I don't understand why do we care about voltage and current supplied to any components, why not just the power, no matter how much are the current and voltage?

For example: a red LED needs 2V to function, but why? Why don't we say the LED function at - just assuming - 10 W, whatever the combination of voltage and current?

  • 10
    \$\begingroup\$ Why do we care about velocity and time? Why not just distance? For example I need a car. But why? Why just not say I need to travel 100 miles? \$\endgroup\$ – Eugene Sh. Aug 10 '18 at 16:05
  • \$\begingroup\$ Because most things that draw power "care". Systems that can accept a consistent amount of power across varying voltage/current ratios by means of switching conversion are relatively new, and while now (for limited ranges) compactly practical still have cost and inefficiency. \$\endgroup\$ – Chris Stratton Aug 10 '18 at 16:08
  • 1
    \$\begingroup\$ Let someone shuffle their feet across a wool rug and then touch the tip of your nose. Do you care? There's almost no power, but a very high voltage. \$\endgroup\$ – Elliot Alderson Aug 10 '18 at 16:56
  • \$\begingroup\$ I only care about voltage and current because the devices I use care about voltage and current. If we had devices that didn't care, I wouldn't care either. Why do the devices care? That is a big question. It is just the nature of the devices. Or you could say it is baked in to the fabric of the universe. \$\endgroup\$ – mkeith Aug 11 '18 at 5:31
  • \$\begingroup\$ Red LED's need 2V because 2 electron volts is the energy required to produce a red photon. If you don't have at least two volts, you won't get any red photons. Can't get them from a bandgap with a lower voltage. \$\endgroup\$ – mkeith Aug 11 '18 at 5:37

There is an interesting thought in this question: knowing the power consumption is enough to determine voltage and current.

There is indeed only one solution at a given power consumption for discrete components. A 1\$\Omega\$ resistor will consume 1W when it is running at 1V and 1A. For a diode, the operating point can be found on the IV curves even if it is not straightforward.

For a capacitor we can also determine what voltage it is at for a given energy storage which would be equal to \$P.t\$ where t is time in seconds, and the instantaneous current can be calculated as \$P/V\$.

Using power for analysing circuit behavior can be a powerfull technique (no pun intended). It can help you find the amplitude of an output signal faster than more complex computations.

However, it is not very handy to use power as the unique or even the main specification of these components. For a diode for instance, the voltage level hardly changes for a big current range because the VI curve is almost vertical after the 1.x V knick. And also because the component can operator at many power levels, and currents, but mostly at a "single" voltage level. So for a diode, the operating voltage is a major characteristic.

Once we know the voltage we still have to limit the power consumption. If I say that the LED should operate at 30mW is that a convenient information to work with? No! When I have a known drive voltage, I want to determine the resistor value to determine the set point for the LED. The LED power does not allow me to determine it directly. Voltage and current do! I just have to know how much voltage the resistor needs to handle and how much current I want. If my source is 5V, my LED operates at 2V, I have 3V over the resistor, so knowing that I want about 20mA, I know the resistor's value by doing 3x50= 150\$\Omega\$ (because I know that 50 = 1000/20).

If I were using power, I'ld have 40mW, so I'ld need to convert that to voltage and current, or at least know the voltage. With voltage and power alone, I would use the fact that the resistor consumes 1.5 times the power of the LED because it has 1.5 times the voltage drop. So it consumes 60mW. \$P=V.I=V^2/R\$ so \$R=V^2/P\$ meaning that \$R=9V^2/60mW=9\Omega/0.06=3\Omega/0.02=150\Omega\$. Wow, the same result, but a bit harder to come by.

So rather than computing the interesting characteristics each time, we memorize them or communicate them. It speeds up our work, and keeps us concentrated on the important stuff. Power consumption becomes interesting to know too because in the our world of connected devices, determinining the power budget is important too, but is is less important if you just want the circuit to work.


For example: a red LED needs 2V to function, but why? why don't we say the LED function at - just assuming - 10 W, whatever the combination of voltage and current?

Because you can achieve 10 W in an infinite number of voltage and current combinations. e.g., from \$ P = VI \$ we can see that we could generate 10 W using

  • 1 V x 10 A,
  • 10 V x 1 A,
  • 100 V x 0.1 A,
  • 1000 V x 0.01 A.

enter image description here

Figure 1. A 55 W headlamp bulb - but for what voltage supply?

