Intro: Semi Conducting - a Guide
WORD OF WARNING, THIS IS LONG.
Anyway, Electronics, as stated before by many of my comrades around the globe, have all become the most important things in our daily lives. Cell phones, computers, TV's, keyboards, mice, speakers, everything all uses electricity and of course, SEMICONDUCTORS.
You could easily try and open up your phone, or your xbox, and there's hundreds of big, small, tiny, and huge black things lurking inside them. Those things have silicon, among other metals mashed together inside of them.
Asking around, and getting some odd looks, almost no one knows how a transistor works, a mosfet, or anything except for the most basic of electronic components, and even then that's a stretch.
This guide/bible to the almighty semiconductor will serve you well, hopefully.
Yeah, I wanna win a 3-D printer...(and maybe the laser pointer, but i won't be greedy or lie) I'd like to make encasements and parts for electronics, and not just make 3-D bunnies and boxes all day long like some other guy probably would. I'd like to actually put the thing to work, and make some decent cases out of it! Although my instructable isn't as great as Grenadiers one on HV, I hope that we'll both be able to interest the average person back into electronics and bring back the electrical curiosity that nearly everyone had back then, even if he didn't win. We'll hopefully find a way to do it. And, yes, I do know him, and he was okay with me doing this guide.
All I can really say is thank you for taking your time to read this, if you do. I put quite a bit of time into this instructable guide.
Please try and spread this guide around as much as you can!
And, I know fairly well that many of you instructable users won't be too interested in this because it's not ..."cool". And, I understand your point, to an extent. But, learning all of this allows you to DO cool stuff! It allows you to DESIGN your own cool stuff! Think about it; wouldn't you feel fantastic if you just discovered a new type of semiconductor, or a new way to use an existing one? What are semiconductors used for? Everything. High voltage is just one flashy thing, but seriously, you could make a new ...anything. A new transistor. And believe me when I say this ; anything new and amazing in the electronics world will be here to stay for a long. long. time.
That's why this guide is here. ^^
Step 1: Packages An' Stuff
Before I begin explaining all that silicon, I must say a word about packages.
Every semiconductor that I know of out there exists with several types of packages. These packages are what hold the little chip inside, and serve to allow us with our fat fingers to be able to use them in everything, and plus it makes it a heck of a lot easier, soldering wise.
Some packages have different pinouts. Usually it is wise to look up the datasheet of a particular semiconductor, to check the pins and see what leads where.
Smaller, unheat-sinkable diodes are usually in DO-xx packages. They look like little pills.
Larger diodes are in regular packages used by just about everything else, and are heatsinkable with the addition of the hole in the package.
MOSFET's, transistors, IGBT's, Voltage regulators, and certain diodes can all use the same exact package, for cost saving and ease of use, as most pinouts for a specific package are the same across the board, no matter the semiconductor.
The rule of thumb, heat sink wise, is the physically larger the package, the more heat it can dissipate. A small TO-220 package can usually dissipate around 50 watts, while larger TO-247's and TO-3 packages can dissipate upwards of 150 watts, or even 200 watts!
Whenever you heat sink something, be generous. It is much better to have an oversized heatsink, and a nice cool semiconductor, than a small undersized heat sink and a now melted blob of plastic stuck to it.
Fans also increase the power dissipation of everything. Use them if necessary.
Heat sink goop is very important, along with insulators, if you plan on using more than one semiconductor on one plate.
When applying heat sink stuff, be sure to use a small amount, usually about maybe 3/4's the size of a grain of rice, and screw on the semiconductor tightly. Using an insulator, like a Sil Pad (silicone pad), or a mica wafer, you need to apply roughly the same amount of stuff to BOTH sides, as this helps transfer heat better.
Insulators prevent semiconductors from being electrically connected, when they aren't supposed to. Most packages, at least most MOSFET, IGBT, and voltage reg packages have a metal backing to them; this increases heat transference, but they are often connected to some lead of the semiconductor. Therefore, use insulators to protect things from touching in bad ways.
First package is a TO-3PN package. Always has a metal backing, needs insulator if using more than one on a single heat-sink.
Second Package is a TO-220 package. This one happens to have a metal backing, so needs an insulator.
The package after that one, is a TO-92 package, for small signal transistors. They use no heatsink therefore require no insulators.
Package after that is a SOT-32 package. It has a small metal backing, and require an insulator.
5th Package is a TO-220 package, and has 4 leads. These types of packages can have a number of leads, actually. Plastic backing, needs no insulators, but can not dissipate as much heat as other types due to this.
6th package is a special variation of the TO-247 package, it just has longer leads than standard ones. They all have metal backings, and these can dissipate a good quantity of heat. These are also the more preferred package, because of their low inductance design when compared to the TO-3PN (whcih has that big metal flange on the back). Needs an insulator.
7th Package is a TO-264 package. This is the largest size out of none brick packages. These house the more power capable semiconductors, and can dissipate the largest amount of heat when compared to other packages. These have a metal backing, and from my experience, somewhat thin/weak leads. Be careful when handling these. Insulator is needed, as usual.
8t-9th package is a Dual Inline Package, or DIP for short. These are used all over for IC's and the like, and are easy to handle, easy to solder, and easy to kill, (they can't handle alot of power, or dissipate any power!). They come in two flavours, plastic, and ceramic. Ceramic IC packages are usually older, and can dissipate a little more heat, if any is generated, than the plastic variation. They can have as few as 6 pins, and range all the way up to 40+ pins, in microcontrollers.
The 10th and final package is a SOIC package. This is a Surface mount package, and these are an absolute pain to use, due to their size. A dollar bill is shown in the picture to give you a rough idea of just HOW small these suckers are. You can use them, but you need to usually either 1. Make a PCB specifically for them, or, if you're going to use them on a breadboard/perfboard, then 2. create a DIP package yourself. This can be done with a fine tip soldering iron, and a LOT of patience. There are quite a few instructables out there how to do this yourself in more detail.
There are more surface mount packages, and they all range in size, pin count, and pin spacing. Generally, the one with the largest size and largest pin spacing are the easiest to use. (still hard though).
There is another type of package, called a Ball grid array. These IC's are absolutely near impossible to work with, if you don't have the tools necessary to use them. Generally, avoid them, unless you KNOW what you're doing with them.
Thanks for fixing the notes by the way!
