Are all watts equal?

[kandamrgam]

Hello everyone,

I often hear sentiments like 10W from class-D amplifier is equivalent of 5W from a class-A amplifier. Aren't watts always just that, watts?

When buying an amplifier, all I have to look into is the wattage rating at low/high/whatever impedance, right? Or are there other factors which affect the delivery of the said watts?

Just trying to get some rational answers from the scientists here..

Regards,

 

[Uzsidoboz97]

It just reminded me of this, sorry if inapropriate

 

[raindance]

So long as the spec is watts RMS for both amplifiers and measured for the same frequency range and at the same distortion level, watts are watts. Oh, and measured for the same duration.

The only reason for old fashioned class A or vacuum tube amplifiers to sound more powerful is gain. They typically have far too much of it and very little turning of the volume control gets it loud. Gain and power and not related at all but are often confused.

 

[nerdstrike]

Consider the petawatt laser. It sounds like you could blow holes in planets with one, but the wattage is the result of an extremely short pulse time and a long interval between shots. The time window of the measurement matters. Also, you have to keep your mirrors really really clean.

Cooling can matter too, e.g. a big class A amp will make your room toasty in no time and class D less toasty due to their relative efficiencies when not running at full power.

 

[rdenney]

High gain, low power. It gets loud at low volume settings but clips at slightly less low volume settings. Low-gain, high power may require the whole range of the preamp's volume knob but it ultimately gets louder. And maybe when the low-power amp clips, it does it in a way that is less noticeable.

Power is the product of voltage and current. Voltage and current are amplitudes of waveforms. As the voltage increases, the demand for current increases, and that current waveform may lag behind the voltage waveform by some phase angle that depends on the load. If the amp is unable to deliver the required current at any given instant, the voltage will be unable to increase and the amp will clip (i.e., the waveform will go flat). Instantaneous power is the product of those two waveforms at a single instant--the rate at which energy is delivered.

We measure power as a surrogate for the maximum voltage and the maximum current capabilities of an amp. But where do we measure it? At the peak of the voltage waveform, the peak of the current waveform, or some average? RMS (root mean square) is just a way to average out a alternating-current waveform, but it's just one number. Normally, when an amp is rated for RMS power, it will have the power to deliver the entire waveform that the RMS value describes. But this may be tested with a sine wave, where the RMS value is 71% of the peak. At the opposite extreme, the RMS voltage of a square wave is equal to the peak voltage.

And then there's the question of where this is measured in the relationship between power and distortion. The most conservative power capability measurement is the RMS power for a sine wave measured at the point where distortion starts to climb rapidly, which is often called the knee of the curve. That's usually where Amir reports maximum power. Another is to measure it at the rated distortion provided by the manufacturer, and another is to measure it at some minimum acceptable standard of distortion, such as 1%, which is where a lot of manufacturers report it. The power vs. distortion curve answers all of those for those willing to understand it.

The real question is: Can an amp deliver the current necessary to fill out the voltage waveform? It's the voltage that creates the EM field in the voice coil of the speaker and causes it to move. Any amp being asked to produce, say, 10 volts RMS will need to be able to deliver 1.25 amps of current into an 8-ohm speaker load to avoid clipping, assuming zero phase angle. The amp at that RMS value is producing 12.5 watts RMS. That 12.5 watts RMS will be the same no matter what the architecture of the amp is.

Some amps clip hard when their power limits are reached, and others increase in distortion more gradually. If power is measured at the knee of the distortion curve, the latter may seem to be more powerful than the former even if it is rated lower. Proponents of this or that type of amp are often depending on these games to justify their point. But reading those curves as tested will expose the reality that a single numerical power rating might obscure.

Rick "educated consumer" Denney

 

[mhardy6647]

Consider the petawatt laser. It sounds like you could blow holes in planets with one, but the wattage is the result of an extremely short pulse time and a long interval between shots. The time window of the measurement matters. Also, you have to keep your mirrors really really clean.

Cooling can matter too, e.g. a big class A amp will make your room toasty in no time and class D less toasty due to their relative efficiencies when not running at full power.

thus the importance of the watt-hour, which is a measurement of energy. The watt is a measure of power, with (SI) units of joules per second -- a measurement of the rate of energy transfer or conversion.

