FM
in the mid-seventies and supplied my
LP disc cutter with high-quality ABC
orchestral concerts. It is still working
perfectly (mono only, of course) with
its original four 6AM6 valves. Today,
Australia’s electronics magazine
it sits under my bench, and I listen to
it now and again simply for the sentimental satisfaction of keeping it alive.
Brian Wallace,
Dora Creek, NSW.
Windows updates and
DMM AC calibration
In your July editorial, I noticed that
you pinged Microsoft for their dodgy
fix that didn’t fix. Shock horror. You’re
surely not suggesting that Microsoft
sells software of dubious quality. Let’s
not forget that one of the first, if not the
first, vulnerabilities exposed in Windows XP was inherited from Windows
NT, and Microsoft knew about it for
years but did nothing to fix it.
I think you need to keep in mind that
the primary aim of almost every company is to make profits, and manufacturing something or supplying goods
or services is just the process they use
to achieve that end. Providing superior products, services etc comes at
an extra cost which has to be passed
onto customers or taken out of profits.
Some companies have successfully
managed to provide quality as well as
make a profit, although many more
have delivered quality and gone bust.
So to stay up with the competition,
flashy new thingamabobs are often
the more cost-effective route to staying in business.
Henry Ford is reputed to have said
that “There is one rule for the industrialist and that is: Make the best quality
goods possible at the lowest cost possible, paying the highest wages possible.” Modern manufacturers have
pretty much ditched the last part and
are concentrating on the second at the
expense of the first. It is a cut-throat
game.
Having said all that, I must add that
my observation is that most goods I
use these days are of higher quality
and provide greater utility than they
ever have in the past 60 years, and they
are relatively cheap. Smartphones and
computer systems would be the most
glaring exceptions to that. Fortunately,
although life could be better, it could
also be much worse.
I’ll finish this diatribe with an
observation from the late, great Douglas Adams: “The idea that Bill Gates
has appeared like a knight in shining
armour to lead all customers out of
a mire of technological chaos neatly
ignores the fact that it was he … [who]
led them into it in the first place ...”
siliconchip.com.au
Can you
deliver reallife technology
solutions?
Is the answer automated smart grids that connect
industry and households to renewable energy?
Could you engineer robotic systems that improve safety
and increase efficiency in manufacturing and farms?
Work alongside USQ’s world-class
lecturers and help solve current industry
problems with a USQ engineering degree.
Study 100% online, on campus
or a mix of both.
#1 uni in Australia for graduate
starting salary in engineering.^
Experience more opportunities
and personalised support with
our small class sizes.
Join this in-demand industry and study
electrical, mechatronic, computer systems
engineering and more.
USQ Engineering
^ Good Universities Guide, 2022
CRICOS: QLD 00244B, NSW 02225M | TEQSA: PRV12081
Helping to put you in Control
Spectrally Flat Class C Pyranometer
DeltaOHM PPYRA03AC – Spectrally Flat Class C Pyranometers
according to ISO 9060:2018. Complete with
levelling device and calibration report.
SKU: OHM-001
Price: $1545.50 ea
UC100 USB Motion Controller
Directly replaces the traditional Parallel port
with modern USB. Can control up to 6-axis with
Mach3, Mach4 or UCCNC software. Up to 100kHz
step frequency operation, Fast communication
with data buffer for robust and stabile operation.
SKU: CND-001
Price: $198.00 ea
M18 Shielded Inductive Proximity Sensor NPN, NO+NC
IPSI/L-18NOC5B Shielded (flush) M18, 4-wire
NPN-style output with NO+NC contact type. It has
a sensing distance of ~5 mm. Screw on connector.
Choose 2m cable with straight or 90 degree 4 wire
M12 connector.
SKU: IBS-0280
Price: $30.25 ea
Programmable Range Pressure Sensor
Novus Automation’s NP620 Programmable
Range pressure sensor with input range of 0
to 10 Bar. 2-wire 4 to 20 mA output, 0.25%
accuracy and 1/2” BSPP process connection.
SKU: NOS-206
Price: $259.60 ea
604 Differential Pressure Switch
Huba 604 is used as a DP flow switch in
ventilation ducts for the control of filters and
fans, and in control
systems for the control of dampers. Pressure
range 0.2-3mbar option 1-10, 0.5-5, 10-50.