Step 2: Diodes
Diodes are perhaps one of the more common components, and were among the first to be invented actually. A diode is a type of doped p - n device that prevents current from flowing in one direction, allowing for alot of interesting things to be done.
A diode can be used to rectify alternating current, which will be explained in a later step,
A diode can be used to create a radio, because of the above property, it can demodulate the carrier signal and output a pulsing DC wave, which is the sound you could hear. Most radios nowadays take advantage of more complicated silicon, increasing the part count dramatically but also increasing loudness and ease of use, along with better sound quality of course.
Diodes are quite inexpensive, compared to other semiconductor devices, because it's relatively easy to manufacture compared to other doped silicon, as well as because of mass production and demand.
There are TONS, and I mean TONS of different diodes, but they all perform (nearly, there is an exception) the same thing as every other diode.
Diodes to exhibit some odd characteristics, mainly one being the voltage dropped across it. Most standard diodes have a voltage drop of from .4 volts to .7 volts. Certain diodes have even lower voltage drops, some Schottky diodes have voltage drops of only .2 volts! This voltage drop is important, because it allows you to determine how much heat a diode will create, based off of how many amps you plan on putting through it, and if you need to heatsink it or not. Most diodes are pre rated for you, however.
Step 3: Signal Diodes
A signal diode is typically a small diode, sometimes in a glass case or in a plastic case. They can handle low currents, and low voltages but they have an advantage of having extremely fast recovery times, however some are somewhat slower compared to the fast ones. The purpose behind this, (why not just use a rectifier diode?) Is because of space constraints. There are a LOT of parts on a computer circuit board, so they really have to get parts down to small sizes.
Signal diodes are obviously used for clock signals in digital devices, and serve to prevent a reverse signal from damaging the microcontroller.
List of some Signal diodes
1N34A (germanium diode)
Step 4: Rectifier Diodes
Rectifier diodes are the more common ones out of all diode types, because they're used so frequently in nearly everything.
Rectifier diodes are used to rectify alternating current, or current that changes state from one voltage to another. Because a diode prevents current from flowing in a certain direction, they can prevent that change from the negative state, and thus allow DC current to be created, with additional components of course. 4 diodes are arranged to do just that, and it's called Full Wave rectifying, where both sides of the wave are turned into a DC pulse wasting very little electricity from the AC current. Half wave rectifying is the equivalent of taking one diode and putting it across an alternating current source; this wastes one half of the sine wave.
This property allows these diodes to be used in just about every wall power source you can imagine. They all change that 60 hz sine wave from your wall outlet into clean, DC power for your computer, cell phone, or whatever your electronic device is. (with a smoothing cap too!)
However, unlike signal diodes, these diodes can have a wide range of recovery times, to standard diodes, which can effectively recover at 1 kilohertz AC current and lower, to Ultrafast diodes which can recover extremely fast, all the way up into the megahertz region. The voltage drop for most of these diodes are roughly the same, around .6-.7 volts. High voltage variants can have massive voltage drops, from 25 volts to 45 volts, but when you're working with voltages anywhere from 5 kilovolts, to 50 kilovolts, 45 volts is not going to matter too much.
These diodes can also range from low current, 250 mA, to hundreds of amps. Usually, however, if the diode is going to be conducting more than 4 amps, the diode NEEDS a heatsink, or it will overheat and release it's magic smoke.
Some rectifier diodes
IN4007 (the IN indicates standard recovery)
UF4007 (UF indicates ultrafast recovery)
Step 5: Schottky Diodes
Schottky Diodes are a very special type of diode, formed not by doped silicon, but by contacts created by a semiconductor and a metal. This property was inherently discovered in Cat's Whisker Diodes and radios, where a small piece of metal, sometimes a rusty razorblade or a piece of Galena (a semiconductor), was poked and prodded with a needle until the radio started to transmit sound. The purpose behind this was to (here's that word again!) rectify the radio signal, so that it creates pulsating DC current to the speaker, generating sound waves. This worked fantastically, but it was extremely unreliable, and hard to make because 1. Even the slightest vibration could cause the diode effect to disappear, or 2. the actual diode itself was hard to find, as the needle contact needed to be in a specific spot where it would rectify the radio waves.
Schottky diodes have the lowest voltage drop of all the previous diodes mentioned; sometimes as low as .2 volts!
These diodes due to their construction also have no recovery limit, meaning they can rectify just about any signal, hence the ability to rectify 100 + megahertz radio waves, or even higher, upwards of 100's of GHZ!
So, this diode sounds pretty magical, and amazing, right? Why are we even using those other diodes,? Especially since this one is so fantastic? Well, the Schottky diode has one limitation, and that's voltage. Schottky's can only handle upwards of 200 volts, limiting their use to signal purposes and small voltage rectification. Current wise, however, they can handle just as much current as any other diode, but just as before they will need a heatsink if you plan on rectifying over 4-5 amps.
There is another type of diode called a Tunnel diode. This type of diode is similar to a Schottky diode in a few ways, mainly the fact that it's low voltage, AND low current. A tunnel diode is a special type of negative resistance diode. This means, that as voltage increases, current decreases! This effect only occurs in a certain voltage range, however, and makes tunnel diodes only useful for low power applications. Not to mention, they can't handle much current either. They can oscillate at pretty much any frequency, however, from 1 khz to 1000 of GHZ. They're pretty neat, and this is sort of the basic principle behind Magnetrons, the diodes used inside of microwaves to oscillate at 2.43 or so GHZ.
Some Common Schottky Diodes
Step 6: Zener Diodes
Zener diodes are a special, specific type of diode. They still rectify current, have a voltage drop, and a reverse recovery time, though this part is negligible in most means. However, unlike other diodes they do allow voltage to flow in both directions, under a specific circumstance though. This is called the "zener effect" and is much like avalanching, which will be explained with TVS diodes.
The zener effect is an interesting property of zeners. Effectively, once a certain voltage is reached the zener starts conducting in both ways, but CLAMPS the voltage down hard to that specific zener voltage. Think of it as a controlled voltage drop device, no matter what the current or voltage across it is, it always drops that specific amount of voltage. For example, a 5.6 volt zener diode has 2 volts flowing from cathode to anode. It will not conduct. If the voltage is brought up to, say, 6 volts, then the zener diode will begin conducting, and will drop 5.6 volts off of that voltage, even if it's 100 volts, it will still drop 5.6 volts.
