One of the most-often criticized man in the world is the person who has
changed just about everyone's life.
It is the inventor of the software we all use everyday on our computer.
Maybe things have changed slightly in recent times but the originator
who changed our computer from a blue screen with writing to a
full-colour active screen, was BILL GATES with his MICROSOFT packages.
He has been criticized by lawyers, politicians, accountants, professors
and engineers, but not one group has banned together and produced a
Agreed, there were some rival start-ups, but they all wanted $1,000 for
a package when Microsoft was only $200.
Most inventors invent one item. Bill Gates invented one item and then
went on to improve and expand it. That takes brilliance.
He could have sat back and counted the millions from the first release,
but he went on to produce the next improvement and the next. Each stage
was a huge advancement over the previous.
But the one thing no-one realises is this:
His programs required bigger and better and faster computers (operating
systems) and he was forcing hardware manufacturers to come up with
advancements that were 10 times better and faster and larger in storage
capacity, all along the way.
My first computer had 10 meg hard drive. The next was 80 meg and then it
was hundreds of meg. Now we are taking about tera byte.
We went from 5.25 floppy disks to 3.5 and these only stored from 360kB
to 1 meg of
data. Programs came on 12 discs.
And then the internet was born.
Bill Gates finished up with so much money he now has full time
staff dealing with humanitarian activities throughout the world.
Only an ignorant person would criticize him.
Only a person who has never done any programming.
And especially at a time 30 years ago when there was no hardware to run
your programs and no-one to ask for advice or assistance.
It was like building on sand.
He only had speeds of about 1MHz to 4MHz, a capacity of 64kB and
a 10Meg hard drive.
Florida Poly has a HUGE focus on robotics with their Engineering
Masters program, and would make a perfect resource to help turn a hobby into a
And yet he released programs requiring an upgrade of 1,000% !!
That's forward thinking in a time when IBM said the world would need 8
computers. He said he wanted to put a computer in everyone's home.
That's when they thought he was mad.
And he was wrong . . . every house has more than 3 computers!!!
The capacity of early computers was so small they could not run any of Bill
And when you think of it. All his programs were worthless. You had to buy
And 30 years ago, the term obsolescence, had not been discovered. Everything
lasted until the day it died.
He changed the world.
Not only did he introduce the word "updating," but introduced the term:
And it was the introduction of "copies" and "clones" and "knockoffs" that
brought computers to a cost below $1,000.
Yes, we have an enormous amount to thank him for. Maybe the would would have
progressed to the same extent without him, but you can only say this if there
were parallel inventors with the same forward thinking.
Yes. I know he took a lot of ideas from others and was helped along the way.
I am only using his name as the "Front-Guy" and saying the whole revolution
was brought about by a very small group of thinkers and now this type of
thinking has been imparted to millions of others and that's why we have an
industry called the IT industry.
The biggest problem with University Graduates is this:
They "thin and expect " they idea, project, circuit will work the first time.
They get lulled into a false concept of the "world of reality" by being involved
in markings as high as 99.85%.
And when the circuit doesn't work, they are LOST.
Although my circuits work 100% of the time, they all go through a series of
steps and stages of improvement and adjustment until this result is achieved.
Even a microcontroller program will take 100 or more steps to get it finalised
and sometimes a PC board will take 3 or 4 changes and improvements to get to the
Most of the time these changes are improvements and even total re-design with
fewer, cheaper components.
But, in the end, the result is a robust, reliable project that anyone can build
and get it to work as soon as it is turned ON.
As I have said for the past 50 years, it does not matter if the project
does not work the first time. This is when and where you will stat to learn
It's only after diagnosing, testing and fixing a fault, that you will really
Anyone can build electronics projects and, in fact, out first 6 transistor
superhetrodyne radios were build by 14 year-old children on the floor of their
apartments in Hong Kong and sold in Australia for $2.50. But if the radio did
not work, it would take a capable person to locate the problem.
The same applies today.