SKU: NOS-2600
Price: $307.95 ea
Thermostat Controller with NTC Sensor and Buzzer
Panel mount thermostat with included NTC
sensor on 2 m lead. Configurable for a huge
range of heating and cooling applications.
Fitted with Buzzer for alarm. 100 to 240 VAC
powered.
SKU: CET-0012
Price: $121.00 ea
2.0 N·m NEMA 23 Integrated Stepper Motor
iST-2320 2.0 N·m NEMA 23 stepper motor
with integrated driver. Standard pulse and
direction (or CW/CCW) input. Advanced antiresonance DSP driver.
Comments on component shortages
SKU: SMC-126
Price: $241.95 ea
For Wholesale prices
Contact Ocean Controls
Ph: (03) 9708 2390
oceancontrols.com.au
Prices are subjected to change without notice.
8
Silicon Chip
Also, I thought about R.M.’s problem of measuring AC
for calibration purposes (Ask Silicon Chip, July 2021,
page 110). Most DMMs do not have very high accuracy
on the AC ranges, usually much poorer than the specified DC accuracy.
For example, Keysight U123x meters have specified
DC accuracy of 0.5%, AC accuracy of 1% for the low
voltage ranges, and frequency accuracy of 0.1%. I used
to own one, and I thought it was a great general-purpose
instrument. The corresponding figures for the U1282A are
0.025%, 0.3% and 0.005%. I own one now but rarely use
it because it’s expensive, and I don’t want to damage it.
1% is a pretty typical accuracy for the AC ranges on
DMMs.
You recommend that R. M. use a “low-distortion sinewave oscillator”, but consider how much distortion contributes to the RMS value of a (nominal) sinewave. Because
you have to use the root-of-sum-of-squares method to add
distortion and fundamental the distortion has to exceed
14% before the RMS value increases by 1%.
A 1V RMS sinewave with 1% distortion has an RMS
value of 1.00005V RMS, which is well within the accuracy of even five- or six-digit DMMs.
I realise that “low-distortion” is a relative term, but
even a cheap function generator should deliver acceptable performance.
Besides, if you are using the LTC1966 as a reference, it
should be irrelevant what the waveform is so long as it is
within the meter’s crest factor capabilities.
The exercise is complicated by the LTC1966 working
over a fairly small range, up to 500mV max, so it cannot
be used reliably to calibrate higher voltage ranges without also calibrating the gain circuit. Also, strict attention
must be paid to the effects of loading, both resistive and
capacitive. If the multimeter is not a ‘True RMS’ meter,
the need is less.
Phil Denniss, Darlington, NSW.
Nicholas comments: You are right; a pretty basic sinewave generator should be fine for calibrating most DMMs
on AC voltage ranges. The main requirement is that sinewave amplitude must be more precisely known than the
DMM’s measurement tolerance.
Also, some of the blame for how buggy computer software has become must lie with consumers. The reason
software companies prioritise adding bells and whistles
over fixing bugs or improving performance is that they’ve
figured out that is why people pay for their products.
Consider that Microsoft and Apple essentially form a
duopoly, and by refusing to license macOS to be run on
hardware they do not sell, Apple allows Microsoft to act
as a monopoly. Monopolies rarely lead to good outcomes
for consumers. Linux is making inroads into the desktop
market, but only slowly.
Concerning your October Editorial Viewpoint, I am not
surprised that components are also in short supply. This
suggests that the situation will be quite bad for Silicon
Chip and hobbyists because we are at the bottom of the
pecking order.
The inability of hobbyists to play with electronics simply does not rate against manufacturers who need to keep
their production operational.
Australia’s electronics magazine
siliconchip.com.au
Design Contest Win $500+
Dick Smith challenges you
Win $500 by designing a
noughts-and-crosses machine
that can beat 14-year old me!
Dick Smith has described in his new autobiography how one
of the turning points in his life, at age 14, was succesfully building a ‘noughts-and-crosses
machine’ (also known as tic-tac-toe) that could play the game as well as anyone. Keep in mind that
this was in 1958, when nobody had computers; it was a purely electromechanical device.