Anyway, this effect is incredibly useful in alot of things, such as protecting MOSFET gates, or protecting sensitive circuitry from overvoltage. They are also used in voltage regulators, as the voltage reference for the circuit to choose from. Zener Diodes are not rated by amps, but by wattage instead, since the input voltage can be anything. Common zener diodes range from .25 Watts to 1 watt, although some zener diodes do go up into 10 watts, or even more.
Common Zener Diodes
1N4744 (15 volt)
1N4742 (12 volt)
1N4734 (5.6 volt)
Step 7: TVS Diodes
TVS diodes work in a very similar manner to zener diodes. They also clamp the voltage to a specific point, but instead TVS's are used to protect very large devices, and are even sometimes used as Lightning protectors. They drop off a LOT of voltage. They can handle much more power than zener diodes, and are often large in size because of that.
TVS diodes, much like Zener diodes, are rated in wattage, and are generally over 1 kW of input power because they can withstand very high amperage at a very fast pulse rate. This is often why TVS diodes are so large, because they don't effectively need a heatsink as it only changes the amount of time it takes for the diode to cool down. However, this means that TVS diodes are only good for nS (nanosecond!) pulses of high voltage/high current. Any longer, and mister TVS goes boom.
Littelfuse happens to make Lightning Protection diodes, and I nicely asked for some samples and they gave me some. The diode you see in the picture is a 1500 watt TVS diode. (surprisingly small!)
There aren't really any common TVS diodes, so to speak, but they have common brand names, like Transil (stsemiconductor) or Transorb (Vishay)
Step 8: LED's and Laser Diodes
Pretty much EVERYONE know's what an LED is. It stands for Light Emitting diode, and exhibits the same exact functions as a rectifier diode. They use extremely low power, and are, with few exceptions, physically small and generate very little heat. LED's come in a vast variety of types, not just colors either. Infrared LED's and UV LED's are some to name a few, and even then there are more flavors of these types of LED's as well. Wavelengths, viewing angles, and all sorts of different characteristics exist for LED's. However, there does exist a problem with LED's. An LED can not withstand very high voltage; only around 1.2 volts, (low power common ones anyway!) and need to be current limited otherwise they will self destruct. It's incredibly easy to current limit LED's with a simple resistor (I assume you guy's know what that is) to limit the current to an acceptable level. Every LED has different amperage ratings, mostly in the mA range for small ones. Some LED's, because of their brightness do output respectable amounts of heat, and therefore require a good heat-sink. Laser diodes are great examples of this;
High power laser diodes (diode? Hello...) have to have pretty large heat-sinks, and thus large enclosures and cooling systems to prevent them from popping. Laser diodes are basically a special form of LED; instead of shooting out light everywhere, they concentrate it into a single beam path, and thus form the laser itself. This prevents the energy from being spread out, and therefore, lasers in general can be seen, very very far away. There are several different types of laser diodes too, IR ones, color ones, (all colors too, but each one costs a different amount of money due to intensity and manufacturing processes. Red is probably the cheapest of the colors).
The smaller lead on the LED is the cathode, or negative lead.
An LED can rectify current just as any other diode, as previously stated. You can use it with AC current, and it will created a half wave rectified DC pulse, or, if you're feeling weird, using a full bridge you can make an LED bridge Rectifier (now who would want to do that?)
LED's have voltage drop, too. About the same amount as rectifier diodes too.
LED's are used in just about everything, (well almost everything) including speakers (you know the little light settings that show volume?), computers, mice (the lazer underneath, unless you're still using ballsy mice), cell phones (backlighting, and keyboard), as well as normal, standard lighting.
Laser Diodes are used in quite a few things as well. Laser mice, (that's a duh), your CD drive, blu-ray drives, fiber optics use laser systems to project their information over a long distance... Laser dot sights in the military, Laser engravers use CO2 laser systems (not a diode, though, something different). Clubs use them in their rave parties. ;D
Step 9: Varicaps, or Variable Capacitance Diodes
Variable Capacitance diodes are a replacement for mechanical butterfly type capacitors found in radios. They use a special type of diode that enhances the capacitance (this is found in all diodes, actually but most manufactures tend to try an avoid this to prevent ringing and oscillations).
The capacitance is a variable of the voltage input through it. Different voltages tend to change the capacitance of the diode, allowing for fine controls and the removal of mechanical variable capacitors. They are cheaper to make, and are found for obvious reasons inside of radio sets and the like.
Step 10: Bipolar Transistors
Bipolar Junction Transistor
Ah, the transistor. The most commonly made semiconductor in the industry, this device quickly replaced the vacuum tube in many ways, for many reasons. In the early days, if there was a transistor in a radio, it was more than likely the most expensive part in it. Nowadays transistors are in everything, from cheap to expensive. They have three pins on them, the collector, the emitter, and the base.
There are also two types of transistors; PNP and NPN transistors. Each one has a different property, and therefore different uses, as explained in step 8 and 9.
Transistors have a few problems with them, mainly anegative temperature coefficient, and voltage drop. A negative temperature coefficient, basically means, when a transistor heats up, it starts to conduct more current. This increase in current, causes the transistor to heat up further, and further, until the transistor is either saturated, or, if the transistor has inadequate heat-sinking will blow up. A voltage drop is how much the transistor basically, loses voltage when it is on, or even off. You can figure out how much heat the transistor will make, by taking the voltage drop of the transistor, and multiplying it by the current you're trying to amplify or switch. P = Ed * I where P equals power in watts, E equals the voltage drop, and I equals the total current needed to switch/amplify. A transistor also has a property called Saturation, and this property exists for essentially all semi-conducting devices. Saturation is basically when a transistor cannot be on "any further" than it already is, that means that the load will have the maximum voltage drop across it while the transistor a minimal voltage drop across it. This reduces heating a LOT for all semiconductors, and is generally a good idea to keep a transistor in saturation if you are switching with it.
Transistors are used in just as many things as diodes, as they too are an incredibly common component. They are used in computers, radios, (as amplifiers), TV's, (amplifiers and switches there!), and just about every electronic thing you can think of, more than likely has at least one transistor in it.
Step 11: NPN Transistors
We begin at the base of the transistor.
BOOM. A signal is applied to the base, and turns on or off the transistor. The transistor suddenly starts screaming, as the current tickles the transistors arm. It starts to let go of the current at the collector. The current then flows from collector to emitter, at a certain amount based off of something called gain.