Products are made by the millions by either robots and unskilled workers, but
the eventual success lies in the technical department to sort out any faults and
get it to a stage of being produced with 100% reliability.
So, there is still a need for the "thinker."
And that's what we are providing.
The tools and knowledge to get you into the world of designing and producing
products that will assist, help and advance the world, in general.
And the best place to look is the area of aids for the sick and disabled.
The field is enormous and it just takes a little time to associate yourself with
those in need to see what devices can be invented.
I will leave the rest to you.
Universities are constantly emailing me in the hope I would recommend one or
more of their courses.
I would not recommend ANY University course.
I have experienced FIRST HAND the course structure, notes and access to
practical learning in a number of electronics courses.
Before you even start to think about doing any type of electronics
course, spend $1,000 on kits from Talking Electronics and a few robot
kits and learn the basics and advance yourself to a stage that you can
understand what is going to be taught in the course you are
These courses cost $30,000 or more and offer little more than glossing
over the basics and using some exotic equations to make the course sound
But, in the end, the graduates have hardly soldered a wire and could not
build or diagnose any of the Talking Electronics beyond a beginner.
A $30,000 course is $29,000 wasted and you would not even get $1,000
worth of projects to take home.
The last University professor stripped the valuable parts out of the
projects and left the rubbish for the students to take home.
If you want to test the skills of a graduate, take a faulty project to a
University and ask the class to repair it for you.
That's when the truth will hit you in the face. No-one will volunteer.
The funniest photo you will get from a University
Electronics For You
Another junk circuit from
None of T.K. Hareendran's circuits work. He has
been dumping his untried, untested, junk circuits into Electronic For
You magazine for years and never bothered to reply to my emails.
The 100k resistor will only allow 3.2mA to flow at the top of the wave
and this will only occur every half cycle. This reduces the average to
less than 1mA !!!
The blue LED drops 2.7v and so the whole circuit only has 2.7v to play
The drop across the collector-emitter of the transistor is 0.3v so how
do you expect the blue LED in the rainbow LED to illuminate if only 2.4v
But if you look at the whole project, why turn the LED on with a
transistor etc when the circuit takes just as much current when not
The whole project is a FLOP.
What is the purpose of the 150 ohm resistor? Who cares about
the diameter of the LDR?
Apart from that, the circuit is live and DANGEROUS.
Never make any circuits that connect directly to the 240v.
They are ILLEGAL in Australia and India should wake up to this
He posted a video of the circuit and he thinks: "Just because the
circuit works, it must be designed correctly."
The voltage across a LED varies considerably and on some occasions, the
circuit will work very faintly.
But I am criticising the lack of understanding and the lack of technical
know-how and the lack of "tolerance" in the components.
I remember an Indian flip flop circuit from 40 years ago in an
Australian magazine. I built it and it did not work.
I found out later that the variation in gain of the transistors needed
the resistors changed to get the circuit to start working.
And the same with a micro flasher from England. I never got the circuit
to work and again it relied on adjusting the components to get the
circuit to start oscillating.
It's a fluke if you can get them to work.
He posted a video of his circuit working.
The voltage across a LED varies considerably and just because a
particular set of components work in the laboratory, does not mean the
circuit is suitable for publishing.
You need to get your values CORRECT and UNIVERSAL - so that
anyone with any set of components will be able to get the circuit
When I design a kit or any project, I build a number of prototypes and prove the fact
that it will work perfectly because none of the components are
It just separates the DESIGN ENGINEER from the dodgy ENGINEER.
Here is an OR-gate from an electronics forum:
Here is a reply from a technical engineer:
Here is my reply:
The whole point of this website is
to teach you how to simplify things and think like an engineer.
Whenever your design a circuit, go through each component and ask: "Is this
Sometimes, the saving of one or more components can be the difference between
profit and loss, when you are selling thousands of items.
But it also shows you are good engineer. That's why you have to thimk !!