Email Design to Enter
Design your own noughts-andcrosses circuit and send your
submission to compo<at>siliconchip.
com.au including:
a) Your name and address
b) Phone number or email
address (ideally both)
c) A circuit or wiring diagram
which clearly shows how the
device works
d) The display can be anything as
long as it’s understandable
e) Evidence that your device can
always play a perfect game (it
never loses)
f) A video and/or supply images
and text describing it
g) Entries requiring software
must include source code
The deadline for submissions is the
31st of January 2022.
➠
➠ Win $500 + Signed Copy of Dick Smith's Autobiography
➠
Four winners to be decided, one
each for the following categories:
➊ The simplest noughts-andcrosses playing machine
most ingenious noughts➋ The
and-crosses playing machine
youngest constructor to
➌ The
build a working noughts-and-
DICK
SMITH
crosses playing machine
most clever noughts-and➍ The
crosses playing machine not
using any kind of integrated
processor
The entry we judge overall to be
the best will also be featured in
our Circuit Notebook column and
receive an additional $200.
‘Businessman, adventurer, philanthropist…Di
ck Smith is a true Australian legend.’
JOE CITIZEN
Conditions of entry
Dick Smith writes
1) You must be a resident of
Australia or New Zealand
2) One entry per family (Silicon
Chip staff and their families
are not eligible)
3) Submissions will be confirmed
within 7 days. If you do not
receive a confirmation of your
submission, contact us to
verify that we have received it
4) Chance plays no part in
determining the winner
5) The judges’ decision is final
6) The winners will be decided by
the 3rd of February 2022 and
will be notified immediately
By 1958 I’d advanced from building crystal radio sets to designing
and building what I called a noughts and crosses machine. It really
was an early computer. I used second-hand parts from a telephone
exchange to build it. It would play noughts and crosses against anyone
and no one could beat it.
This was a great boost to me, because while I was no good at rote
learning and theory, I was fine at practical things. The fact that my mind
was capable of working out how to build this complex machine gave
me confidence as I left school. Now I just had to find a job.
Because this was such a turning point in his life and he’s so
enthusiastic about youngsters learning electronics, he’s putting up
$2000 of his own money to award to people who can come up with
a modern version of his noughts-and-crosses machine. Silicon Chip
will judge the entries.
Winners will be announced in the March 2022 issue of Silicon Chip
magazine and will also be contacted directly for payment information.
siliconchip.com.au
Australia’s electronics magazine
December 2021 9
POWER
SUPPLIES
PTY LTD
ELECTRONICS SPECIALISTS TO
DEFENCE AVIATION MINING
MEDICAL RAIL INDUSTRIAL
Our Core Ser vices:
Electronic
DLM Workshop
Repair
NATA
ISO17025
Calibration
37 Years
Repair
Specialisation
Power Supply
Repair to
50KVA
Convenient
Local
Support
SWITCHMODE POWER SUPPLIES Pty Ltd ABN 54 003 958 030
Unit 1 /37 Leighton Place Hornsby NSW 2077
(PO Box 606 Hornsby NSW 1630)
Tel: 02 9476 0300
Email: service<at>switchmode.com.au Website: www.switchmode.com.au
10
Silicon Chip
But hobbyists have one advantage; we can use recovered components, whereas manufacturers cannot afford
the risk.
For many years, I bought old equipment and PCBs to
get expensive and rare components, and in doing so, I
collected a lot of common parts.
The result is that the shortage is not a hindrance to my
experimentation. I realise that I am in a unique position,
but there will still be old equipment with components
that can be retrieved. One just needs a good hot air gun
to make for easy removal.
If you don’t already know, Wiltronics are still advertising surplus components. They are: BAS16 SOT-23,
BC848C SOT-23, BC849C SOT-23, all in 3K reels <at> $30 per
reel; TIP32B TO-220 as 500 pack <at> 73¢ each; 1SMB5941
47V diode 1.5W in 2K reel <at> $40; and 16V 5W Zener
diode in pack of 1000 for $50.
You can find these online at: www.wiltronics.com.au/
product-category/semiconductors-surplus/
The Dick Smith noughts & crosses competition intrigues
me. I have no intention of entering it, but it is interesting that noughts & crosses is being revisited when it has
been researched to death. The creation of an unbeatable
machine is a trivial exercise. I will be very interested to
see the winning entries.