An NPN transistor may have 100 gain. This would cause, well, let's put it into a scenario. You have 10 mA of current pushing at the NPN transistor's base. The transistor "lets go" of 1 amp of current at the collector, and lets 1 amp flow through it. This amount of current, times the voltage drop will give you a rough estimate of how much heat your NPN transistor will generate, and whether or not you'll need to heat sink it. NPN transistors are used for controlling positive currents.
Common NPN transistors
Step 12: PNP Transistors
PNP transistors work in a very similar way to NPN transistors, but instead of positive voltage. But, they are the opposite in terms of operation. They are on when there is NO base current, and turn off when a current is applied. How much on or off is determined by the gain of the transistor. They are used to control (typically) negative voltage.
PNP transistors, other than that change, are essentially exactly alike NPN transistors, in that they are used in amplifiers, radios, computers, everything electronic.
Common PNP transistors
Step 13: Darlington Transistors
Darlington transistors.... It always sounds like a British term, right?
Well, apparently the people in Transistor's and CO decided, that 100 beta gain isn't enough. They wanted more. So, they did something anyone else would have done, and stuck two transistors together.
What? Right, two transistors, stuck together in a fashion called a Darlington Pair. They work in the same exact way as a regular, single transistor, but they amplify many many times more. A typical Darlington transistor, such as a TIP127, or the TIP122 are transistors that have gain's of about 1000 beta (the unit for gain is the beta). Wut. Well, this means that a 100 uA (nano amp) Signal will be output at about... 100 mA, if I'm right. This means you can do some really neat stuff with darlington's. You can make a finger or touch button using this, because your skin has very small currents in it, and that current is enough to let the darlington conduct. Pretty neat huh?
Darlington's are used in places where a signal is exceptionally weak, and work rather well at touch sensitive screens, or also for amplifying signals from radio's. They can also be used for driving larger power transistors, such as ones that have a very little current gain. Arranging them properly allows the gain of each to be multiplied by the other. For example, a 100 gain and 8 gain transistor are put in a darlington configuration. The 8 gain is the transistor on the bottom, and the 100 gain is the upper one. The upper one will increase the current gain of the 8 gain one to roughly 800 gain. Total gain = T1G X T2G. The plus side of this, is the higher gain transistor doesn't have to handle a lot of current. It can even be a small signal one.
Common Darlington Transistors
Step 14: Totem Poles
A totem pole is a certain type of driver, used to amplify square waves (and sine waves, with additional components to prevent distortion!) into a stronger square wave for driving other types of semiconductors faster.
A totem pole consists of a PNP and NPN transistor arranged emitter to emitter, with their bases coupled together. The output signal is taken from where the emitter of the NPN transistor and the collector of the PNP transistor meet. The plus of this is the ease of use, but the downside is that transistors do make heat, and this may (depends on your application) create enough heat to warrant a heat sink.
A totem pole driver can be used to more effectively drive MOSFETS, (which will be explained in the next step, actually), and IGBT's. The increase in current allows the semiconductors to turn on and off faster, reducing heating.
You can use just about any NPN and PNP transistor combination, as long as they can handle the amount of current you need to drive the mosfet or IGBT with.
Step 15: MOSFET
Metal Oxide Screen Field Effect Transistor
The MOSFET can be compared to the transistor as a much more effective switching device. There are three pins, just like transistors, and they are called the Drain, Gate, and the Source (ooo fancy).
The Drain and Source can be compared to the collector and emitter of a transistor, respectively. The gate is exactly like the base, except a MOSFET is a voltage controlled device.
Well, this means that instead of applying current you apply voltage. A MOSFET is a very ON or OFF device.Mosfets are much more suited to switching than being used linearly, but they CAN be used linearly. The heat created by a mosfet in it's linear stage (inbetween on and off) is about the same as a transistor. The only reason that mosfet's aren't suited for this is the voltage level threshold for mosfets is different for each device. That can not be said for BJ transistors. A transistor is best used for amplifying signals, and a mosfet is best used for switching. Before, transistors did both jobs, and frankly they sucked at switching.
MOSFETS, like transistors, have voltage drops, saturation, and something new, ON state resistance. This is basically how much the mosfet resists, as it is saturated or turned fully on. MOSFETS have increasing on resistance as voltage handling ability increases.
Basically, try to think of the MOSFET as a variable voltage controlled resistor. More voltage = more on, but this can only be turned on more to a certain point, until you start damaging the FET. Most MOSFETS have a limit to how much voltage can be put onto the gate, usually it's around +- 20 volts, but some can handle +-30 volts. Exceed this, and you will decrease your mosfet's life on this earth. The on state resistance does allow you to calculate how much heat your mosfet will generate by allowing you to use Ohm's law. (W = I2 * Ron) There are two more unique properties to mosfets, and those two are the intricate body diodes, and the turn off and turn off time. The body diode is inherently formed when a MOSFET is birthed, due to the alternating P- N channels of doped silicon. You cannot remove this diode, it is created as soon as the mosfet is created. If this diode is too slow for high frequency applications, you may need to bypass the diode and add your own ultrafast one.
Now, Turn off and turn ON times are extremely important, when working with frequencies over 5 kHZ or so. Turn on time is basically the amount of time it takes for the MOSFET to turn on, once the gate has received a signal. MOSFETS, when compared to their newer cousin, the IGBT, (to be explained in a bit) are a LOT faster, and have turn on times of anywhere from 20 nS (nanoseconds!) or 20 billionths of a second to turn on, to, any time in the world. It depends on which MOSFET you are referring to. Turn off times, and delay's are often a little bit longer than turn on times. Turn off times and delays can range from 30 nS to 1000 nS. If you turn on/off the mosfet faster than it can, it generates heat, and a LOT of it, plus your circuit will/may fail to work properly. Mosfets also have something called Input capacitance. The way a Mosfet is made basically creates a dielectric within the mosfet. The input capacitance varies, but for the most part it is around 2200 pF to 10000 pF. This complicates driving mosfets, and is the exact reason gate drive anythings exist. The more current that is available allows the input capacitance to be charged up faster. This may not sound like that big of a deal, especially with the little capacitance there is, but at high frequencies, this causes mosfets to turn on and off more slowly, inducing heating and problems. Ringing on the gates is also a massive problem, as mosfets have capacitance, and there is inductance on the leads. This basically creates an LCR circuit, and can damage/kill your mosfet if it's bad enough. Gate resistors help prevent this, along with snubbers. (usually involve resistors and capacitors, and sometimes zener diodes). MOSFETS that have more current handling ability generally have more capacitance than ones that have lower current ability. This generally has to do with how big the die size is. Bigger die = more capacitance for you + more current.
