Won’t Somebody, Please, Think Of The Transistors!

At what age did you begin learning about electronics? What was the state of the art available to you at the time and what kinds of things were you building? For each reader these answers can be wildly different. Our technology advances so quickly that each successive generation has a profoundly different learning experience. This makes it really hard to figure out what basic knowledge today will be most useful tomorrow.

Go on, guess the diode!
Go on, guess the diode!

Do you know the forward voltage drop of a diode? Of course you do. Somewhere just below 0.7 volts, give or take a few millivolts, of course given that it is a silicon diode. If you send current through a 1N4148, you can be pretty certain that the cathode voltage will be that figure below the anode, every time. You probably also have a working knowledge that a germanium diode or a Schottky diode will have a lower forward voltage, and you’ll know in turn that a bipolar transistor will begin to turn on when the voltage between its base and emitter achieves that value. If you know Ohm’s Law, you can now set up a biasing network and without too many problems construct a transistor amplifier.

You probably consider this stuff to be the innate knowledge that everyone should have if they are involved in electronics, but how did you gain that knowledge? Did you read it in books? Perhaps you were taught it at school, or at university, or maybe you figured it out yourself in a misspent youth surrounded by parts scavenged from surplus electronics. Either way, given a handful of discretes, you’ll be able to make them do something. It’s not as though you create complex transistor circuits in the way your 1960s equivalent would have done, but it’s still pretty fundamental to working with electronics even in 2018.

I’ve had a couple of conversations in the past few months with friends who are following university teaching careers in the electronic engineering departments of different universities on different continents, and who both expressed a similar refrain. Their students arrive with an extensive familiarity with electronics, but a lot of the basic innate knowledge described in the previous paragraph is missing. It’s not that they are somehow not up to scratch, they’re as clever as they have ever been. But for them the experience of elementary electronics has been one of breakout modules rather than discrete components, and of single board computers with every conceivable interface ready for them to use. They are the exceptionally lucky beneficiaries of advancing technology alongside globalised manufacture. This brings us the best the world can offer for the price of a fast-food meal or a movie ticket, but this comes at the cost of abstracting the action further away from the engineer.

There is a whole Hackaday article about this transistor.
There is a whole Hackaday article about this transistor.

In the past, an aspiring electronics enthusiast would have started with simple transistor circuits such as a light sensitive switch, a simple electric organ, or an AM radio such as the one featured in the children’s book we reviewed last year. They used the minimum of components yet delivered a tangible result, and from them a basic understanding of topics such as biasing could be gained. And once they had tired of the book projects they had a ready supply of almost limitless free components in every dumpster since consumer electronics of the time contained plenty of through-hole discretes.

You might say that kids today live in a world of game consoles and 24/7 media so therefore wouldn’t be interested in a simple circuit. While it’s true that an AM radio isn’t going to capture their hearts, it is unfair to them to make unfounded judgement about their capacity for seeking knowledge. Our technology-focused youth are smarter than some adults give them credit for.

A very cursory search will find projects using a Raspberry Pi as an organ or as a light-sensitive switch. Is learning about discrete components any less valid than learning a bit of Python? Ideally, they should be learning both. But let’s face it, spending time learning Linux and myriad programming languages is an important part of the upbringing for the electronics hackers of today.

So what is to be done, to bring back a bit of basic electronic knowledge for the next generation? Certainly not throwing away their Arduino or Raspberry Pi, after all the tech-inclined youth of the 8-bit era or before would have given anything to own one of those! We’re also not interested in going misty-eyed over circuits and components from the 1960s, it’s perfectly acceptable to be into retro electronics, but the kids are going to be learning their craft in the 2020s and need stuff that’s relevant. There is no reason why a single-board computer can’t interface to a discrete circuit rather than an op-amp or a logic gate, indeed for some applications such as adding an audio output to a Raspberry Pi Zero it is something of a necessity.

Perhaps if you take a look at catalogues containing kits aimed at a beginner audience you might begin to understand where a start could be made on broadening their scope. I can see that it is easy for a supplier to bundle a pile of sensor breakout boards with a few sets of rainbow jumper leads to go with a breadboard and a single-board computer, but the addition of a bag of passive components and some 2N3904s on a tape would hardly add much to their bill of materials. But maybe it’s also for us to step up to the plate and provide an example or two. When so many of our projects on sites such as our own Hackaday.io tend to feature a single board computer, can we really blame the youth for following our lead?

146 thoughts on “Won’t Somebody, Please, Think Of The Transistors!

    1. Those were awesome. I bought my son a smaller one when he was little. He will graduate in May with his electrical engineering degree and start working at a steel mill — in Texas!

    2. That is one of the saddest thing I have ever lost to a move.
      I still have the box of components and the breadboards that I added to the kit to expand outside the few parts it had, but the kit and book were never found after my move to the city.

      I need to see if I can get a copy of the book to toy with again.