Regarding Mr Smith himself, I am impressed that he
created an unbeatable machine at age 14, especially back
in 1958. He was very fortunate to have access to telephone
exchange parts as such things were almost impossible to
obtain in the early 1960s, when I was his age.
As I age (now 70) and suffer more and more from knotted neurons, I have become interested in technical people who have achieved something remarkable. Their stories are far more interesting than those from other areas
of endeavour.
The microprocessor used in the USB Cable Tester project of the October Silicon Chip edition is a new one to
me, and I was curious to look at its specs.
Since the COVID-19 pandemic has caused a shortage
of microcontrollers, it has occurred to me that Microchip and other manufacturers may use the pandemic to
rationalise their products.
In particular, older ICs that use more silicon real estate
than current products would undoubtedly be targets for
phasing out. Since Microchip has a vast range of microcontrollers, I would expect them to take advantage of the
situation.
The PIC16F18877 looks like a drop-in replacement for
many earlier microcontrollers, including some very popular ones that have been around for some time and are in
short supply. Microchip has been very good at maintaining common pinouts on many of their microcontrollers,
which is to their credit.
However, the move to the ‘swiss army’ microcontroller does leave a bit to be desired because that increased
complexity also leads to microcontrollers which are much
harder to understand.
George Ramsay, Holland Park, Qld.
Comments: we reckon most people who’d be interested in
entering the competition would already know how to win
the game. The real challenge is coming up with a clever
circuit that uses minimal components, especially if no
microcontroller is involved.
Australia’s electronics magazine
siliconchip.com.au
By the way, while the situation has stabilised somewhat,
we can’t see the shortages going away until late 2022 at
the earliest, and more likely 2023-2024.
As far as we are aware, Microchip has never phased
out a product, and we don’t think they will start any time
soon. However, their older products become more expensive over time, making switching to newer devices attractive. They generally perform better and have more features at a lower price, so it makes a lot of sense to migrate
code where possible.
We are sure you are right about other manufacturers
taking this opportunity to cull their ranges, though.
Yes, Microchip does a great job of maintaining pin compatibility. For example, the PIC16F1887x family looks
like a drop-in replacement for the venerable PIC16F877.
We don’t think the new micros having more features is
a problem. You don’t have to use all the features; many
of them do nothing unless enabled, but it’s nice to have
them present in case you need them.
The processor cores are still pretty easy to understand
and work with, although Microchip’s stablemate AVR
processors are significantly easier to understand at a low
level than the PIC series.
A familiar tale of woe
I read with particular interest your column in the October 2021 issue with respect to the global silicon chip shortage. Your Mosfet lead time quote is the worst one I have
heard so far (2.5 years!), but for all the other MCUs I use
at RICTECH, the lead-time is usually quoted as November 2022 at the earliest, so at least a full year of waiting
at this stage.
I sell the Colour Maximite 2 units on my website, and
I have one left and no chips to get any more made for at
least a full year, probably longer at the rate things are
going. I will continue to offer it once I can get the chips
again, but who will remember or even want one in a year?
I was also interested in your Tele-com project starting on
page 30 of the same issue. It’s a clever idea, and I remember
my older brother building something similar back when
I was just a nipper, although it was 9V battery-based and
used the handsets only, not the whole phone.
Graeme Rixon, RICTECH(NZ) Ltd.
Notes about connecting to Micromite via Bluetooth
I have been doing some more work with my Micromite
project that you published in the September 2021 issue,
and thought it would be a good idea to pass on some
additional advice.
When using the Bluetooth terminal on your Android
phone, you can often connect OK, but the screen is blank.
This is because the program auto-starts and is out of sync
with the terminal.
Powering the Micromite on and off a few times will
sometimes get things back in sync, but the best way is to
send Ctrl-C to the Micromite from the terminal.
That will halt the auto-running BASIC program and
return you a cursor input prompt. You then just type in
RUN, and things will start at the beginning for you.
The Bluetooth Terminal App recommended in the article has the facility to program a macro key. I have programmed my M1 key to send Hex 03, which is Ctrl-C.
Tom Hartley, Allens Rivulet, Tas.
SC
siliconchip.com.au
Australia’s electronics magazine
December 2021 11
|