Mosfets ALSO have an important factor to note; avalanching. This is the same avalanching I was talking about with TVS diodes. MOSFETS can avalanche voltage across them as excess heat. This makes these types of mosfets perfect for fly-back transformer purposes, as voltage spikes can kill a mosfet that isn't avalanche rated. Not all are avalanche rated; the datasheet will tell you if they are or not. Moving on, there are roughly two types of mosfets, P channel, and N channel, the N channel mosfet being the more common one out of the two.
Mosfets are also Positive temperature coefficient, meaning the hotter they get, the LESS current they conduct. It makes it extremely easy to parallel these devices, albeit with additional driving techniques.
MOSFETS are used in many many digital devices, due to their simple nature as on off devices. (think binary, what is it?)
Because they are used for switching, they can be used as relays, light blinkers, sound generators (like synthesizers), Switch mode power supplies, CPU logic, gate logic, oscillators, like your laptop's backlight, if it still uses CFL's that is.
MOSFETS are used in a quite a few hobbyist's circuits as well.
Step 16: N-channel Mosfet
The N-channel mosfet is the most common, and well researched mosfet there is. Compared to P-channel fets, they're cheaper, easier to use, and generally better in all terms in ability. (power wise, current ability, voltage ability, etc)
They use a positive voltage signal on their gates to trigger the switching.
N channel MOSFETS have a wide range of power, and a wide range of packages. They are available in surface mount and through hole versions. Their power handling ability ranges from under 1 amp all the way up to 100+ amps in brick versions, though these are hard to drive, due to their input capacitance and turn on/off times.
There are signal N-channel mosfets, in T0-92 packages, and power mosfets, such as the IRFP250 mosfet, a common and cheap fet used by hobbyists all over.
The basic principle behind an N-channel fet is a positive signal is applied to the gate, and it switches on. N-MOSFETS are generally used in everything, just like transistors, because they replaced what the transistor was horrible at. They're commonly found in computers, amplifiers (those expensive ones, you know?), and all sorts of hobbyist circuits.
Common N-channel fets
(all are made/created by International Rectifier, the company who actually invented the mosfet. They even own www.mosfet.com!)
Step 17: P-Channel Mosfet
P-channel mosfets work in the same manner as an N-channel fet, but instead of controlling/controlled by positive voltage, they are controlled by negative voltage signals to the gate. They are off when the voltage to the gate is +V, and on when the voltage is negative, or zero.
P-channel fets are often physically smaller, and are usually found in the TO-220 case. They can not handle as much power, or as much voltage as most N-channel mosfets. Their On state resistance is usually considerably higher than N-channel fets as well.
They're used in switch mode power supplies to keep the ground at ground. If you used an Nchannel fet, it would be above ground a certain voltage.
Common P-channel fets:
Step 18: IGBT
EYE GEE BEE TEE
Insulated Gate Bipolar Transistor
Now, you may be thinking, "Oh Jared, its just a transistor with a fancy name!" WRONG.
An insulated gate bipolar transistor, is the equivalent of taking a mosfet, and regular bipolar transistor, and shoving them together. They made babies!
And boy, do these babies pack a punch.
IGBT's are voltage controlled devices, just like MOSFET's. They combine the high current capability and the low saturation voltage (complete on) of a transistor, and the simple gate drive of a mosfet. They have three pins, just like every other device. The pins are labeled Gate, Collector, and emitter. The collector is the equivalent of the drain of a mosfet, and the emitter, the source of a mosfet.
Basically, they're pretty snazzy. But, we don't live in a perfect world.
IGBT's have their fair share of issues as well. They have a negative temperature coefficient, like transistors. Meaning, the more hot it gets, the more current it conducts, causing thermal runaway and a dead IGBT (this happens in a split second for the most part).
Compared to MOSFET's, IGBT's are somewhat slower in their turn off and turn on times. They can usually handle anything from 0 hz to around 40-50 khz, for the most part, though faster IGBT's do exist. It really depends on the actual device itself.
IGBT's are renowned for their current handling ability. Most can handle upwards of 50 amps, and more, at voltages above 600 volts! They are perfect for low frequency, high current, high voltage purposes, where mosfets are good for High frequency, lower current, and lower voltage purposes. But, like mosfets, they also have input capacitance, and ringing issues. Take the same measures you would with a mosfet to prevent this, using resistors and a clean signal source. Shorter leads also help.
IGBT's do NOT have intricate body diodes, like mosfets do. However, manufactures often do you a favor, and put a high speed high voltage diode in with the IGBT package, for ease of use. Be sure to check how fast the diode is when using the IGBT with high frequencies. But, there are exceptions, some IGBT's don't have diodes in them at all, and may require you to add one outside of the package. You need to put the anode of the diode (pick an ultrafast high amp 1000-1200 volt diode) to the emitter of the IGBT. This will allow voltages to freewheel if needed, preventing your IGBT from blowing up.
IGBT's are used in welders, induction heaters (I love these), microwave ovens, and in quite a few applications where MOSFETS have also been used. I have yet to see an IGBT amplifier, though.
There is only one type of IGBT. No N channels or P channels!
Step 19: JFETS
A JFET is like a garden hose.
Well, it is!
A JFET can be thought of a negative mosfet. The more voltage that is applied to the gate, the more resistance there is from the drain to source. (they have the same labels as mosfets!) Once a certain voltage is reached the drain to source can be pinched off completely, acting as an open switch.
They are the simplest type of transistor, and can be used as a switch, or, as previously mentioned, a voltage controlled resistance.
JFETS come in two flavors, P channel and N channel, also like mosfets!
P channel JFETS are controlled by positive voltage, and N channel, with negative. Why they decided to switch these around with MOSFETS, I will never know.
JFETS are not good for high voltage, or high current. They are perfect for volume control in stereo recievers, though ,and signal applications in computers.
They can be mostly found in TO-92 packages and smaller.
Step 20: SCR (thyristor)
ESS SEE ARRR / THY WRIST OR
Silicon Controlled Rectifier Or Thyristor
A silicon controlled rectifier is sort of like a MOSFET or IGBT. They have three leads, called the anode, cathode, and gate.
