As active devices go, it doesn’t get much simpler than a diode. Two terminals. Current flows in one direction and not in the other. Simple, right? Well, then there are examples with useful side effects like light emitting diodes. [GreatScott] points out that there are other useful diodes and, in particular, he posted a video covering Schottky and Zener diodes.
These special diodes have particular purposes. A Schottky diode has a very low voltage drop and fast switching speed. Zener diodes have application in simple voltage regulation.
If you ever wondered, these exotic diode names because of their inventors. [Walter Schottky] was a German and although some people call it a hot carrier diode, most people use the inventor’s name. We don’t know of another name for [Clarence Melvin Zener]’s invention.
If you wonder why you care about high speed in a diode, [GreatScott] has a good demo of rectifying a signal with a regular diode and a Schottky. At a certain frequency, the normal diode starts conducting when it should be off at a certain frequency. The Schottky diode is able to turn off faster, so it can handle a much higher frequency.
There are other exotic diodes including PIN diodes, Gunn diodes and more. After the apocalypse, you might want to try making your own with sodium bicarb. Oddly enough, we covered [Afroman]’s video last year that covered similar topics, if you want a second point of view.
It always bothered me that diodes are considered active components. The definition that I learned was that active components can change their behavior based on an outside signal, and passives just do what they do. You can control whether a transistor passes current via an electrical signal, but a resistor always resists.
A diode has more complex characteristics than a resistor, but they’re still set. You can’t send the diode a signal and have it switch polarity or whatever. So wouldn’t that make it much more similar to the passives group than the actives?
Hmm… A common diode changes its dropout voltage according to temperature! Temperature is an outside signal :)
I’ve grown with different definition, by the way: active components are the ones with nonlinear characteristics.
Wouldn’t also resistor then change it’s resistance according to temperature? :)
Haha, it will :) But to a much lesser extent. Maybe we need to get rid of “passive” thing and divide components into “very active” and “not-so-active”?
Damn, I thought about it a little and realized that a capacitor will fall under this definition as well :/
Maybe this has something to do with history? Tube diodes required external power to work (so you can certainly call them active), and ordinary ones might have fallen into the same category just by analogy.
I thought again and I guess I found the reason:
C=U/Q
R=U/I
L=Phi/I
There are no such nice-looking linear expressions for dioide.
I like your train of thought, but here is a simple explanation of why diodes are considered active components:
Think of the diode as a resistor that changes resistance according to its bias voltage. If you apply a large positive bias voltage to the diode, its resistance drops very low (almost 0 Ohm) and conducts current freely (aka, the diode is “switched on”). On the other hand, if a small or negative bias voltage is applied to the diode, then its resistance stays very high, causing it to resist the flow of current (aka, the diode is “switched off”).
I think it’s the nonlinearity, really; you can, for instance, feed a diode a pair of signals through a resistor divider, and get behavior that’s not just the sum of the behavior of them separately. You could even make one of them be a signal and one of them be a control, and make do without a transistor that way. You can’t do that with linear passives, the response of a pile of linear bits is always the sum of the response to each input separately.
You could do what youe mention, with RF AC, to such an extent you could make logic gates and switches out of diodes. To such an extent you could build a digital clock without a SINGLE transistor! Just using diodes for all the switching, using AC for power, and switching the bias to control it as you mention. Then you could probably get it submitted to HAD, and quite possibly win a prize.
Whether you do or don’t win, though, might not be universally remembered. It was a few years ago. Great idea at the time, couldn’t figure out how it wasn’t impossible til I read the maker’s report on it all.
heh, yep. Should also be reasonable(ish) to get an analog amplifier (albeit the way I can see you need a current signal in and out) by using the exponential region of the diode to take a logarithm, adding that and a bias with a resistor network, and feeding that through a second diode to go back to current with an exponential. Same math as a slide rule.