In designing a device we have to make some decisions. Let's take the example of a 55 W auto headlamp bulb: somebody has decided that the standard battery voltage is now 12 V for cars and 24 V for trucks and buses. That means that the 12 V bulb will require 4.6 A and the 24 V lamp will require 2.3 A.

If you are designing a product from scratch then you have many options.

The LED example in your question is rather interesting in that the forward voltage, VF, is determined by the semiconductor used to generate the light.

enter image description here

Figure 2. The IV curves for typical small red, orange and green LEDs. Source: LEDnique - parallel LEDs.

Note that the curves in Figure 2 are a fixed property of the material used to dope the semiconductor to get the desired colour. You can't change it.

From the comments:

... why normal diodes need 0.7V to work?

Because that's the voltage required to enable charge carriers to jump the depletion zone in the P-N diode junction.

A p–n junction diode is made of a crystal of semiconductor, usually silicon, but germanium and gallium arsenide are also used. Impurities are added to it to create a region on one side that contains negative charge carriers (electrons), called an n-type semiconductor, and a region on the other side that contains positive charge carriers (holes), called a p-type semiconductor. When the n-type and p-type materials are attached together, a momentary flow of electrons occur from the n to the p side resulting in a third region between the two where no charge carriers are present. This region is called the depletion region because there are no charge carriers (neither electrons nor holes) in it. The diode's terminals are attached to the n-type and p-type regions. The boundary between these two regions, called a p–n junction, is where the action of the diode takes place. When a sufficiently higher electrical potential is applied to the P side (the anode) than to the N side (the cathode), it allows electrons to flow through the depletion region from the N-type side to the P-type side. The junction does not allow the flow of electrons in the opposite direction when the potential is applied in reverse, creating, in a sense, an electrical check valve. Source: Wikipedia Diode [emphasis mine].

0.7 V is sufficiently higher.

  • \$\begingroup\$ Actually I meant to choose the LED because I know it's basically a diode, again back to my question, why normal diodes need 0.7V to work? I mean can't a high current substitute the low voltage ? \$\endgroup\$ – Mohamed Abduljawad Aug 10 '18 at 19:25
  • \$\begingroup\$ See the update. \$\endgroup\$ – Transistor Aug 10 '18 at 19:34
  • \$\begingroup\$ @Transistor Could you maybe add that different components are more sensitive specifically to Voltage and current? Especially for things like capacitors, diodes, transistors, transformers, it is necessary to view voltage and current separately because especially with voltage, exceeding ratings even momentarily can cause component failure, while remaining well below rated power. Other than the power used in the explosive deconstruction of components, anyway. \$\endgroup\$ – K H Aug 10 '18 at 22:10

If you are provided with power rating only, you can achieve that with infinite combinations as 'transistor' mentioned. But the electrical components have certain limitation to the voltage applied between them. Apply higher or lower voltage than mentioned, you end up getting bad result. But the thing is different for current. The components take as much current as it need, no matter how much is supplied. Say you have 10W bulb,Then you can try the combination of 5V*2A or 20V*0.5A and many others. As I mentioned earlier, the voltage applied must be fixed to certain level. Lets say the bulb has 10V mentioned on it. So if you apply 5V or 20V you wont get the result you are looking for. So the bulb needs to be operated at 10V*1A (since power is 10W). And you need to choose power supply rated 10V, 1A. This reading lets you know that the supply is capable of maintaining 10V at its terminals and 'at least 1A current.' In case of LED the voltage is fairly constant for certain range of current.


Voltage and current are fundamentally different things. They are not just different aspects of power. Is it possible to have a voltage difference without a current, and (in a superconductor) a current without a voltage difference. As others have said, electrical components constrain the voltage-current relationship, but that doesn't mean we can only care about power. For one thing, it's a lot easier to make a circuit that provides a fixed voltage or current than a fixed power. Measuring the voltage and current is also the easiest way to measure power.

Electricity can carry Information as well as power. For information, you don't need both voltage and current, and you usually try to minimize one of them. Consider a digital logic circuit -- how would looking at power be better than looking at voltage?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.