Why Sony watts aren't as good as _____ watts

One again I’m writing something as an electrical engineer who knows just enough to be dangerous. Please keep this in mind. One more note, I’m not deliberately picking on Sony. This really applies to any low cost receiver.

A friend of mine is considering buying some new equipment. He currently has a Sony receiver rated at 100wpc. I was showing him a Rotel amp rated at 75wpc and I assured him that it should have just as much real world power as his Sony. This got me thinking about why are Rotel, McIntosh, Bryston etc watts better? After consulting with my brother, who unlike me has designed and built an amp, I came up with this idea.

It all comes down to power supplies: This is not entirely true but I’m going to assume that both amps have all the heat sinking they could use and I’m not looking at little things like tonality and such. I think it’s safe to assume that receivers have a lot of sound quality compromises but I’m looking at watts of power. Power needs a good power supply.
Most amps (all?) will clip when one of two things happen:
1. The output voltage can’t go any higher.
2. The output current can’t go any higher.

All power supplies have voltage and current limits. For the sake of comparison I’m not going to look at loss in the amp circuit, I’m really interested in what can be delivered to the speakers.
An amp delivering 100watts into 8 ohms is delivering 3.54A at 28.28V. ie the voltage at the speaker terminal is 28.28V, the current through the wire is 3.54A.

Amp manufactures like to advertise big power supplies which can deliver high current. A voltage limited amp (ie all the current it needs) will double in output power every time you halve the speaker load. The opposite is true in a current limited amp. In a current limited amp the power HALVES each time you halve the load (lets assume you Sony can drive those Magnepans)
Some quick math (remember Power = Current^2 * Resistance) yields the following numbers:

Good beefy amp (lots ‘o current)

Low $ amp (meek current)

In both cases the limiting factor at 8 ohms is assumed to be voltage (this is usually the case.)

Numbers mean a lot in low cost products. The manufacture will rate the highest power possible, but cost is everything. The manufacture will use a power supply just big enough to get the numbers marketing asks for (100watts into 8 ohms). Since the power supply can provide more current the amp’s actual power drops as speaker impedance goes down. This is just the opposite of those big He-Man amps.
But my speakers say 8 ohms… A quick look in a Stereophile speaker review shows all sorts of pretty charts. One of them is speaker impedance vs frequency. This numbers are all over the place. A quick look at the B&W 302 shows this 8 ohm baby swings from 3 to 20 ohms. At impedances over 8 ohms both amps are limited by voltage. When things go below 8 ohms beefy becomes stronger while meek gets meeker.

Now the question of how many Rotel watts = 100 Sony watts. That really depends on what impedance you are interested in but… a 75 watt good amp will deliver 86 watts into 7 ohms. This is the same at the 100 watt meek amp. Drop the impedance any lower and the Rotel looks better and better.

Any questions?
I'm not sure what your take is on power, but I think we probably agree. Here's my take:

In a perfect world each time you halve the load impedance,
the power will double. Unfortunately, all amps are not created equal in this regard. Take a well known amp like a Krell. It has such a huge PS that it WILL double it's power output as the load is halved. The reason for this is it's voltage doesn't droop. Another amp not quite so beefy can't maintain its output voltage and so power will only increase by little, if at all.

Another thing I like to look at is the way an amp is rated.
Some amps rated at 100 watts actually only deliver that power into a narrow band of frequencies. As an example, the FTC mandated spec is supposed to be 100 watts from 20hz-20khz at .1 thd. But many manufacturers fudge on this, especially tube amps. But hey, I own tube amps and they deliver all the power I need driving magnaplanars.

The bottom line is that if you have speakers with a tough load (low impedance, high phase angle, etc.) you need a Krell or something like it. Otherwise, you're spending a lot of money on a big PS. Huge PSs are what mostly drives the cost of amps into the stratosphere. Remember, an amp puts out voltage, the speaker(load) draws current. Of course, if the amp can't deliver the current required then its power will not increase.

Jim Robinson
Power supply is much larger and current delivery is higher on quality amps.
Well, all printed specs are only a guideline. No one takes the time to design and amp that is exactly XX watts. The watt specs are always on the low side, partially for legal reasons (false advertising on power), and many times geared to get a certain low distortion reading. It is more likely that a mass market receiver will be less conservative to impress Joe Consumer with the power.

I can remember checking out HT receivers for a friend. A certain brand had a receiver that was rated at 50 watts at 8 ohms and 110 watts at 4 ohms, which does not make sense. The 100 watt receiver at 8 ohms was 150 watts at 4 ohms. It is more likely the 50 watt receiver was a 75 watt receiver, but they probably wanted to justify asking $100 more for the 100 watt model, so they understated the other model. They probably costs the same to manufacture in China, so if they can get you to spend an extra $100 to get double the power, they'll profit.
correction in the second to last paragraph,

"Since the power supply can provide more current the amp’s actual power drops as speaker impedance goes down. This is just the opposite of those big He-Man amps."