I received this PC board today:
It is not mine and was received by mistake from the PC board
But it is a good lesson for everyone.
The board has no markings or component values.
This is a big mistake.
You might think you will remember, but when you have hundreds of different PC
boards and lots of different versions and improvements, you cannot possibly
All my boards can be assembled without any further information. All the
component values are shown and the name of the project with
Some of my customers buy a kit and don't assemble it for 20 years. All these
marking are not just for my benefit but for EVERYBODY.
A fully documented board shows professionalism as well as convenience.
We have passed the stage of paying extra for double-sided boards with a legend on
top and bottom as well as plate-through holes and tinned lands. All this
beauty is included for FREE.
There is no excuse for not producing a perfect board and something that someone
can take way and assemble without frustration.
One of the websites provided an article on soldering.
It suggested one of the cheap soldering irons that gets "as hot as hell."
Here are my comments:
The soldering iron kit you offer is absolutely WORTHLESS. For
the same price you can get a temperature-controlled iron on eBay.
0.2mm solder is a bit fine.
0.5mm solder is the best and here is the way to set the temperature of the iron:
Turn the temperature knob down to very low and turn the iron ON.
Gradually increase the temp and wait for the tip to heat up.
When the solder begins to melt, you have reached the temperature called the
Now increase the temperature slightly so you will actually solder at a higher
temperature and this will allow the solder to "run" all over the joint and not
The most important part of the solder is the flux.
It is important not to "burn the flux" but allow it to get hot and it becomes an
ETCHANT and "digs away" the oxidised layer on the solder-land - just just soap
on our hands.
The solder should be on one side of the wire you are soldering and the iron on
the other and you should just touch the solder to achieve THERMAL CONDUCTIVITY
to melt the solder and allow it to run all over the wire and the land.
This should always be one in less than ONE SECOND because sometime you are
soldering LEDS and they will be damaged if it takes 2 seconds.
The website did not show any valid images on how to solder and no-where did it
say to tin only one land for a surface-mount component and then place the
component and "stick it in place" before fully soldering the other end and maybe
coming back to the first end.
You don't mix the markings on a component with component-values on a circuit.
Sometimes I put additional component-values and component-markings on a PC
board to prevent wrong values being fitted. Such as 220 for a 22 ohm surface
mount resistor and 4702 for a 47k surface mount resistor.
The 223 and 102 on the circuit above should be 22n and 1n.
You don't need 470u across the battery because the current will be less than
10mA and 10u will be sufficient.
1uH for the coil is too large. 0.1uH (100nH) is approx correct.
With this simple mistake, the circuit will never work.
A FIRST LAB IN CIRCUITS AND ELECTRONICS
This textbook is filled with mistakes and terrible descriptions as well as
teaching quite complex things to beginners, without covering the really-needed
Here is a positive/negative output supply with the "batteries" identified by a
single cell instead of a voltage. The output is marked "+" and "-" but it will
be reversed when the pot is at the lowest position.
The text says: "You have made an AC to DC converter."
The input shows DC !!!
If the input is AC, the graph shows the output voltage will follow the input but
get clipped at a voltage equal to VDC plus the 0.6v across the diode.
As soon as you supply sufficient voltage to the gate to turn the MOSFET on,
current will start to flow and if the voltage is increased further, the MOSFET
will short-circuit the power supply and either the MOSFET will blow up or the
power supply will "cook."
There is no current-limiting LOAD resistor.
The 100u is around the wrong way. The Function generator will produce a voltage
higher than the base-emitter voltage.
What is a LOGIC LEVEL INDICATOR?
A Logic Level Indicator is generally a LED that shows when an output is HIGH.
The term should be LOGIC PROBE because the output will be oscillating at 100Hz
to 10kHz and a Logic Probe will indicate the percentage HIGH and percentage LOW,
by the illumination of the LEDs and the pulse LED will show a signal is present.
He also says: We see capacitors with 1000 on them to indicate 100 with no
zero's to mean 100p.
