It depends on what you define as “moderate listening volumes” but here are some examples.
Assuming 8 ohms (and assuming I did my math correctly):
1 watt = 2.8 volts and 0.4 amps
5 watts = 6.3 volts and 0.8 amps
10 watts = 8.9 volts and 1.1 amps
20 watts = 12.6 volts and 1.6 amps
Both.
For those about to rock…?
Lame AC/DC jokes aside, if it’s audible, then it’s basically AC somewhere in between 20 Hz to 20,000 Hz. The human hearing range is actually a bit narrower than that, especially as we get older, and some people have better hearing than others, but that’s the common range of frequencies that people use for “audio range”.
The coil is literally just a coil of wire, suspended next to a permanent magnet. As you vary the voltage/current through the coil, it makes an electromagnet which pushes and pulls against the permanent magnet. Since the coil is suspended on a moveable cone, the coil (and the cone along with it) move closer to and away from the permanent magnet as the polarity and strength of the voltage/current vary.
Volume is amplitude, so the voltage and therefor current vary. It will be related to the amplifier’s power supply, and how the transistors are biased to have a voltage that will vary as loudness changes.
Speaker output impedance should be really low (compared to e.g. a line output) so I suppose it acts more as a current source?
The magnetic field produced by a coil should be proportional to the current.
Good question about the average voltage; If we apply Watt’s and/or Ohm’s law as P=VI then the voltage should be proportional to the square root of the power… (so a 1000W speaker would take 10x the voltage of a 10W speaker if you crank it up)
There are some exotic speakers that don’t use coils. For example, the Plasma tweeter of the Hill Plasmatronics. Those use very high voltages at low current.
There are also piezoelectric speakers, which also use high voltage / low current. They are not favorably regarded by audiophiles, but they are loud.
The conventional loudspeaker driver, and the enclosure design is based around the amplifier as a voltage source. If you crunch the physics this leads to a flat frequency response. There are reasons why a current source might be advantageous - in particular it mitigates dynamic changes is voice coil resistance due to ohmic heating. But you need to apply a filter to modify the frequency response back to linear - so it isn’t viable unless the system is built as a unit. So is a very niche idea.
Loudspeaker efficiencies vary a lot. It is also important to not confuse sensitivity (sound output for Volts in) versus efficiency (sound output for Watts in.) They get sort of normalised at 8Ω, by setting the standard for sensitivity measurement to be 2.84 V RMS. So when you see a similar 4Ω speaker with higher sensitivity, that is usually a function of the lower impedance increasing voltage sensitivity, not improved efficiency. Whilst domestic HiFi is usually 8Ω, professional systems are often 4Ω drivers, and some car systems may be 2Ω in an effort to get power within a limited voltage budget.
There are also large scale speaker systems, where you hang speakers across the wires wherever you need a speaker. Each loudspeaker is designed (usually with a transformer) to accept a fixed peak voltage and has an impedance that has it draw its rated power at the nominal voltage of the line. Commonly you will see 70V and 100V systems. The amplifier is a voltage source, but is sized to be able to power whatever the load is.
Speaker sensitivities range from say 80dB for some domestic HiFi, to close to 100dB for professional sound reinforcement. Measurement made at 1 metre.
The question of AC versus DC is reasonably easy. Loudspeaker cones need to move forward and backward from their zero position. That needs alternating polarity of input voltage.
How you get there depends upon the amplifier topology. Most modern amplifiers have positive and negative rails and switch between them to deliver output of either polarity. It wasn’t always so.
You can make simple amplifiers with only a single power rail, were they don’t reverse polarity, so the same polarity just varying in amplitude. However you need to AC couple the output of such an amplifier - otherwise at idle it is delivering half the output rail and will cheerfully cook your speakers. A large capacitor on the output, or an output transformer. Once you do this the output is AC anyway.
These single ended designs are either the simplest and cheapest amplifiers, or the most esoteric and stupidly expensive Audiophool deigns. Many single ended audiophool designs are have additional circuitry to avoid the output capacitor. The Zen amplifier series by Nelson Pass is a modern exemplar of the possibilities, and hobbyists still build them.
Once you get to modern switching amplifiers it all gets a bit evil.
Just to address the OP’s questions more specifically.
At moderate listening levels, the answer is surprisingly low. A reasonable domestic HiFi speaker could be cheerfully idling along at a fraction of a Watt. So a couple of volts tops. A good HiFi however needs to be capable of reproducing transients. That depends on what sort of music you listen to - modern ultra compressed pop has essentially no transients - Mahler’s 6th and you have huge dynamic range and serious transients. They can require large peaks of power. (In reality, any recorded music is compressed to some extent, I don’t care how much you spend on a HiFi, it won’t match live music.)
Assuming a nominal 8Ω loudspeaker, current is also quite low at moderate levels. Ohms law. So say 2 volts, I = V/R current is a quarter of an Amp.
Both. Again, Ohms law. Also Power = IV = V^2/R
So a doubling in voltage results is a doubling of current and quadrupling of power. As above for how much so.
A lot has been said above, and the linked to thread from years gone past is little short of a trainwreck.
AC.
If one want to start arguing the minutiae of what is AC and what is DC, you discover that the pedants have left a huge fraction of reasonable signals as neither one nor the other.
The basic loudspeaker, as invented by Kellog, simply uses an electromagnet to drive the cone back and forth. Very little has changed in the century since the invention. Current on one direction drives the coil one way, current in the other direction drives the cone back the other. The current flows in response to the applied voltage. No Volts, no Amps.
If we want to get more technical, we can note that the elecromagnet (often just termed the motor) can and does act in reverse. The cone resits movement and the resistance acts to create a voltage that opposes the applied voltage. (Of then termed the back electro-motive force aka Back EMF. EMF is a fancy way of saying voltage.) The entire assembly of cone’s mass, suspension, air in the box and the resistance of the air to the cones motion creates a mechanical system that reacts to the applied force from the motor - and reflects that reaction back into the electrical world. The fabulous result is that the speaker basically behaves exactly as we want, and it is reasonably trivial to create a loudspeaker. The design of loudspeakers for bass reproduction was famously (at least for DIY HiFi geeks, by Thiele and Small half a century ago. Their work both describes the operation of, and provided design guidance, on how bass speakers operate.
thx great answer … and many above were also quite helpful.
… and of course you are right … at one moment in time, I had an inexpensive (I always stayed at the inexpensive side of hifi) chinese 1 watt Miniwatt tube amplfier …
… which - linked to (very sensitive) full range speakers made a lot of beautiful noise for that awsame watt of music.