But, instead of the ability to turn on and off, they SCR's do what's called 'latching', and won't turn off again until the power is turned off to the circuit. Current flows from anode to cathode, like a diode, but has the switch, the gate.They have three leads, per usual, and come in a much wider range of packaging then previous semiconductors.
What is this good for? Well, they're great for power control, motor controls, lamp dimmers, and high voltage/high amperage current controlling!
SCR's are very very generously rated. And they mean it! A typical, small package SCR like the one in the photo can handle around 50 amps continuously, and almost 560 amps pulse, at almost 1200 volts! That's insane, especially for something so small.
SCR's can also be used for more dangerous, fun things, like Coilguns. The high current capability makes them fantastic as a mechanical switch replacement, due to the absence of wear and tear. SCR's can handle the 400 + amp pulses that a capacitor discharges. But, surprisingly, SCR's do not need heatsinks when used for pulses!
Because the pulse is so small, very very little heating actually takes place. It doesn't have a CHANCE to even heat up.
SCR's come in a wide variety of packages, ranging from little TO-92's, to bulky, beefy Brick packages and puck packages. Stud packages also exist!
Not really any "common" SCR's. Here are some various ones though.
Step 21: Triacs
A Triac is basically an SCR, except, SCR's are good for handling DC current. TRIACS are used for AC current, and are built a little differently.
Triacs can typically handle around half the amperage of the same make of SCR. A 50 amp SCR has a 25 amp Triac. This makes them unsuitable for coilguns, as SCR's do a much better job of doing so.
Triacs are useful for light dimmer circuits with AC current, (wall outlet anyone?), motor speed control (AC motors), and high voltage/High current control (again, AC).
Triacs are basically built with two SCR's put in anti-parallel. (meaning that the anode of one SCR is connected to the Cathode of the other!) They both share the same gate, meaning they are both triggered by a signal to the gate, hence the ability to handle AC currents.
Other than being able to handle AC current, Triac properties are the same as SCR's.
Step 22: Last of the Discretes
That was the last of the discrete semiconductor devices. Now, we move onto the more complicated stuff, the small, small stuff.
IC's, Voltage Regs, stuff like that.
IC's are basically combinations of everything I stated above, in a neat, little mass produced package that makes it so much easier to use, than 500 discrete components!
Voltage Regulators are a special type of IC, heatsinkable, and output a neat, fixed voltage signal using a voltage reference, sometimes a zener diode.
Step 23: Voltage Regulators
(lasers involved here, hold on)
A voltage regulator is a relatively simple device. You put direct current, in pin 1, and out comes a nicely regulated voltage out of pin 3. There are a few flavors, of voltage regulators though, and they all do essentially the same thing, but differently or for a different purpose.
There are Positive voltage regulators, and negative voltage regulators. They both come in a wide variety of packages, from TO-92's to TO-3's, and even DIP packages sometimes. They require decent heatsinking depending on how much voltage is needed to be dropped off from the incoming supply. Most voltage regulators have a certain limit to how much voltage can be input into it, and all have some sort of protection to keep itself, and the circuit it's supplying from blowing up. Voltage regulators can handle voltage inputs under 70 volts, and lower. The LM350 can only handle 32 volts input! But, there are also adjustable voltage regulators. These can be made to output different voltages, other than the predetermined voltage on a regular voltage regulator, which makes them perfect for linear power supplies. They do tend to generate a healthy amount of heat, and can use a transistor to bypass the regulator itself, and allow more current to flow then the regulator would normally accept.
Most regulators can accept 1-2 amps, with overcurrent protection. This basically means that if the current draw is over the specified amount, the regulator "shuts off" and prevents more current from flowing until the load is removed, or decreased. Some voltage regulators lack this, and can supply as many amps as you want, provided adequate heatsinking, unless you want a blob of plastic instead of a regulator.
Voltage regulators are used in alot of power supplies to make sure the voltage doesn't go over a certain point. Switch mode power supplies, linear power supplies, some chargers, all use voltage regulators to maintain a specific voltage over time. They're also perfect for powering logic circuits, and lasers! How? Well, a laser diode requires a specific amount of voltage. You can use a voltage regulator to keep the voltage at a certain point, then use either a zener diode + a resistor or just a resistor to drop off the necessary amount of voltage, and to keep the current under a certain point. Otherwise, you're going to fry your laser!
Common Voltage regulators.
LM350 (adjustable Voltage regulator)
LM317 (adjustable, higher input voltage)
LM7812 (12 volt regulator)
LM7912 (-12 volt regulator)
Note: There are a certain type of voltage regulator, called switching regulators. These are generally a little more complex, as they require more external components (usually) but they give the benefit of much increased efficiency, so basically, less heat, and less power wasted. They use what's called a buck converter.
Step 24: Integrated Circuits
Integrated circuits can not possibly be explained in 4000 words or less.
I'll try my best to shorten it up, so that you won't be sitting here for hours reading this.
An integrated circuit is a bunch of extremely small components put together in a package, usually DIP's for human hands, and surface mount packages like SOIC or TSSOP packages, which are extremely small.
Integrated circuits do practically EVERYTHING. They make your mouse work. Your computer work. Your computer's power supply to work, your speakers, your cell phone, your monitor, your keyboard, your mouse, your ipod, your ipod's charger, your printer, your scanner, pacemaker, TV, just about EVERY electronic device has these inside of them, because they do so many things!
They come in so many packages, flavors, types, it is quite impossible to explain them all!
Here are a few small ones, that are used often for hobbyists.
555 Timer. The 555 timer is a simple oscillation circuit used to generate square waves. This chip comes in an 8 pin DIP package and can be bought at radioshack, or pretty much any electronic supplier. This chip is the most widely made chip on the planet. There are millions of these chips made everyday, because they can do so much. For a hobbyist, they can be used to make a flyback transformer driver, a nifty little circuit that's fantastic for beginners to learn about AC current, and transformers.
Op Amps are a special type of Ic that can amplify something thousands of times. They are perfect for many things, and can be used for oscillators, schmitt triggers (which can turn noisy signals into nice clean signals), feedback mechanisms, voltage regulators (they're inside of them!) and audio amplifiers. They can take a very small signal, such from say, a crystal radio, and turn it into something suitable for driving regular headphones. They have two inputs. Inverting, and non-inverting. They can output both positive and negative voltages. They're also good for voltage comparing, which puts a voltage through the output, depending on which one is greater. A good example would be 5 volts on the non inverting side, and 10 volts on the inverting side. The 10 volt side would be output, unless the 5 volts increased to 15, then it would output 15 volts.