SHOW ME WHERE ???
In his kit of parts he uses 1/2watt 10% resistors. Who uses 1/2watt 10%
The text book is an absolute DISASTER.
He has obviously never used a CRO in his life, otherwise he would have included
This is the most important table for a CRO and is placed on the side of the CRO
so you can quickly see what frequency is being displayed.
That's why so many visitors to Talking Electronics website said they have learnt
more from the website than all the teaching they got at University.
Here's another stupid design:
The gain of the transistor is only 30 at 100mA and the current capability of the
transistor is about 100mA, so it is not suitable for the circuit. You
can't use a "100mA transistor to do a 100mA job."
The power delivered to the LED will be about 300mW at the most. You can't get
any more than this from a 9v battery. Do you think a 9v battery is the
"power-plant of the future?"
That's only the first problem.
You will find the reverse voltage is more than the LED will tolerate.
Here is the specification:
Maximum reverse voltage between ANODE and CATHODE: 5v.
You have a 25 turn to 50 turn transformation ratio so the reverse voltage can be
18v or more.
Reverse spikes can damage a LED much faster than the blinking of any eye.
Just because a circuit works and illuminates a LED does not mean it is a good
The whole circuit is designed incorrectly and the turns-ratio should be 25:15 -
apart from all the other mistakes.
It is very complex to provide a battery charger and here the
problems with the circuit above.
The 3R9 1watt resistor will only allow a maximum of 500mA before the resistor
will burn out.
The voltage across the collector-emitter terminals is given as 4v in the data
If you are detecting 13.8v you are not providing "head-room" to charge the
The voltage across the 1k5 will be 13.8v + 0.7v + 1.7v = 16.2v If
you have 18v supply, the voltage will be 1.8v The current through
the 1k5 will be 1.2mA The maximum gain of the transistor is
100 so 120mA will flow.
If we allow 2v across the transistor the voltage drops will be 13.8v + 0.7v + 1v
across the 3R9 plus 2v across the transistor = 17.5v If
the supply is 18v you have no "headroom" so that a CURRENT can be delivered.
It seems that only 120mA will be delivered and all you need is a resistor
permanently connected to the battery and it will be trickle charged all the
This circuit does almost NOTHING.
Here's an over-design from
It is a temperature controller for a USB soldering Iron:
The fact is a USB socket can only deliver 5v at 1 amp. This 5 watts. 5 watts is
only good for the smallest joint and will be quite unsuitable for any component
that needs to be removed quickly or has large thick leads.
If you have one of these worthless irons and want to reduce the temperature, all
you need is 3 x 1 ohm wire wound resistors and an alligator clip. Put them in
series with one lead and connect the alligator clip to the other lead and add 1,
2 or 3 ohms to the resistance of the iron to reduce the temperature. You
can also get 3R3 if you want the iron really cold to do plastic work.
For $14.00 you can get a 240v temperature controlled 20 watt iron and avoid any
hassles from ebay.
For $25.00 you can get a temperature controller soldering with a base station
that indicates the temperature on 7-segment displays. Everything else is either
over-priced or just RUBBISH.
I have been using Australian-made solder for the past 60 years and never had a
I though 60/40 solder made anywhere in the world would be absolutely identical
to the type I have been using.
But I recently bough Chinese 60/40 resin cored solder and it was absolutely
useless. I don't know if the combination of tin and Lead was exactly
correct or if some other "filler" has been included or if the "reject" metals
have been added.
The solder did not melt at the normal temperature I have been using for 60 years
and the solder did not "run" over and around the joint and the surface of the
joint was was shiny.
No-matter what I did, the result was terrible.
I then bought eutectic solder that has the combination in the ratio of 37% to
63%. This ratio produces a solder with the lowest melting point. This
solder worked much better but was still not as good as the Australian-made
You must ONLY BUY eutectic solder and make sure the seller guarantees it to be
eutectic as the label on the reel does not indicate this.