    3. That was my first one too. My parents made me throw it out after it got too junky and they got me a 300 in 1 to replace it. That was the one with the breadboard and some of the springed parts around it. It was never the same. I might try to find a 200 in 1 on ebay one day out of nostalgia.

      1. I had the 200 in one I think, still have it although I think it’s buried in a closet at my mom and dad’s. I remember once about 8-9 years ago when I worked for Subaru, part of their electrical class was hooking up a few lights and a switch in one of the 30-in-one kits. I had all three circuits built and done inside of 5 minutes and was in the middle of building an amplifier to connect to my phone when one of the other students started looking at what I was doing. About 20 minutes later I had it completed, and the instructor was like “Hey, do you want to teach the rest of the class?” So I did while he went outside and smoked :-P

    4. I’ve been hooked since I made a red LED blink with a 30-in-1 kit when I was 8. This was the mid 1980’s. Only in the last year have I been able to delve into electronics further, finally stretching my ham radio legs after getting my General. Experiments in HF have me doing Arduino stuff now, and also building small circuits from scratch. I am one of those mentioned in the article though: I know to use certain components to make a emitter follower, but still don’t know why that works lol. I have a lot to learn. It’s easy to make huge progress and still be missing some of the underpinnings. I’m a good example of that.

    5. I started with the 50 in one. When I mastered all 50 projects, I saved up and bought myself the 200 in one. But it never really felt the same. While it had new projects, the choices were too wide for my chilld brain to put all together. the 50-in-1 allowed a ” comparatively finite” set of permutations and experimentation was easier. IMHO. At my low level of understanding the 200-in-1 was just too much kit.

    6. I wasn’t too impressed by the Radio Shack 100 or 200 in 1 kits, because they weren’t expandable. What I had was an Allied Radio Knight Kit 100 in 1, which used springs that plugged into a piece of pegboard, and you used the springs to both hold the components and wires. The components just came in a box, and then there were permanent screwed-down springs that connected to things on the front panel, like a potentiometer, milliammeter, variable capacitor, DPDT relay, photoresistor, morse key, and speaker. Oh, and in addition to the three PNP and two NPN transistors supplied, there was a 12AU7 dual triode in a socket. This was good for things that the (germanium) transistors weren’t particularly good at, like circuits requiring high input impedance. The power supply consisted of a 24V transformer (with a 12V CT filament winding), a bridge rectifier, and filter capacitor. This provided about 30 V, which was good for tube, transistor, and hybrid circuits.

      The best part was that it had a wide range of resistors and capacitors, and you could easily add other parts not included in the kit, making it a true breadboarding system. And you weren’t even limited to the size of the pegboard, since the springs would plug into standard pegboard as well. Think of it as a solderless manhattan breadboarding kit.

      This is it, on the right, unfortunately they don’t show the back side.
      http://www.mequonsteve.com/knight-kit/1966-84-85.jpg

    1. +1!!

      Then there could be a contest to build something useful in 3 IC’s or less, and it can’t be something that the chips were originally designed to do. Something on the order of building an amplifier from logic IC’s.

      1. Or people who use ADCs.

        Changing the current through a diode 20% changes the diode voltage about 5mV.. 1LSB of a 10-bit ADC with a 5V reference. If you have a 12-bit ADC with a 3.3V reference, 1LSB is equivalent to about a 3% of change in diode current.

        Diode voltage also changes about 2.2mV per degree C.. 2.7 LSB for a 12-bit 3.3V ADC, or 1.8 LSB for a 5V 12-bit ADC.

        If you don’t know about those effects and don’t control for them, putting a diode in circuit measured by an ADC can drop the useful resolution to 8 or 9 bits no matter what it says on the tin.

        Calculating diode voltage isn’t much harder than using Ohm’s Law.. dV=ln(dI)/39.. and you can do good approximations in your head with a few rules of thumb:

        – changing the voltage 100mV changes the current ~50x
        – 60mV -> ~10x, 18mV -> ~2x, 2.5mV -> ~10%
        – changing the temperature 1C is about the same as changing the voltage 2.5mV
        – 0.645V @ 1mA @ 25C is a good starting point

    1. It depends, really.

      Let’s define “turned on” as 10mA, reverse saturation current as 10pA, temperature as good old 300K rule of thumb, and the emissivity factor as 1.2 (1.0 just isn’t realistic for real diodes), and the Shockley equation therefore gives us 0.65 volts across the diode.

  1. “At what age did you begin learning about electronics? What was the state of the art available to you at the time and what kinds of things were you building?”

    Key on a kite string. :-D

  2. You don’t have to go back to diodes and transistors to find the erosion of basic technical knowledge. Love to see how many arduino/pi folks (I have nothing against them) can craft machine code and get it into and running on a PIC, small ARM, what have you.

