"Watts" Versus "Current"

Current is a component of watts (power) as Dan just indicated. Therefore, all watts are not created equal.

1 amp (of current) X 10 volts = 10 watts of power. But 10 amps X 1 volt also = 10 watts. Further, the basic W =A x V really works for direct current (DC) only. The formula is a bit more complicated for AC (as in: music signal). But leaving that aside for now, how the power is created will either be better or less good for certain kinds of speakers.

Why? Because every kind of driver generates a reactive signal BACK TO THE AMP. When the signal from the amp moves a cone back and forth, for example, the action of the cone's voicecoil in the magnetic field actually GENERATES a reverse electric current BACK to the amp! This reverse current adds to the impedance (load) the amp "sees".

What's CRITICAL (in amp selection) IMO, is HOW how that (let's call it 'phantom') impedance is created. If the amp sees a capacitive type of load (an electrostatic speaker) it needs for its available watts to consist of lots of amps (current). If it sees a resistive type load (like cones, domes, ribbons and planar magnetics) it needs for its available watts to consist of lots of volts (to overcome the reverse voltage created by those kinds of drivers.)

I'm going to generalize here, but 1 tube amplifier watt (a typical push-pull circuit with an output transformer, nothing exotic!) will be rich in the current it can make available (for that 1 watt of output.) 1 solid state amplifier watt (a typical push-pull circuit with the output transistors directly coupled to the speaker) will be rich in the voltage it can make available (for that 1 watt of output.)

You can guess where this is headed ;-) Tubes are better than silicon for capacitive loads, and silicon is beter than tubes for resistive loads. All right -- this is a sweeping generalization! HOWEVER, because the impedance of all speakers varies widely depending on frequency (the rated impedance of speakers is only a nominal figure) there could be times when you'd need a HUGE tube amp to drive those dynamic drivers (God forbid acoustic suspension woofers!) properly. Or if you are driving electrostatics (panels only, we're talking about) you might need a HUGE solid state amp to give those stats the current they need when a tube amp of far less output would sound as good (and likely better IMO ;-)

People into fine audio are generally at the 3-digit Stanford-Binet level. But many (probably most) have had little occaision to become familiar with the nature of (yikes!) circuit design. But if you will take a little time (good bathroom reading for instance as it takes several passes to sink in ;-) and go through this excellent and (reasonably) clear, easy to understand online tutorial, you will find yourself more informed than many of the self-appointed/annointed 'know-it-alls'/'experts' that haunt the halls of audio-underworld ;-)
http://www.allaboutcircuits.com/vol_1/index.html
.

You're quite right, nothing really happens until those watts are called for, but when they are, the device appears to the amp as a certain kind of load. And it's the nature of the speaker load (or more accurately, the nature of the speaker's reactance) that will determine mix of amps x volts the amp must supply.

So it's not just that speakers present different impedences at different frequencies, but that those impedances can be resistive or capacitive. Remember that current flows easily through a voice coil (so volts are called for), but almost not at all across the [air-space + step-up-transformer] assembly of an electrostat (so current is called for.)
.

OK, smarty pants!
What color amp should I buy???

does it matter? surely whatever color(s) it is, it's bound to match several things in your pièce derrière le garage ;-)

Swamp, not to take away from Dan's similar explanation, but if you read my post at the beginning of this thread, you'll realize I said the exact same thing:

Therefore, all watts are not created equal. 1 amp (of current) X 10 volts = 10 watts of power. But 10 amps X 1 volt also = 10 watts. Further, the basic W =A x V really works for direct current (DC) only. The formula is a bit more complicated for AC (as in: music signal). But leaving that aside for now, how the power is created will either be better or less good for certain kinds of speakers.

Why? Because every kind of driver generates a reactive signal BACK TO THE AMP. When the signal from the amp moves a cone back and forth, for example, the action of the cone's voicecoil in the magnetic field actually GENERATES a reverse electric current BACK to the amp! This reverse current adds to the impedance (load) the amp "sees".

What's CRITICAL (in amp selection) IMO, is HOW that (let's call it 'phantom') impedance is created. If the amp sees a capacitive type of load (an electrostatic speaker) it needs for its available watts to consist of lots of amps (current). If it sees a resistive type load (like cones, domes, ribbons and planar magnetics) it needs for its available watts to consist of lots of volts (to overcome the reverse voltage created by those kinds of drivers.)

.

Swamp, in the beginning, most of us (I know I did anyway) conceptualized 'impedance' and 'resistance' as being essentially the same thing; because they are both expressed as ohms. and we were taught (first) that ohms are a measure of resistance. Strictly speaking, that is NOT TRUE! It is IMPEDANCE that is measured in ohms, and RESISTANCE is simply ONE KIND of impedance ;-) Impedance is indeed the better term, because it refers to any condition that IMPEDES the flow of current (resistance being just one such condition.)

So, for instance, when a manufacturer specifies their speaker is "4 ohms" that's NOT what you will read if you just put an ohm meter across the speaker terminals ;-) That 4 ohm "nominal impedance" is in fact a "weighted average" of the impedance (to current flow) that the amplifier "sees" at EACH frequency across the entire spectrum. But that speaker is by no means 4 ohms across the board!

Further, when a speaker manufacturer specifies "power handling capacity" (like, 50W - 200W) they're not just saying the minimum watts you need to 'make it play', or the maximum watts before you 'fry' your speaker ;-) What they are really getting at is the RANGE of (amplifier) power ratings in which the speaker will produce the whole frequency spectrum (and without distortion.) For example, below 50 watts, some parts of the frequency spectrum (associated with hi impedance) may not reproduce at all, or only with distortion. On the other hand, driving the speaker above 200 watts, at certain (other) parts of the frequency spectrum associated with lo-impedance, you could indeed fry something (like a dome tweeter's voice coil, or burn a hole in your electrostat's diaphragm!)

Did you know that a MartinLogan panel's IMPEDANCE can drop to less than 1 ohm at 15KHz?! So if you were using a ss amp with a rated power of say 200W/ch @ 8ohms, then @ 4 ohms (following Ohm's Law) it could deliver 400W, @ 2 ohms, 800W, and @ 1 ohm, 1600W!! But at each stage, the VOLTAGE part of those bigger watts would be increasing more and more, compared to the current part. So if you were playing a LOUD 15kHz with a solid state amp into your MartinLogan, huge voltage would build up across the air gap in the panel, and pretty soon ZAP - lightening! (better known as arc-ing) and you've burned a hole in your diaphragm!

OTOH, due to the "impedance matching" capability of transformers, a tube amp won't produce this result with an electostat (unless it's a mutha of a tube amp, and your hand slipped on the volume control!) Tubes have a high internal impedance (think of a high revving 4-cyl engine) and speakers generally have a low internal impedance. Solution for getting the power to the road? A 6-speed gear box! Better known as an output transformer. If you select the correct 2, 4, 8, 16 ohm tap (gear) then you can match the amp's power (torque) to the speaker's load (incline of the hill) for maximum power transfer! (Don't ya just love physics?!) What this means (broadly generalizing ;-) is that a tube amp can supply the same watts, comprised of the same proportion of amps and volts, into any impedance. Only one solid state amplifier I know can do this, and that is McIntosh; BECAUSE they ALSO use output transformers on their SS amps, even though they don't really have to -- but it keeps THEIR clientèle from making bad amp/speaker choices ;-) I just happen to be a McIntosh "clientèle" but not because I don't know how to select an amp! ;-)
.