Gate Drive IC's are a type of integrated circuit that take a signal, usually and most often a square wave, and turn it into a more powerful signal with more current. This is great for driving MOSFETS and IGBT's at high speed, because they allow the gate capacitance to be charged up much more quickly. There exist low side, high side, inverting, and non inverting drivers. Low side drivers are the easiest to use, and thus the most widely used! They do exactly what was said above. High side drivers often exist with low side drivers and can be used to drive a half bridge, or two mosfets/IGBT's in series. This series modification allows the mosfets to effectively share the load, and reduces heating in each FET/IGBT. High side driving isolates the signal from the signal source, preventing back EMF or voltage spikes from killing your control circuitry. But, this comes at a price, and high side drivers often fail, and are a bit harder to use, requiring more components and time. Inverting drivers take the signal from the signal source, and flip it. If there was a positive part of a square wave, they turn it neutral, and if there was a neutral part, they turn it positive. This allows two low side drivers to be used from one signal source, and lets you use a half bridge without a half bridge ic. (they use high side driving).
There also exists logic gates. These are AND, NAND, OR, NOR, XNOR, NOT, and XOR gates.
Each performs a specific logical function. To understand logic gates, you need to understand binary. 1 = On, or + v , 0 = off, or 0 v.
Now that you understand binary, you can understand logic gates! An AND gate has any number of inputs, and one output. One output is universal across all gates. For an AND gate, all the inputs need to be high, for the output to be high.
For a NAND gate, you invert this. All inputs need to be high, for the output to be low.
For an OR gate, If any input is high, the output is high.
For a NOR gate, if any input is high, the output is low, otherwise it will be high.
an XOR gate will be high if an odd number of inputs are high. I.E. 01 in binary will be a high, while 101 will be a low. (each binary number is an input)
An XNOR gate does the opposite, and will be a low if an odd number of inputs are high.
A NOT gate does an inverting function, it flips the input signal from a 1 to a 0, or a 0 to a 1.
There are several different logic types, CMOS, TTL, nMOS, RTL, DTL, Ternary, and ECL. Each one has logical functions, but use a different type of technology to achieve it. CMOS uses 3.3 V to indicate a high, and uses Complementary Metal Oxide semiconductors to do so. (complementary N and P channel mosfets)
TTL is transistor transistor logic. It uses +5 volts for a high.
nMOS uses N-type metal oxide semiconductors. (only N channel mosfets)
RTL is resistor-transistor logic. +3.5 volts is considered a high.
DTL is Diode transistor logic. +3.5 volts is also considered a high.
Ternary logic uses unknown, and true falses instead of high and lows.
ECL logic is emitter coupled logic, and uses Bipolar transistors. +3.9 volts is a high.
All of these gates make up CPU's and microprocessors inside of your computer, cell phone, ipod's, cd players, etc.
They are the basic of basic logic functions.
Step 25: Vocabulary Section
Reverse Biased - This means that voltage is flowing from cathode to anode, or trying to flow through. This often results in avalanche breakdown, or in zener diodes, the zener effect.
Forward Biased - This means that the voltage is flowing from anode to cathode. This will result in a voltage drop across the diode.
Biased - Bipolar junction transistors use this to prevent a full turn on in amplifiers. If there is a voltage that exceeds the saturation voltage of a transistor, the transistor will not amplify the signal, and as long as that said voltage is high, the transistor will continue to act as an open switch. Resistors are often used to limit current, and thus keep the transistor at and below saturation if used for amplifying purposes.
Avalanche - see zener effect.
Zener Effect - The zener effect and avalanching are essentially the same thing; a diode, or a transistor (meaning mosfets and igbt's too!) is reverse biased, and a breakdown voltage occurs. This results in a very specific amount of voltage being dropped off of the semiconductor, and is controlled in zener diodes, and in some transistors. This effect is useful for circuit protection, voltage regulation, voltage references, and as well as flyback topologies to prevent back EMF.
EMF - Electromagnetic field. This can be created by a variety of things. Mostly high voltage, high current, radio frequencies, and inductance. When an inductor (a coil of wire in many cases) is shut off from a current source, the inductor tries to keep current flowing. This results in a sometimes massive spike of voltage and current, and can very easily damage sensitive electronics. Radio frequencies, high voltage, and high current EMF's are a result of induced charges into wires, or noise, and like with inductance spikes, they too can easily damage sensitive IC's if not properly handled. A snubber is an effective EMF killer. Avalanching and the zener effect also come into play into some snubbers, and protect sensitive mosfet gates from overvoltage.
Snubber circuit - A snubber circuit can consist of several things. The more simpler ones contain only resistors, and capacitors. They effectively absorb the high voltage spikes, and release them into ground. This however can cause problems, such as ringing and occasionally back current release to the mosfet gates, causing heat. Zener dioes and TVS diodes are also used in some cases to keep voltage levels at a reasonable level. Snubber circuits are also used in MOSFET gate drive techniques, to prevent ringing on the gates.
Ringing - Ringing is the result of gate capacitance, and inductance of the wires. This can cause undesirable effects in transistors; such as blowing up, melting, and excessive heating. Easy way to get rid of this? Use snubbers, ferrite beads, and shorter lead lengths.
Negative Resistance - As voltage increases, current DECREASES. Shown in tunnel diodes, as well as fluorescent lights; that's the reason fluorescent tubes require ballasts.
Voltage Drop - A voltage drop is a specific amount of voltage essentially lost over a device or thing in a circuit. Zener diodes have a specific controlled voltage drop.
Analog Electronics - electronic devices that work on many voltage levels. There is no specific on or off, mostly inbetween. A sine wave is a perfect example of analog.
Digital electronics - Electronic devices that work on two voltage levels, on and off. There is no inbetween, and a good example would be a square wave.
Square Wave - A series of pulses whose voltage is either on or off in specific times called Duty Cycles. The duty cycle can range from 0% (fully off) to 100% (to always on).
Sine Wave - A sine wave is a type of wave that varies widely over a negative and positive voltage level. This type of wave has the least amount of EMI and generally is the most "beautiful" out of all other types of waveforms, due to their smoothness. They can be created a wide variety of ways, LC tank circuits, generators, and digital sine waves. (explained in a moment).