    1. When I was introduced to the PIC processors in the mid-90s, and had an Atmel distributor trying to convert me, the biggest issue was that every microcontroller had its own instruction set, and nobody had an affordable C compiler available for any of them. Now it almost doesn’t matter what CPU you’re using – PIC, ATmega, ARM-M0, ESP8266, they can all use the same development tools, or at least something close enough not to matter. I consider this a good thing – the learning curve for each instruction set architecture just isn’t worth it if you’re just using it for one project.

      I spent a lot of time in Z-80 machine code (not even assembler – just octal opcodes!) when I had a hand-wired Z-80 board in 1979, and in 6809 assembler (the EDTASM+ cartridge) when I had a TRS-80 CoCo. But once I upgraded to x86, C became my “bare metal” programming environment, and I’ve only gone back on the rarest of occasions. When each keyword translates to just a few machine instructions, it’s hard to justify going any lower-level than C. Yes, it’s useful to know what those machine instructions do, but actually using them is not effective use of my time.

    1. I don’t find it among car engine wrenchers and hot-rodders or machinists. They know really basic engine stuff up to modifying the code in engine controllers and stoichiometry.

  3. “You probably consider this stuff to be the innate knowledge that everyone should have if they are involved in electronics” — Well, no. I just started learning about this stuff like a year and a half ago or something and it’s been pretty slow going, too.

    1. I can second that. There is a lot of information on the web, as well as a lot of circuits. But there is not much that ties it together. That means it takes a lot of time and effort to learn it, by yourself at least. Time i don’t have.

      I would love if there more ‘labs’ out there. Something like a basic circuit that you expand in several distinct steps, where the added value is explained and shown. All I can find are expensive and timeconsuming university kits.

      1. I’m quite frustrated by the fact that there seems to be this huge gap between Arduino etc. – level easy, lower-end stuff and then more in-depth, more complex stuff. The easy, low-end stuff just explains how to hook things up together and stuff, but never bothers to go beyond that, and the higher-end stuff just expects you to already know all the math and everything involved.

        I just…I don’t even know what I should learn in order to progress from the easy stuff! A university or something isn’t an option for me, nor is some over-the-Internet course that costs several thousands, so I’m just kind of left hunting random videos and stuff with no proper guidance from start to finish.

        Some sort of a properly constructed course that goes into the maths, design pitfalls, understanding how thick one’s traces need to be on a PCB, and so on, that was also friendly for those of us with light, kinda starved wallets would be very welcome.

        1. The mid-level stuff is out there. My suggestion to you is that amateur radio may be where the broadest range of complexities exists. You can pick anything from radio transmitters and receivers made with a dozen components, to projects implementing software-defined radio in FPGAs, and literally everything in between. Radio communications may not be everybody’s interest, but as a way of getting familiar and then proficient with electronics, it can’t be beat.

          1. A Google search for “one transistor radio” is a starting point; look for web pages by radio amateurs who have many projects – there are quite a few of them out there.

        2. I learnt most about electronics technology (of the time) as a childhood hobby long before a professional education so it’s doable without education but of course the education helps.

          Learning electronics is like building a house, First you have to learn how to formset and lay concrete, then bricklaying, carpentry, plumbing, electrical etc.

          You take on one thing at a time. Google is your friend. YouTube videos are a poor source for information. Component specification documents were my main source along with many magazines of the era.

          Wikipedia has all the basic Math and most of the complex Math.

          If you have started with Arduino then you can easily move into C as the Arduino IDE is basically a C platform. There are many example libraries.

          There are a lot of design hints for PCB design as well. The starting point is choosing you layout tool. Many of them are very advanced and have a steep learning curve but there are still quite a number of simple CADs that are easy to use. I personally prefer simpler PCB CADs for development as I make prototype PCBs myself.

          As for track thickness, there are charts available on the web showing current vs track width for 1/2 oz copper. Be conservative.

          If you say more precisely “what” you want to learn then perhaps I could offer more info.

          1. “If you say more precisely “what” you want to learn then perhaps I could offer more info.” — That’s one of the problems: I don’t know what I don’t know, so I don’t really know what to learn! ;) That’s kind of the hard part when you don’t have any sort of a structured course to follow.

            At the moment I just randomly swing in every which way, just trying to increase the breadth of all things electronic in general. For example, I backed this “FPGAs for newbs” – thing ( https://www.kickstarter.com/projects/1013562009/fipsy-the-fpga-breakout-board-for-beginners/description ) on Kickstarter, because it seems fun enough and I’ve already bought the ‘Learning FPGAs’ – book on O’Reilly ( http://shop.oreilly.com/product/0636920053576.do ) for later. I’ve also bought this one course on Udemy, where you build a quadcopter from scratch. I’m exploring Lora atm., and I’ve been wondering if I should try and understand SDR-stuff. Then there’s the thing where I’d like to build my own SBC capable of running Linux — just using some low-end SoC that’s already supported by OpenWRT/Lede would do — but I don’t quite have the required knowledge like e.g. should I try to match trace-lengths and if so, where and why, if I should care about stuff like EMI and, again, where and why, and so on.

