Do current sources exist practically or is it just a concept? If there exist a practical current source, can you give some examples.
An ideal current source is a concept, and does not exist, cannot exist. Other useful concepts like an ideal resistor, an ideal voltage source, a 'light inextensible string' or 'frictionless surface' (I loved those two in physics lessons) don't and cannot exist either.
However, not only are these concepts useful mathematically, but they are also very useful as a first order approximation to the behaviour of many real physical things, and can be approximated to a greater or lesser degree of completeness by real physical things.
So we can approximate a current source by a transistor, with the base held at some potential which keeps a voltage across, and therefore a current through, an emitter resistor, which current appears at the collector. This current source has a non-infinite output impedance, varies with temperature, and cannot deliver infinite voltage into an open circuit, but is still useful, in certain applications.
An inductor, with a current flowing through it, at the time the switch controlling current to it opens, is best thought of as a current source. How to cope with the continuous current still coming from it, with the energy behind it to raise the output voltage to any high voltage sufficient to cause breakdown, is a good mental model for how to design the current paths and control components that are there to limit the voltage rise.
When a switch mode converter is controlled to vary its output, 'constant voltage' and 'constant current' are the two main models for what the control inside tries to do at the output terminals.
In the same way, real voltage sources cannot deliver infinite current, real resistors have some residual series inductance, and the only use for a light inextensible string is for tethering a spherical cow.
Of course they exist.
There are many cases, where current sources are applicable:
- LED Lighting
- Microelectronics (such as amplifiers, measurement devices, ...)
- Motor controllers
Often, they are realized by a control circuit (including an operational amplifier) that monitors the current using a shunt resistor, controlling it using a transistor.
You always want a current source, when your "output" is somewhat proportional to the current (motor torque, light, loudness)...
If you mean the dual of a battery, yes some such devices exist. Photovoltaic cells and betavoltaic cells are examples of devices that produce more-or-less constant current up to some maximum open-circuit voltage, just as a battery produces more-or-less constant voltage up to some maximum short-circuit current.
That said, in most electronics, current sources and sinks are artificially produced by manipulating energy from a relatively constant voltage source. A JFET or depletion-mode MOSFET plus a resistor or a simple combination of BJTs such as a Widlar current source are examples of such arrangements that can produce a wide range of currents.
Define "exist". Of course, the perfect current source does not exist, because that would require it to produce infinitely high voltage in cases where the effective resistance across its terminal approaches infinity (i.e. if the source is left "unconnected"); infinite voltage doesn't exist. But the same goes for the perfect voltage source: no voltage source in this world will produce infinite current if shorted. In the end, it boils down to conservation of energy.
Now, for "practical" current sources: Of course they exist! They're a very important utility in very many (if not most) analog semiconductor circuits.
For example, if you did this:
You can transform a fixed voltage source to a fixed current source (by the way, this small circuit with the two transistors is called a "current mirror" and is really, really common in electronics). You'll find this in a lot of circuits – imagine you want to drive a series of LEDs with exactly 20mA, but you know that LEDs aren't perfect, so no two LEDs have exactly the same forward voltage, so driving them with a constant voltage and a series resistor isn't as good as actually driving them with a constant current. Or, for example, to imprint a specific current into a bigger transistor's base, for specific amplification usage.
Elementary, current sources are quite common in physics. Think about an LED: by pushing in electrons with a specific energy ("energy per charge" is voltage), you can fill one "hole" at lower energy levels in your semiconductor, emitting the energy difference of the electron before entering the semiconductor interface and after recombination as photon. So you get a fixed relation between "number of electrons per second" (==current) and "photons produced per second" (==brightness). The same works the other way around, too: If you shine a light on an LED, the amount of current it produces has a fixed relationship to the number of photons hitting the semiconductor "just right". Hence, with fixed brightness, you get fixed current – a real-world current source (in fact, if you optimize for efficiency at absorbing photons rather than emitting them, you get what is known as a photodiode).
I don`t think so. For my opinion, each current needs a driving voltage (perhaps, there are some "exotic" exceptions where we have a movement of charges - current - without an applied voltage).
However, in all of our classical electronic applications we realize devices which we call "current source" based on voltage sources, which allow a current which is - more or less - independent on the applied load (using a suitable control mechanism, regulation loop).
In particular, this is true for a continuous flow of charges (current).
EDIT: For my opinion, the "crux" is in the word "exist". I have tried to explain my view: Yes - there are electronic devices which we call "current source" because they can "deliver" a nearly constant current - independent on the connected load (within certain limits, of course). But - is the word "deliver" correct? Do these sources really deliver a current ? I think, they rather allow a current (which is driven - as each continuous current - by a voltage). And that is the reason that - from the physical point of view - we really do not have a "current source". The current of this source is determined by the driving voltage and a corresponding very large source resistance.
What about lightning ? It comes from megavolts so the peak current is essentially the same whether there is a short to ground or to + or - 1Kv .The output pulses from some photomultiplier tubes can be looked at this way because the peak output voltage is proportional to the load impedence over a significant range .A permanent magnet DC generator would make a reasonable constant current source if it were driven with a constant torque source.