Here is a TEMPERATURE ALARM from a suppler on eBay. It is not expensive
- but it does not work.
You can imagine the frustration of the hundreds of buyers having the same
The supplier was not interested in the mistakes.
Here is the circuit.
Here are the problems.
The first thing you have to know is the type of thermistor. It has a cold value
of 1,000 ohms resistance.
There are two types of thermistors. One is PTC (Positive Temperature
Coefficient) and its resistance INCREASES when it get hot. The other typ is NTC
( Negative Temperature Coefficient) and its resistance DECREASES when it gets
hot. The type used in the circuit NTC. When its gets hot, the
resistance decreases to 600 ohms and less.
Now we can see how the circuit works.
The 10k 10-turn pot is adjusted so that the first transistor is turned ON.
The first transistor turns ON the second transistor and the second transistor
turns OFF the third transistor. This means the buzzer does not produce a sound.
When the thermistor heats up, its resistance decreases and the voltage between
emitter and base of the first transistor decreases and this turns the first
But the second transistor is still being kept ON via the 39k resistor and thus
the buzzer does not produce a sound.
When the 39k is removed, the circuit works perfectly.
But the buzzer only gets a voltage of about 1.8v because this is the maximum
voltage that will develop across a red LED.
To make the buzzer louder and prevent the red LED being damaged, a 100R should
be included in one of the leads of the LED.
Finally, the circuit did not work for me because the 10-turn pot was "102"
(1k) and the resistance of the thermistor did not go low enough to turn OFF the
The CAD package use to produce the printed circuit board created a GROUND PLANE
between all the tracks and a beginner could easily leave too much solder on a
pad and the solder will touch the ground plane and create a "short."
Lastly, the first and second transistors are connected from the positive and
negative rails without a current-limiting resistor.
When they are both turned ON, a very high current will flow through them. This
is especially the case when the supply is increased to 6v. The transistors get
Now you can see why you have to be very experienced in designing a circuit. If
now, you can introduce 5 faults into a simple circuit such as this, and produce
frustration for everyone who buys the kit.
I have told D.Mohankumar to stop putting his JUNK CIRCUITS on the web and
messing up young enthusiasts with poor designs and circuits that do not work.
I have covered this circuit before but he continues to put it on the web.
A 1 watt LED takes 300mA at full brightness and this means 300mA will flow
through the 10 ohm resistor and drop 3v.
The battery only supplies 4.5v and the drop across the transistor will be about
0.3v. The LED has a characteristic voltage across it of 3.6v when fully
Work it out for yourself. The voltage across the resistor will be 4.5 - 3.6 - .3
= 0.7v When 70mA flows, the voltage across the resistor will be 0.7v.
When 70mA flows through the LED, the voltage across it will be about 3.2v and
thus the current will increase to about 100mA. This means the LED will dissipate
about 300mW - a lot less than 1watt.
As soon as the voltage of the battery reduces, the brightness will reduce too.
It's just a poorly designed circuit.
The charging current will be quite small - let's see: The 4.5v transformer
will produce 4.5v x 1.4 = 6.3v.
There is a drop of 2 x 0.7v across two diodes in the rectifier = 4.9v
The voltage of the battery will rise to 5v when it is charging. This means we do
not have enough voltage to charge the battery.
The transformer needs to deliver 5v plus 0.7v plus 2 x 0.7v = 7.1v
Another project that has not been tried and tested.
Professor D.Mohankumar is just making a fool of himself, continuing to publish
projects that simply will not work.
I am highlighting these faults and showing how to work out the current that will
flow in the charging process and the illumination of the LED.
These values are much less than expected by
Professor D.Mohankumar and that's why it is important to do a little bit of
calculating before designing a circuit, so you know if it will be operational.
In this case the circuit is a DUD.
Not only do we get a lot of bad engineering from Indian websites, but many of
the world-wide electronics magazines also have terrible faults.