            Eh, anyway, I’m mostly just complaining because I like structured courses — I’m so harebrained that all this unstructured learning makes things slower and harder than it could be — and not because there’s a lack of stuff to learn, if you get what I mean.

  4. I ended up learning electronics from PCs down, and mostly not through academics, so I’ve definitely been through the phase where I could make an amplifying bandpass filter in an Arduino (sample, math, PWM output, RC filter), but would draw a total blank if asked to accomplish it with transistors.

    I came to analog circuitry relatively recently, and have really enjoyed seeing the fog lift from the workings of circuits that I’ve generally taken for granted. Building a photodiode transimpedance amplifier that’s orders of magnitude more sensitive than I could get from going straight to digital was deeply satisfying — to quote Asimov, “And it was amazing the feeling of power that gave him.”

    1. >” I could make an amplifying bandpass filter in an Arduino (sample, math, PWM output, RC filter)”

      I very much doubt you could actually do it without understanding of the fundamental circuitry, because the filter part would require an active filter of a higher order than a simple RC lowpass to not kill your frequency response or let an unacceptable amount of ripple through.

      I mean, you could -do- it, but the results would be practically unusable. For example, 16 MHz arduino running the 8 bit PWM output has the Nyquist limit at 32 kHz, and if you want your ripple to be less than 1%, so your effective output error would be around 2 bits, you’d have to filter it down to 300 Hz with a simple RC lowpass filter. That means your device’s effective bandwidth would be 0 – 300 Hz and there’s not a whole lot you can do with it.

      That’s the difference between top down and bottom up learning. The Arduino crowd gets the broad strokes, but the details are lost, and so the results too are very coarse.

  5. It is perfectly okay to built your knowledge upon the abstraction layer the generation before yourself built. Without that we would still code assembly for each and every processor seperately.

    But extend that concept to a perverted level and you end up with things like javascript frameworks that mimic operating system functions so you can build applications upon that like a browser that can run another javascript framework.
    And as kids these days(tm) are only proficient in using web technology languages, it apparently became perfectly acceptable to ship a complete browser(!) with your application if you want to run your silly 5kB js-“app” (+ 500MB js-framework code) on a desktop machine (I’m looking at you electron). I’m too old for that shit.

    1. Coding is a good analogy here. I was rather surprised to learn you can now graduate from a good, 4 year comp sci program and not know how to write some C. Tools and tech have progressed and abstracted away the lower levels.

      On some level, it worries me that most “new” coders I’ve met have no real understanding of how a computer works. They’ve heard of pointers and bit fields but it’s not something they fully understand in practice. On the other hand, there’s a bunch of stuff to learn and only 4 years to do it. Something has to fall off the agenda.

      1. Typical EE program in the EU is three years; yet the Europeans, for a significantly smaller cost, manage to graduate kids that are functional and useful for basic stuff. In the United States, four years may or may not produce someone that knows how to bias a bipolar transistor or program in C.

        In the USN and USMC, we had avionics ‘A’ schools where you learned basic electrical and electronic theory and principles and troubleshooting before being sent to a ‘C’ school to learn your specialty. Used the same test questions (transistor theory, AC electricity, network theory, etc) to screen candidates until about 10 years past, when the boss said to stop because too many were being disqualified. At this time, we no longer interview candidates have less than 10 years of experience.

        FWIW, most of my relevant electronic theory was from ‘B’ school. Did not learn much more electronics in college, but did learn much math and physics and literacy crap. And they no longer have a class B avionics school. Now get off of my dirt….

        1. interesting you characterized math, physics, and literacy as crap. Persons great with all three of those, is why you where able to learn what you value. That or you are pissed off because you have no grass or lawn to chase me off of. ;)

          1. A limited grasp of higher math can impact your studies. I never did truly understand div, grad and curl…
            …So Maxwell’s equations are a mystery…
            …which has impacted my understanding of EM propagation, antenna theory, and waveguides [magic rainpipes]

          2. Alan Campbell: I disagree. At the time I learned Maxwell’s equations, it was explained using concepts such as “area” and “flux”, and while the div, grad, and curl equations, and even the triple integral versions were given, the explanations were framed in concepts that could be visualized. The higher math can add some elegance to the concepts, but I challenge your assertion that without them, Maxwell’s equations are a mystery. Maxwell’s equations CAN be explained in simple ways, because all four describe simple concepts.

    2. Most programming jobs now-a-days are for web development, so that’s what the schools teach. The few jobs that exist that are lower level, pay a fair wage, but it’s not glamorous. THe lower level jobs tend to attract the geeks that want to learn low level anyway. Of course there is spillover when management hires “any old coder”(tm) for embedded work, and then yells at the hardware team that the system needs more memory and a larger processor.