Someone already invoked the mythological "RMS watt" -- no such thing per se (RMS volts, sure, but not RMS watts) -- but, yeah, the notion is valid. "Sine wave power" or "continuous power" measures, ahem, continuous power output capability over some period of time... perhaps, theoretically, forever.
Put a big ol' bank of capacitors somewhere and a laser or an amplifier or whatever can deliver lots of power... for a short period of time. The amount of power it can deliver over a longer period of time will (almost always) be less.

 

[olieb]

Are all watts equal?​

Regarding your question about amps the answer is yes, as has been explained.
But as always, the first watt is more equal than others.

 

[solderdude]

Hello everyone,

I often hear sentiments like 10W from class-D amplifier is equivalent of 5W from a class-A amplifier. Aren't watts always just that, watts?

When buying an amplifier, all I have to look into is the wattage rating at low/high/whatever impedance, right? Or are there other factors which affect the delivery of the said watts?

Just trying to get some rational answers from the scientists here..

Regards,

With class-D (actually all amps that have a regulated power supply) the peak/music power rating is almost the same as the continuous rating.
When an amplifier is using a design with a big transformer in it then there could be a difference between continuous and peak/music power.

When using an amplifier to listen to music the music power rating is probably the best comparison.

Then there is another issue.... speaker impedance which is never exact 4ohm or 8ohm.

Another confounding factor is the published specs. These could be substantially different than the actual performance.

It thus has nothing to do with the class (A-AB-T-G-H) but more with the type of power supply used, the ability to drive different impedance speakers and above all the listed numbers versus the actual numbers.

Then there is the difference between the number Watt meters indicate and actual Watts.

Technically Watts is Watts but Watts in a spec sheet may differ from Watts in reality.

To make things even more complex... amplifiers do not output Watts but rather a voltage and the load determines the current (and thus Watt).
1W in 8ohm = 2.83V
2.83V in 4ohm = 2W
1W in 4ohm = 2V

Amplifiers have an output voltage limit and an output current limit.

 

[neRok]

It just reminded me of this, sorry if inapropriate

View attachment 477077

Haha. There isn't even 21 months!

 

[raindance]

Oh, I should add, zero or low feedback amps usually have gobs of gain. I went to a high end audio show some years ago and listened to a flea watt SET amp driving a pair of large Zu speakers. The amp was clipping continuously, but all the seats in the room were filled with audiophiles nodding their heads to the fantastic sound and the dude doing the demo said "it is only on quarter volume, so it still has lots of power in reserve". The sound was so bad (can anyone say HARSH?) it actually hurt my ears. It just goes to show, fidelity is very subjective. At the same show my favorite speakers were a pair of JBL Everests driven by a sensible amplifier. They were smooooth...

 

[DVDdoug]

If you look at the reviews/measurements here, you can't tell the class from the measurements (or specs).

If you were measuring the wattage consumed at the wall outlet compared to full-power amplifier output, at full power output and maybe idle, the efficiency would be a big clue! Except if it were a tube amp you might be fooled by the inefficiency.

 

[egellings]

The watt, like the meter or yard, is a unit of measure. There is only one size of the yard; yards don't come in different sizes. Same for electrical watts. Some amplifiers of the same wattage may sound louder or softer depending upon how the amplifier and speaker interact. An amplifier with a higher output impedance will not be able to deliver all its power into a speaker because due to amp's internal impedance, some of that power will be dissipated internally in the amplifier itself, due to the voltage divider effect of the amp's and the speaker's impedances.

 

[Coltr@ne]

http://www.audioreview.it/wp-content/uploads/prove/storiche/marantz-sr-14.html/MARA_CCL.jpg

 

[mhardy6647]

"We" (the hifi/audiophile axis of not-particularly-evil) would probably do ourselves a favor by eschewing measurements of power and just go for energy.
E.g., measure the watt-hours of energy an amplifier is capable of delivering while a specific recording of The Planets is played through it. Normalizing the downstream part (load) might be tricky, but it's a solvable problem!