It should read "... supply CAN'T provide more..." That should make more sense.

Jwrobinson - you are correct about how things are rated. I had a low buck Sony receiver rated at 60wpc into 2khz. No mention of 20-20khz rating. We have the same view on power supplies. Big supply means you have more power as the impedance drops. If the PS is just big enough for an 8 ohm load (ie it's current and voltage limited at 8 ohms), it will stay current limited into a lower impedance load. Do a little math and the amp is becoming less powerful as the impedance drops.
Of course this is a simplistic take on the subject and looks solely at output power, which we all know is not the best way to judge amps. I came up with this to answer my friend's question, "Is a 75w Rotel as powerful as my 100w Sony." This is a somewhat technical answer but my friend is an electrical engineer and who understand it.
Hi Nikkidanjo

You say "Of course this is a simplistic take on the subject and looks solely at output power, which we all know is not the best way to judge amps."

I guess I would agree. For example, Class-A amplifiers are generally capable of a very modest current, usually barely above that theoretically needed to drive the speaker. I have not heard anyone claim they are rubbish, because of the low current capability. You do point out a couple important formula one which you clearly state... Power equals current squared TIMES resistance. The other at the bottom of your analysis/numbers( which you do not state)is Power equals Voltage squared DIVIDED by resistance. This might make things easier to understand for folks not familiar with some of this.

I remain,
Open them up and see. When driving an amp hard, quality counts. Even Sony vs. Sony ES will use many different parts with the same power output. When you start working the amp hard, those .02 vs .005 distortion ratings make a big difference. I think we all know that heavier guage internal wiring, better caps, better solder, etc... all add up to lower noise which allows us to drive the amp harder without any audible/damaging distortion. I'm no engineer but, I do know that my 20 watt class A amp weighs more than my 840 watt theater receiver!
What were the apms driving when the power was spec? Most Japanese amps drive only a 1khz tone and not pink noise full bandwidth from 20hz-20khz You will read it in the manual something like 100w @1khz.

Other thing is 100w @ 8ohm @1khz at wiat distortion level? The higher the distortion figure, the higher the power.

Power suplplies are not stiff so when driving real speakers as opposed to "dummy loads" will clip as you know speaker impedence is variable with frequencies. An 8 ohm speaker might have bad impedence curves which swings fro, 2-18 ohms.

If the power supply is stiff enough, power rating will double when impedence is halved.

Just remember, power or watts is not everything in high end audio. There are just too many variables to change the figure. Anyways, just listen to the music and forget about the numbers.

Hope it helps.
I'm not sure I agree with Genesis168 about stiff power supplies. Geez, that's a term I haven't heard in years. I think it originated back in the days of the high-powered Receiver wars (mid 70s?). A stiff PS usually referred to a Receiver that had high power but little headroom. In other words, they were rated as high as possible and not necessarily across the entire freq range. Some loosely regulated PSs were rated very conservatively but with as much as 6db headroom.

Also, someone wrote about Class A amps not having much current capabilities. I've owned a few that were rated
Class A up to a certain power rating and then sort of transitioned over to Class AB. I don't think Class A amps are low current capable per se. As we know, a Class A amp's
output devices are always on. Class AB devices are only on part of the time, with each output device driving only half the waveform. Of course, more power can be had by using multiple output devices. One of the advantages of of Class A (or so they say) is a lack of crossover distortion. One of the drawbacks is that Class A amps run hot especially when they are at idle and not driving a load.

I want to qualify everything I have said by stating I am not an EE, but I've been involved with audio(hi & mid) for more than 30 years, so I don't get involved with formulas and such.

Jim Robinson

Jwrobinson, I like your dB headroom spec. Naim is one of the few companies who has listed dB headroom. It was at least 6 dB and probably much more.
Dynamic, instantaneous current delivery can maker a 30 wpc amp like Naim sound much more powerful. I also think it contributes to a power amp presenting more detail.
I'm waiting for the tube crowd to jump all over this one.
Jwrobinson, "stiff" was ment to be a word NOT what you were reffering to in the 70's.

What I was meaning was that the power supplies in cheaper gear were not "big" enough or will "droop" when a tough load is driven causing the amp to clip.
There's obviously a lot more that goes on than this, because the Sony won't sound as good as the ______ even when neither amp is being called upon to deliver anything above a few watts max, probably less than 1 watt average for normal moderate volume listening with most non-exotic speakers. Differences in amplifiers' music reproduction capabilities don't just make themselves known around only the clipping point.

You're right, stiff is just a word. But a stiff PS is one that is highly or tightly regulated. Its rated power being somewhat higher than a loosely regulated supply. The problem as mentioned before is that there is little headroom, though it probably won't matter if both types are running at a watt or less.

Go to www.passlabs.com click on ARTICLES once there click on POWER SUPPLIES. This is from Nelson Pass whos is truly one of the most gifted of the Soild State designers ever,if not the most gifted of all. This article will answer your question completely.