  6. An Arduino doesn’t really teach you anything about electronics.

    If you want a good solid foundation in understanding electronics, universities have basically been teaching it the same way for 50 years.

      1. PID was glazed over in the last weak of classes. Focus was intro. Favorite was VFD. My son actually compressed the three uses of the 555 circuits into one quad nor chip in Minecraft in less time than it took me to read the print on the dipchips!

  7. The focus of my undergrad EE degree was electronics, so at one point I knew a good deal about how transistors worked. The focus of my graduate EE degree was physical electronics, so at one point I knew a good deal about why transistors worked.

    I am still employed as an EE and I still do quite a bit of design, but I can’t say I’ve ever used really any of the transistor knowledge that I once had, and it’s mostly gone now.

    1. Same. I’ve never had to code ASM, build a class a anplifier or anything like that. You just buy a chip from Maxim or Allegro that fires the work. I don’t see any issue using it even teaching at higher abstraction layers. Most people complaining about it never use Maxwell’s equations (I’ve never had to fight anything since school). You don’t use those because we abstract it to make it easier and more useful. If you had to understand everything about everything then nothing we have nowadays would exist.

      Now by the time you graduate with a BSEE or BSCompE you should know how to figure out how bias transistors if you ever needed to, or know OF ASM and understand how to figure out how some code works, but I wouldn’t expect everyone coming into the program to know any of that. The key is the ability to learn and understand how something works should you need it, not memorize all the covers from the entire field which you will never need or use in industry.

  8. Starting with those basics often pushes folks out of engineering. Start with some magic. The magic may just be the motivation needed to trudge through learning the basics.

    It’s doubtful that anyone looked at the forward drop of a diode and was inspired to build a blinking led let alone a ham radio…

    1. Will admit that the initial three or four semesters of an EE program are full of ‘gateway’ courses intend to weed out the pretenders(Physics, Calc, Chem, etc). So what should we start with?

      Part of the value of an EE program is/was an indicator that a person found a way to get through some demanding stuff while being bombarded with many distractions.That is, they had a focused and disciplined approach to solving problems for both short-term and long-term projects, or they a large supply of passion and determination.

    2. The first mod [hack?] I ever did was adding a small battery and resistor to a crystal radio. A small DC bias, bringing the diode closer to conduction.

      These days, I’d probably use a small solar cell.

      1. Interesting what was the intent and result? The diode is there to rectify radio signals. Unless you where very far remove civilization, there’s nearly always enough RF energy to cause germanium diode to conduct, or Galena if you are speaking of an actual true blue crystal radio

        1. Biasing the diode closer to its knee is actually a very good idea – it’s the best/simplest/cheapest improvement you can make to a crystal radio. My own experience with crystal radios in the Los Angeles area (hardly “very far remove civilization”) was that only four or five stations were really listenable.

  9. I started super early, IC’s were big, and my first hw was an elenco 30 in 1. I made an am radio. Later when I was a bit older with income to spend on kits, I got a talkingelectronics pic lab 1. Taught myself asm and how to work with mcu in risc and after a single project entirely in asm went on to teach myself C.

    Its interesting that we are reading this post , Ive been getting asked what kits to use to learn electronics / arduino and caught myself complaining about even the best kit choice I found for a lack of caps and resistors just the other day.

    That AM radio was epic. You see I loved golden age radio as a kid and it enabled me to hear the local weekly broadcasts. I can still remember being amazed that the bits I cobbled together formed an almost magical device capable of extracting my precious old time radio shows from thin air. I think todays kids are so spoiled with iphones ipads pcs, touch screens heck even smart tvs that expecting them to get excited with anything short of an arduino or raspi is an unreasonable expectation. Instead perhaps jumping on the bandwagon of the duinopi train and producing kits for TOOLS that will elevate their abilities.

    For instance a logic probe like the elenco one I still use is a perfect candidate for a little math tutelage using transistors and only them. The parts are cheap as can be, and if our goal is educating and not getting rich, we can do it with out the 30$ markup.

    Then we need some sorta multi channel logic scope (over time display) so we could do the same with a logic like tool (again if our intention is to teach we can make sure it is affordable.

    Lastly we will eventually need a scope.

    Arduino was intended to get people interested, AND to allow them to quickly make things. Artsy people could now make blinken lights to enhance their art installation. So arduino / raspi are doing their intended jobs perfectly. What is missing now is the stepping stones to making actual electronic engineering (math) be exciting. The obvious way to get their attention is to save them money, and teach them to make projects you could do on an arduino for pennies with out the arduino, and have the tools to work with it in the process.

  10. Some time ago I was buying some bipolar transistors, got a pack of 100, and then had a talk with my dad which sounded like this:
    – Why do you need 100 transistors?
    – It’s the smallest pack available.
    – Nice, when I was young I had to probe a crystal to find good spot for making a crystal radio (diode). You have it good now.