I have found a home made digital clock that used the 60hz as a reference but every time a small spike caused by turning on a light on the same circuit the clock was on caused it to malfunction.
https://hackaday.io/project/11677-the-diode-clock
PIN diodes are used to switch RF signals (controlled with a DC bias) all the time – even common rectifier diodes (e.g. 1N4007) can be used in this way – at higher frequencies, any diode with a finite reverse recovery time would work. We just usually ignore this effect. The “ideal” diode would probably be classified as a passive device, but I never find any in stock…
You could even switch with an ideal diode – “on” is just always biased to forward, and “off” is always reverse. The time delay isn’t required to get that nonlinear action.
Only slightly less ideal diodes even have a nice essentially-exponential region you can do log and exponent with.
A better definition is: a device that can change their behavior based on a signal. It doesn’t matter if it is an outside or inside signal. A diode changes its voltage drop based on the direction of current that has flowed through it. Drive current from the anode to the cathode, and the voltage drop will be very low. Drive current from the cathode to the anode, and the voltage drop will grow by an order of magnitude, or two.
The distinction between ‘passive’ and ‘active’ is easiest to break down in terms of network theory: a ‘terminal’ is a point where you can connect to a circuit (like the leads of a resistor), and a ‘port’ is a pair of terminals where AC current can flow. You can measure the current that flows through a port, and you can measure the voltage between its terminals.
A passive component has one port and a well-defined relationship between voltage and current: V=IR, dv/dt=I/C, di/dt=V/L, etc. If you know the current, you have all the information you need to calculate the voltage, and vice versa.
An active component has two or more ports whose current-to-voltage relationships are linked: the current through a mosfet’s drain-source port depends on the voltage across its gate-source port. You can’t describe the drain current without taking the gate voltage into account.
That matters because circuits connected to a device port can’t see through the port to know what’s on the other side. An LED controlled by a mosfet’s drain-source port doesn’t know there’s a gate-source voltage on the other side. All the LED sees is a port whose current changes independently of anything the LED is doing.
In those terms, a diode is passive: it has two terminals, one port, and a well-defined current-to-voltage relationship. It makes sense to say that the forward voltage of a Schottky diode is lower than the forward voltage of a silicon diode for the same current, and that the forward voltage of a red LED is usually about 1.7v
All components have temperature coefficients, but those only add a free variable to the equations.. V=I(R+rT) They don’t change the number of ports, or the fact that you can calculate V, I, or T of a passive if you know the other two.
The reason diodes aren’t grouped with resistors, capacitors, and inductors is that they’re ‘non-reciprocal’: the current-to-voltage relationship depends on which way the current is flowing. If you put 1v across the terminals of a 1n4007, the current might be about 1A or it might be about 1uA.
Non-reciprocal components break the math used to describe multi-port networks, so diodes get pushed off in their own category.
I learned that active component are just semiconductors. Which includes Diodes.
Also a diode does change behaviour depending on the input. Does it not block a signal when in reverse? Do zener diodes not pass through a signal when it is above their set rating? A transistor has 2 inputs, a diode has just 1 which gives a false sense of being a passive, but it is still a active component.
How about, “An active component requires power for normal operation”… and we don’t care if you bias the diode, for any reason.
Where’s the hack
It’s right there, under the 555.
“Another” name for the Zener is the “avalanche diode”. Which is actually the correct name. Most so-called Zeners don’t work on the Zener principle, so they’re just avalanche diodes, not Zeners.
The why and how of all this, ask someone else. I just know the factoid version.
Zener effect happens at voltages below ≈5V. Avalanche effect happens at voltages above ≈5V.
5.1V diodes are neat because one effect has a positive temperature coefficient, the other has a negative temperature coefficient, and 5.1V diodes are right in the middle, so they have a minimal net temperature efficient.
This barely scratches te surface of diodes.
Solar cells are also “just” diodes.
So are varicaps
https://en.wikipedia.org/wiki/Varicap
And if you’d ever dare call a Gunn diode a passive device your’re only confiming you’re an idiot.
https://en.wikipedia.org/wiki/Gunn_diode
Heh. I’m glad I’m not the only one who sees the similarities.
Can think of multiple uncommon uses of diodes. Got an outside thermometer probe on a wire? Likely the sensor is one or several small signal diodes. Want to drastically improve reception of an old receiver? Try swapping out the 1N34 detector diode (1VFD) for a 1N60p (0.25 VFD). Need a broad noise source? Diode!