In a slightly more serious aside spawned in my fertile fetid mind, compare the not-entirely-unrelated but very different measurements of a loudspeaker's ability to convert electrical power into acoustic power. I've already tipped my hand!
"We" usually think, and speak, in terms of sensitivity, a certain signal voltage into a certain load (e.g., 2.83 AC volts into an 8 ohm load, corresponding to 1 watt) produces a measured SPL of x decibels at a fixed distance from the load (typically 1 meter). It's useful in a practical sense ("How loud will this amp let me crank my AC/DC record?"). But -- there are all sorts of challenges with standardization there!

The less commonly encountered but (arguably) more useful measurement is efficiency. What percentage of electrical watts in to a loudspeaker does that loudspeaker transduce out (so to speak) in the form of sound, and measureable as acoustic watts? Apples to apples, and independent of geometry. Not necessarily easy to measure, though, and also maybe not as intuitive as SPL.

A rule of thumb I read somewhere when reading involved paper products and probably taken from Col. Paul Klipsch's musings, is that the acoustic power output of a symphony playing at more or less full tilt (ffff? fffff? that I don't know!) is on the order of one watt.
If it takes 100 electrical watts in to a given loudspeaker to get that one acoustic watt out, the loudspeaker's efficiency is 1% (0.01). Klipsch bemoaned the poor efficiency of his direct-radiator competitors, and trumpeted his horn-loaded systems' demonstrably higher efficiency.

I am still a proponent of high-efficiency loudspeakers -- bu I digress...

 

[10PSPMemSticksForMiku2007]

There is actually an issue with many of the popular class-d amps where they only output their rated power with a high input voltage like 4-5 volts, and when reviewed they are often tested with such voltages, but many consumer devices are putting out 1 to 2 volt signals. In this case some 100-watt rated class D amps might actually put out less power than a 100-watt rated class AB amp when the input is typical 3.5mm/rca unbalanced from a CD player or PC etc. I saw this on the "Scientific Audiophile" Youtube channel.
I'm not saying this applies to all of the Class-D amps but it is an important point to consider for your amp purchase.

 

[rdenney]

There is actually an issue with many of the popular class-d amps where they only output their rated power with a high input voltage like 4-5 volts, and when reviewed they are often tested with such voltages, but many consumer devices are putting out 1 to 2 volt signals. In this case some 100-watt rated class D amps might actually put out less power than a 100-watt rated class AB amp when the input is typical 3.5mm/rca unbalanced from a CD player or PC etc. I saw this on the "Scientific Audiophile" Youtube channel.
I'm not saying this applies to all of the Class-D amps but it is an important point to consider for your amp purchase.

That's a matter of gain.

My Buckeye NC502MP has a sensitivity of about 2.5 volts with my 6-ohm (nominal) speakers. That corresponds to a gain of about 26 dB.

My Holman preamplifier can attain its rated distortion at an output of 8.4V RMS. That puts it in the middle of the road for preamplifiers, certainly not unusually high output. It loafs driving the Buckeye to full rated power.

A standard CD player puts out 2 V at 0dB, 1KHz. My Naim is at 2.1. But those are single-ended outputs, and the Buckeye takes balanced inputs. I used a balanced converter that completes the balanced waveform, adding 6 dB to the voltage of the input device. The 2V coming in will be 4V going out, which is again abundant for the Buckeye, even if I don't use an intermediate preamplifier.

Even the Benchmark AHB, which is absolutely NOT a Class D amp, has the option of low-gain inputs for high-output sources (like their preamp) that will improve its performance.

Rick "read the specs" Denney

 

[10PSPMemSticksForMiku2007]

That's a matter of gain.

Indeed, I'm not saying the class D amps are bad, I'm saying you need to consider their input sensitivity when purchasing your amp, that is, "read the specs." For someone who just wants to plug their CD, phono, or computer into their amp - for someone who is buying a sub-$100 class D amp - that is important.

 

[izeek]

pair of JBL Everests driven by a sensible amplifier. They were smooooth..

maaaan, of that im sure! would looove to hear a pair.

 

[Andysu]

Indeed, I'm not saying the class D amps are bad, I'm saying you need to consider their input sensitivity when purchasing your amp, that is, "read the specs." For someone who just wants to plug their CD, phono, or computer into their amp - for someone who is buying a sub-$100 class D amp - that is important.

i like the class d amplifiers behringer

 

[Petrushka]

Specs confirmed by competent reviews are equal.