    1. Makes you appreciate the design and technology in the old tektronix scopes. They had 50+ tubes in them and a lot of them were twins, but you go through more gain elements than that in one or two quad op amps these days.

  11. Why just last night I was building a device with sets of Darlington pairs. Ok, it was a ULN2003 but I still take old school cred, since the ULN2003 is probably an older chip than most people here :-)

  12. Cub scout meeting. An electrician showed off a dimmer and I was off to the races. Self-taught after that mostly by our friend “springy” the 100 in 1 and some books TTL cookbook etc

  13. Wait until the supply of inexpensive “IC for every application” dries up. We will be back to pulling the components off the boards. And discretes will be the most valuable items. Everything goes in circles …

  14. Quote: “the ULN2003 is probably an older chip than most people here :-)”
    WARNING: OldGuysTellStories….
    Actually I’m older than ANY chip. I started in the very early 50’s with vacuum tubes, like a 1 tube receiver. Fortunately, today I am designing kits for young kids to learn about wires and sensors and microcomputers and code and LCD displays. So I CARE about this!
    I am concerned about the lack of basic electronics knowledge in this era of such high-level hardware and software. I’m trying to think of an example to punch down through the layers of abstraction on an Atmel/Arduino to a discrete transistor. Or even from the RaspberryPi level from a touch screen down to a transistor turning on a small DC motor. Hmmm….
    The most popular kit I make ( http://arduino-info.wikispaces.com/YourDuinoEngStarter ) does have a discrete transistor in it. But I like the idea of a few more discretes thrown in there, and a Capacitor (whatever THAT is)..
    I think there is an opportunity to use Arduino as a signal source and signal analyzer (Hey, that’s Only Code..). Then 75 cents worth of transistors, diodes, resistors and those capacitor things could be used for lots of circuit examples and experiments.

    But somebody has to do it… I would appreciate pointers to any examples you see that might apply. Anybody want to collaborate on this??
    Regards, Terry King
    …In The Woods in Vermont, USA
    terry@yourduino.com

    1. I love that train of thought! I’ve got my brain phase locked loop on extracting binary from sram card with Eeprom backup, 50 pins, looking into universal paralel / series shift registers. Desire is to bust somebodies binary conversion to ascii. But that train has been dwelling in idle a very long time! Glad to hear someone vocalize the concept! Thankyou

  15. I’ve searched the web for a simple transistor and resistor circuit to connect a cheap Chinese induction sensor up to my 3d printer for use as a bed leveler. I need the signal from the sensor inverted. Sounds simple enough. The sensor needs 6 to 36 volts to operate. There’s 12 volts or 5 volts on most 3d printers. Many ways to skin that cat from simple resistor voltage dividers to simple transistor circuits to digital logic chip circuits. Rather than reinvent the wheel I looked to the web first to copy what someone else already figured out. Since I have simple transistors,capacitors, diodes and resistors on hand; that’s what I wanted to use. Nope, not gonna happen. It looks like no one has tried this method. Really? 100’s of 1000’s of these things being converted to use automatic bed leveling and no one hooked up their sensor with simple components to invert the signal? I’m not going to use the resistor voltage divider method other than it relies on a steady voltage to prevent your motherboard from being fried and I’m not really a fan of that plan of attack.

    Well, I’m still a newbie when it comes to electronics but my first introduction was building an audio amplifier from an integrated circuit chip. When I say “chip”, that’s stretching it a bit. Integrated circuits were a somewhat new invention in the early 70’s and this chip alone is bigger than some of the Arduino’s I own! The same thing today would probably fit on the point of a pin. It came shipped in one of those two part plastic hinged clam shell cases; black bottom and clear top that people saved to put other things in. I’ve been tempted to put it up on the internet for sale; it’s gotta be a rare item! It’s an RCA KD 2131 integrated circuit with CB 2H written below the chip number. It’s a ten pin chip but each one of those pins is about 150 thousandth’s wide.

    1. “I’ve searched the web for a simple transistor and resistor circuit to connect a cheap Chinese induction sensor up to my 3d printer for use as a bed leveler. I need the signal from the sensor inverted. Sounds simple enough. The sensor needs 6 to 36 volts to operate.”

      An inductive proximity sensor typically has an NPN transistor (open-collector) output. Maybe PNP, in some cases.
      So what sensor do you have exactly?

      Because it’s an open collector, there’s no dependence between the supply voltage to the sensor and the supply voltage to the system that’s reading that signal. You just need a pull-up resistor to the 5V rail (or whatever the logic rail is).

      If the logic input to the system is an optocoupler, then just sink current through the optocoupler LED through your open-collector sensor. (Not relevant in your case, but not unusual in industrial automation.)

      You can run the sensor on the 12V rail (or, if you’ve only got 5V, test it – it will probably work.)

      Can’t you do the inversion in software? Most decent motion control firmware for an Arduino (or whatever) such as grbl should support this.

