0
\$\begingroup\$

If you have an AC circuit and need to reduce the power going to your load, what is the difference between commonly-known ways of achieving that?

Because most of my electrical experience is related to houses in the USA, I'm thinking of like 120V, 20A circuits with a breaker run through copper wire, where some household appliance is the load.

I am asking because in my beginner-level study of electronics, I have already heard multiple times how using a resistor would be "inefficient" or "no power savings", but without further explanation. I have also heard about how a longer wire would increase resistance too, but never heard of it as being "inefficient" for some reason. Note that much of this is what I've learned from electricians or the internet, not academically.

My understanding is that using a resistor or longer wire to reduce the power would be roughly equivalent in terms of efficiency, as some of the reduction in power will be lost as heat in either method. So, if I had a circuit where the load used 100W and then added in a resistor or made the wire longer such that the load used 50W, the entire circuit would be using a bit more than 50W because of the resistor or longer wire. But my guess is that in practice using a resistor to lower the power (eg, a dimmer lightbulb or ceiling fan) is still a power savings over not having the resistor.

Another guess based on experience is that one of the more efficient ways of reducing power is to just use a transformer to lower the voltage and have your appliance designed to operate at that voltage.

So, I do not why any of these methods of reducing power is significantly different from the other, and why I have heard so much about how a resistor specifically is "inefficient" or would result in the same power being consumed by the circuit as a whole.

\$\endgroup\$

3 Answers 3

1
\$\begingroup\$

This very much depends on the load you are trying to power; this is not as simple as putting a brick in your toilet tank to reduce water consumption. Even an incandescent lightbulb cannot be modeled as a simple resistor since its resistance strongly depends on its temperature, which depends on the current flowing through it.

However, let's take an example of a 10 ohm resistive load with no temperature coefficient. With a RMS input voltage of 120 V, it normally consumes 120^2/10 = 1440 W of power. If you put a 5 ohm resistor in series with it, the combination now consumes 120^2/15 = 960 W, which is indeed less. The current is 8 A, so 8^2 * 10 = 640 W are consumed by the load and 8^2 * 5 = 320 W are consumed by the resistor. Total power consumed has decreased, but now 1/3 of the energy is turning into heat. This is what people mean when they say it is a low-efficiency method of reducing power consumption; before, 100% of the input power was doing something useful; now, unless "creating heat" is something useful, only 67% is. You now also have a several-hundred degree resistor to set your house on fire with.

Many consumer appliances these days use switchmode power supplies with universal input, meaning that as input voltage drops (because you put, say, a resistor in series with it), they will simply compensate by drawing more current until the required power input is achieved. In the meantime, your resistor is dissipating potentially hundreds of watts for no reason. This is probably what people mean when they say "no power savings". If you increase the resistance until the power supply cannot draw enough current to maintain its output voltage, the power supply will shut down or blow its input fuse. Same deal with lowering the input voltage with a transformer, except you are only losing a couple % in transformer losses instead of 10s of % in resistive heating. If you want to reduce power consumption, you must change the load instead of trying to restrict the input.

\$\endgroup\$
2
  • \$\begingroup\$ thank you for walking through that example. i can see how a resistor is really inefficient even if there is an overall reduction in power. do you know if using a long run of wire would change anything? i assume it would just be a resistor by a different name and have no advantage in terms of efficiency. based on an online calculator, it looks like 3,000ft of 12 gauge solid copper wire would be roughly 5ohms of resistance, equal to the in series resistor in your example. so instead of a 5ohm resistor using 320w, it would be 3,000ft of wire using 320w. \$\endgroup\$
    – tau
    Commented Nov 20 at 0:10
  • 1
    \$\begingroup\$ Yes, electrically there is no difference in this case. The power dissipated per length is much lower so it wouldn't be a fire hazard, but at $0.24 per foot, it would be quite expensive. \$\endgroup\$
    – vir
    Commented Nov 20 at 0:36
1
\$\begingroup\$

You are right about resistor losses, it is an inefficient way to decrease power.

In AC an efficient approach to decrease power to load is to use a triac chopper. It cuts part of sine so effectively decreases a voltage suppling the load.

A newer option is some of SMPS topology.

The older way is to use a transformer with multi-tapped secondary. If decreasing power is needed the switch connects the tap with less turns.

I would say decreasing power is easier in Ac than in Dc btw. In Dc it cannot be done without Smps, aka making an Ac or Dc pulses at least by switching the DC on & off.

\$\endgroup\$
1
\$\begingroup\$

So many ways. Dimmers cut off sine wave at certain level leaving only part of power going through.

Variable transformer is another way.

It all depends on the load. Usually load is what defines the power.

In today's electronics one way is to efficiently rectify, get nice DC with a cool power factor, then create whatever AC you want.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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