Here is a typical example:
A 3-amp diode will drop about 0.7 to 0.9 to 1.1v across it when full current
For the diodes in this bridge, it amounts to about 1.5 watts per diode
because only two diodes are conducting at any one time.
However the total for the bridge can amount to 6 watts and the printed circuit
board simply will not dissipate this amount of energy when the diodes are
I have had many TV's where 1 amp diodes on a PC board have eventually blackened
the board and melted the solder and created faulty connections.
If you want long-term reliability, the board must have sufficient trackwork to
keep the diodes to a point where you can hold your finger on the leads for at
least 10 seconds.
I am absolutely horrified that a magazine would allow a contributor to send
in a construction project, showing the final design with no overlay.
I would probably go further to say the board had been created with a "dalo" pen
and etched with Ferric Chloride.
This means the reader has no opportunity to get the board made
and really the whole project is a worthless endeavour.
There are just so many things wrong with presenting this sort of rubbish, that
is makes me wonder why the publishers produce an electronics magazine in the
When they do have a printed circuit board available, the cost is so high that
you could but the whole project read-made for less cost.
The subscriptions to these magazines is so small, they do not publish the
Here's a circuit from Electronics Monthly February 2019 - an Indian
magazine with no technical staff and no understanding of circuit design:
The battery shows a single cell whereas it has to be a voltage
greater than 7v for the 7805 to work.
The 10u and 1u do not show the polarity of the electrolytic.
The "globe" above the 1u has no connections and no-one really knows what it is
suppose to be.
R1 is 100k. (according to the parts list)
This value is FAR TOO HIGH.
To turn off the first transistor, the base voltage must be less than 0.5v
In the dark, the resistance of the LDR is very high. Normally it is about 300k.
So, let's work it out.
Suppose the LDR goes to 100k. The voltage at the mid-point of the LDR and 100k
resistor will be 2.5v. Too high.
Suppose the LDR goes to 400k, the voltage at the mid-point will be 1v. Too high.
It has to go to 800k !!!
Suppose R1 is 100 ohm. The LDR will have to go to about 1k to turn ON the
transistor. This is possible with a bright - focused - light.
When the first transistor is turned ON, the wasted current through it will be
50mA and the transistor will get warm. This is a bad design.
The speaker symbol is not really a speaker but an active buzzer and the diode is
When the first transistor is OFF, the base current through the second transistor
is 15mA. It only needs 3mA to 5mA to operate the buzzer.
Just another badly designed circuit that should not be presented in an
Here's another DIYODE MAGAZINE disaster.
This is a PROBE intended for both TTL and CMOS circuits.
Look at the resistor connected between the probe and 0v rail.
Suppose you are testing the input pin of a microcontroller that has internal 47k
The line will detect a high when 2.2v or higher is presented to the pin.
When the probe is applied, the 47k pull-up will be acting against the 10k
(forming a voltage divider) and the voltage will be right down at the 1v to 2v
level. You have to take into account the 18k pull-up resistor and this will
raise the voltage above 1v.
The PROBE is very badly designed as the input impedance of a probe should be 1M.
This effectively puts NO LOAD on a circuit AT ALL - especially when the lines
are running at high frequency. .
But if you are dealing with a fairly low frequency, the biasing resistors can be
as low as 100k. But when they are 10k to 18k, the probe is quite worthless and
may produce a false reading.
You will be totally unaware that the probe is at fault and be looking all over
the circuit for the problem. A VERY DANGEROUS PROBE.
Here's another DIYODE MAGAZINE disaster.
The resistor symbols are too big and it makes the circuit look childish. The
breadboard layout had no parts identified and these have been added by me to make
the layout useful. The block diagram of the 555 might be ok to show the
placement of the wires, compared to the wires on the breadboard but the whole
purpose of a CIRCUIT DIAGRAM is to show the output lines of a chip in standard
locations on the circuit so you can instantly see what the circuit is
This is the whole purpose of an EDUCATIONAL MAGAZINE and this is where DIYODE
MAGAZINE falls down completely.