      If you’ve got an NPN open-collector output and you really want an “active high” current-sourcing output, then use a PNP BJT with the emitter connected to Vcc, the sensor collector connected to the base along with a pull-up resistor to Vcc and the output from the collector.

  16. Apparently none of you have ever heard of Heathkits. For decades they were the gold standard in kits for everything from their version of the RS 200 in one ( which was 20projects) up to a color TV and HAM gear ( much of which is still in use). I grew up during the transition from vacuum tubes to solid state. My father had a home built ” hi-fi” which transitioned from a monoral vaccum tube system to stereo and then to solid-state via Dynaco kits. Wow what a trip it’s been.

  17. Uncle gave me his war surplus 30 MHz shortwave. Soon I bought a $3.95 copy of ARRL. I was 8 and out in the sticks with limited TV and limited permission to watch .THE channle with parents smart enough to NOT tell me that ARRL was an adult book. Plainly, it is not. Oddly, I started in microprocessors with a small startup and went to waveguides and sattellite and radar working my way backwards to the lowley transistor… and even one tube. During my last Silicon Valley job did I fix a Hartley oscillator… a klystron of 25 Kw used to melt a lamp element inside a quartz tube on a production line. Everything I did came from very basic knowledge, jumping comanies about each year.

  18. In Australia we have been recently treated with a new electronics magazine. DIYode they have been trying to cater to a very wide audience and as a result some of the articles are a bit bland but they have been running a fantastic series on basic electronics as an example this months issue was parallel and series circuits.

    There is hope yet :)

  19. I like the pictures of the inside of a transistor showing a little man watching an ammeter (from B to E) and adjusting a rheostat (from C to E).

    Someone needs to copy that picture, but change the ammeter to a voltmeter and call it a MOSFET.

    1. after an X in one kit I moved on to those as well – still have them for my boys to start on when they are a bit bigger. Though I never built the beer powered radio …..

    2. I had one of those as as kid as well, with the blue plastic breadboard that you screwed things into. Then I moved onto the second book where actual soldering was involved. Did a little bit of electronics in high school but then various factors resulted in me picking computers and programming (another thing I was hugely interested in at the time) rather than electronics. (I will never forget the time I came up with vague never-went-anywhere plans for traffic lights for a particularly busy stairwell at the private school I was at).

      My electronics experimentation would have been late 80s-early 90s and the most sophisticated component I remember you could get from the catalogs the high-school electronics teacher had in the back was a Z80 of some sort, getting one (and the bits to make it work) was out of my league though.

      1. The earlier version of the board was a piece of partcle board with holes drilled in it for the screws.

        The last version I saw had springs that pushed into the holes in a plastic base ( that may have been the jaycar short circuits version)

    1. I too stated in electronic with the Philips “Electronics Engineer” kit. Addictive. Needed to spend my whole week’s pocket money buying another AC126 transistor after a lead would snap off due to continuous bending to fit them into the spring terminals on the breadboard!

  20. Ok now I feel old, Started out having to studying the cathode voltage of the vacuum tube we had to use. as well as the grid bias.. Before I was out of school there were these new things called transistors . They explained that they worked sort of like tubes. Now had to design and make stuff with these. Then while I was working these integrated circuit things started showing up.. Then micro chips. It’s been a long ride.
    I no longer repair stuff for a living, But I still fix computers as a hobbie

    1. I had a similar experience. From Valve Analogue to Germanium Transistors to Silicon Transistors to Analogue IC’s to Small Scale Digital IC’s to Medium Scale Integration to Large Scale Integration to Microprocessors to Micro-controllers to FPGA all seem like just little steps along the way now.

      I can often tell the approximate age of a design engineer by the components they choose to use in schematics especially with Transistors/FETs etc.

      Also, I think you meant Anode Voltage. Cathode voltage is normal 6.3Volts. But then that was a ling time ago.

  21. “…When so many of our projects on sites such as our own Hackaday.io tend to feature a single board computer, can we really blame the youth for following our lead?”

    Careful!
    You’re (a) getting much too close to the truth, and (b) in serious danger of being sacked, because of it.

  22. “… But let’s face it, spending time learning Linux and myriad programming languages is an important part of the upbringing for the electronics hackers of today…”

    …then that’s what you and Eben Upton, e.g., get: hackers.

    “Be very careful what you wish for…”

  23. Nope…. gonna die and take all my hard learned knowledge with me. All those “outdated” component level repair or design skills. The wave of the future is board swap! (Or just replace the whole damn thing.) My suggestion… become a network engineer.

    1. Wow, they’re still going. I bought every magazine of theirs in the 70/80s.

      On their KITs page they still have the TEC1A/B/C microcomputer trainer kit.

      It would still be great as a learning tool for ASM. It is based on a Z80 that has a Complex Instruction Set (CISC) as opposed to modern Reduced Instruction Sets (RISC) like AVR but it has no config registers so it’s dead easy to learn and you only need to know a handful of instructions to get going.