Digital Sine Wave - A digital sine wave is a very precise series of voltage changes to create a sinusoidal pattern out of on's and off's.
Triangle Wave - Similar to a sine wave, but instead of smooth ramping, it's more harsh and sharp like a square wave. They aren't used for much, but in LC inverters are a sign of off resonance.
Sawtooth wave - A sawtooth wave is a very sudden increase and decrease in voltage. They are called sawtooth waves because the pattern of the wave is much like a saw blade. They can be found in certain flyback topologies, and in CRT tube flybacks. Characterized by voltage spikes and sharp corners.
EMI - Electromagnetic Interference, similar to EMF.
ESD - Electrostatic discharge. This is caused by simple static electricity we all know and use, and can kill very overly sensitive electronic components. Use an anti-static wrist strap and always make sure you're properly grounded before touching any sensitive devices.
Mains Voltage - The Mains voltage is a fancy way of saying wall outlet power. Depending on where you live, it can be 120 volts AC, or 220 v AC, both at 60 Hz, or in some countries 50 hz, at a sine wave. This is lethal! Be careful when working with main's voltage, the current wall outlets can provide is enough to hurt you, or kill you. Respect it.
Offline Power Supply - Off the line power supply. Basically means you plug it into the wall to make power.
LED - Light emitting diode, for those who didn't bother to read that section. :)
DIP - Dual Inline Package
SIP -Single Inline Package
P N P - Positive negative Positive. This is the vocabulary used to describe a transistor, and the way the silicon is doped properly.
N P N - Negative Positive Negative. See P N P.
Doped Silicon - No, your semiconductors aren't on crack. The process of doping means to make the silicon not as pure, and makes the silicon function more to our liking. More semiconduct-y rather than just a hunk of shiny metal. P = Positive junction, while N = Negative junction.
Source - The "ground" or output of a mosfet, or JFET.
Drain - the Input of the Mosfet, or JFET. (not to be confused with gate )
Gate - The input signal of a mosfet, IGBT, JFET, Triac, or SCR. They all require a voltage and not current to turn on.
Base - The Input signal area of a Bipolar junction transistor. Requires current to turn on. (not to be confused with emitter )
Emitter - The output of a Bipolar transistor or an IGBT. Usually connected to ground, though in some cases shouldn't be.
Collector - the input of a Bipolar Transistor or an IGBT. This is where the higher power usually comes into play.
Pull down resistor - This allows the gate capacitance of an IGBT/MOSFET to drain away allowing for much faster turn off speeds.
Gate Capacitance - The amount of capacitance a MOSFET or IGBT has prevents the device from turning on fast. MOSFETS generally have higher capacitance than IGBT's. Use pull down resistors, or a negative voltage swing on your gate driving technique.
More Vocab to come?
Step 26: The End of This Guide (and Applications for You!)
So, learning all of this, you're probably wondering what exactly I can do with these things?
For hobbyists, you can search ALL over for ANYTHING to build. Want a transistor amplifier? Look up Transistor amplifier on google!
Want a flyback transformer driver? Look that up, or use the one I have in the pictures!
Want a simple power supply? Look up "LM317 schematic" and you'll get really simple, and easy to read schematics on how to make a 1 amp supply using an LM317.
Radio transmitter? Look up a schematic, crystal oscillators in cans are easy to use and they have all sorts of different ones available for choosing, frequency wise.
More intense stuff includes Telsa Coils, X-ray machines, coilguns, induction heaters, can crushers... These are extremely dangerous, and deal with High voltage, high current, or combinations of both along with high temperatures and risk of eye injury. Use safety glasses. I'm not responsible for anything STUPID you do with this information. Look up and research how these things work; not only does it help diagnose problems, but it also makes you smarter safety wise.
But, generally, when looking for a schematic, try to stick with something that's been PROVEN to work. Otherwise, you may get stuck with something that an idiot slapped together for the heck of it.
Some GREAT websites for electronic enthusiasts:
www.4hv.org (Don't post any Pseudoscience or "Free energy" crap there. ;)
www.hackaday.com (this website is full of neat little electronic things that catch the writers eye. Good for finding cool stuff to build)
There are several other single user websites, created for the purpose of showing off something CRAZY they designed. Look for them on www.4hv.org. You can find me there by the name of Inducktion, just like on here.
This guide will have hopefully explained a LOT about all of those little black things. You can look up part numbers, and find out just what something does inside of something, if you want to waste your time of course.
One big thing about all of this is, use the datasheet! USE THE DATASHEET.
USE THE DATASHEET.
You'll also probably ask where you can get these things if you can't take apart stuff, or need a specific item.
Order stuff online.
www.craigslist.com (good for finding stuff for free!)
Most information on this guide is universal, but every semiconductor is different, and some manufactures like to be weird and change pin-outs on you. So, always check the datasheet for everything about the semiconductor you're using, so you know the ratings, the pin-outs, and max power dissipation so you don't completely fry your poor little silicon.
But, learning all of this isn't just information, it's the ability to create something new. If you're interested in electronics, this guide I've given you may make you create or ideate something amazing, something no one else has created before! But, circuits are also hard to create, and that's why most of the easy ones have already been discovered and used. I don't mean to crush any hopes there, but it's the cold truth.
However, the electronics field is growing, everyday, and yet, there are so many job openings in it even in the midst of this recession we're in.
The reason for this, is not a whole lot of people are really interesting in how electrons flow. Why? Who knows, probably because it's complicated, but electronics has a learning curve. That's why most people drop out of electronics class in college, and in high school if your school offers it. And, electronics isn't exactly the most exciting thing in the world, but it really depends on what field of electronics you go in.
If you ARE interested in electronics, great job. Keep it up, because these jobs pay the big bucks, as my teacher always says. There's a reason why we make it so complicated. ;D
Ah, well, if there is anything else you don't understand, or maybe need help with, you can message me on here, or on 4hv.org. Either way, I'll try my best to give you assistance.
Every picture within this guide (except for picture numero uno, and the circuit pictures below) I went through the painstaking process of taking, or making, using paint. I did use some help here and there from friends and from a fancy little website called 4hv.com. I appreciate it a lot guys. :)
Also, I really hope you guys/girls (not sexist ;D ) out there learned a bit from this guide. And, for you super nerds out there, feel free to drop me a PM if there's something wrong or I'm mistaken somewhere in the guide, or left something critical out. Thank you for reading!