      It would be nice to see an updated version with perhaps a Character LCD or VGA and a better keyboard.

  24. With electron microscopy and similar becoming regulars here i predict that we will soon start with how a transistor is made/works and then work up to high power cpu’s and their programming

  25. There was this thing called a flashlight or electric torch to euro folks. It used something called a lantern battery to push a current through a wire filament which would heat and glow in a vacuum tube and consume huge amount of power to make a little bit of portable light source.. It had an electro-mechanical device called a switch made of brass that would allow the device to be turned to one of two states. Brass also made something called a circuit path. On and off states where useful given the limited run power source. That’s where the disease called curiosity started.
    Snickering.
    Transistor kits and solderless many-in-one multi project kits are still here. Velleman and Elenco come to mind. Some kind of multimeter should be purchased along side these to get some usefulness other than “cool” factor and soldering practice.

  26. Transistors are a piece of cake. “Pull” base away from direction emmitter arrow points and transistor should turn on. Think of E-B junction as a diode. So you should NEVER see more than a diode drop between E-B (Unless transistor is reversed biased). Follow those basic rules you can trouble shoot with a DVM. (Enter FETs, darlingtons et all to confuse the issue.)

    1. Sounds good so far. The next step is to ensure that is a good signal diode, i.e., does the manufacturer classify it as a signal–or switching–diode, as opposed to offering it as a
      ‘rectifier’ diode (or for solar-panel use, where low Vf is mandatory)?

  27. > We have to “…start them off…” with AM radios? My experience is that ‘starting them off” as simply as possible yields the most fruit: a crystal (wind-the-coil-yourself/germanium diode) radio; a knife-switch,…

    I may have read ARRL at 8-10 but I was 11 before I found galena amoung the railroad tie-track base. At 8 I fixed lamps and toasters..At 12, those and lights and switches at friends’ houses. Moms were grateful but admonished to not ho near the fuse box. Those were dangerous, after all. It did teach me go work safely… albeit with power applied. LMAO!

    It has been a great pleasure, ladies and gentlemen (& kids…all.)

    1. Great background!
      [I was forced to “go near the fuse box”, as a pre-teen, after reading about carbon-arc lamps in Popular Science, and making my own with two fat carbon electrodes from carbon-zinc 6v batteries. Wired directly to 120 v, of course. Blew every fuse in the house. DID make a very bright light–for maybe 100 ms or so. I really wish Popular Science had gone into more detail, like the fact that a ballast was needed…]

    1. Only if you don’t know of Jeri Elsworth, who is a self taught yet highly competent engineer. My point was that there are exceptions to the rule that the ONLY path to success is through acedemia. I myself am a self taught engineer. I retired from a broadcast engineering position at a university that required a degree or equivolent. Thirty plus years of experience was more than good enough. As was my hands on experience twenty years before to get my first “engineer” position. I don’t advocate such a path for anyone, but it can and has been done.

      1. > “that required a degree or equivolent.”
        HAHahahaha… ahem… ok. Typo… ????
        But I can say that I read of the guys in sillycom valley, going to school AFter starting the personal computer mess… and stating that it was not so very easy. School is good. It’s nice if that can be half of it. Be glad I never had much of it… eh, Pinky?!? Be verrry glad. Hahahahaha!

  28. Find what makes the kids you know tick. For me it was guitar pedals or other musical efemera.

    I had my 100-in-1, 200-in-1, etc as a kid and built a bunch of different circuits. It was once I found out how to apply what I had learned so far to music that I found a reason the understand what was going on and how to design my own.

    The simplest is a single chip amp, capacitors as needed to get the -1 to 1 V signal in to the amp out at around -2 to 2V, then some diodes to clip that back down. Add resistors or potentiometer to act as the volume knob. Easy to teach, easy to change.

    But then you can break it down. What and why are the capacitor at the input, why does this schematic require voltage added to the input while another doesn’t, what is going on after the amp that affects the sound? That lets them learn about capacitive coupling, how different amps work on different voltages (0 to 5 or -3 to 3, etc), what the amp characteristics are and why that matters, and what each of the different kinds of filters do to the output of the amp. From there, the world of phase changes from each component, how that affects the analog and digital audio.

  29. I just finished reading Charles Platt’s “Make: Electronics”, and am partway through his “Make: More Electronics”. These are exactly what you’re looking for.

    Very hands-on, just enough theory and history to inform you, but overall giving a very intuitive understanding. I started playing with electronics back in ’81 using the books of Forrest P. Mims and George Young, and have just gotten back to it after a brief 37-year hiatus.

    Platt really made me understand the discrete components from a functional standpoint. I’m working on a blog post on that very topic, since all those “extra thingies” surrounding the chips were so mystifying.

Leave a Reply to AndrewCancel reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.