When it comes to true power,
not all amplifiers are created equal...
In recent years, audio manufacturers have
become ever more optimistic with the quoted
specifications of their amplifiers. In the good old
days, if you bought a good amplifier, the power
ratings were often very conservative - usually, if
the specifications said that your amplifier could
output 60 watts per channel, the chances were that
it could most certainly pump out 60 watts of real
music into each of your loudspeakers and probably
some more on top, and do it cleanly. The 1980's
bought the era of CDs and digital music as a whole,
and the craze of compressing music to increase
apparent loudness and volume (music compression is a
subject which I really don't want to get into) -
hence the modern style power rating war was born.
This era saw mainstream amplifiers
become smaller and more lightweight but still have
the same or even greater total power output as
earlier models, and with ever lower levels of
distortion. Surely it was just new technology which
meant that on the surface these new lightweight
audio amplifiers appeared to be superior in every
way to their vintage era ancestors?
New ways to measure amplifier output
power became popular at this time. This was the era
which spawned 'MPO' (Music Power Output) and even
'PMPO' (Peak Music Power Output) power ratings which
ended up being used to death to greatly increase the
apparent muscle of small consumer grade mini systems
and low end integrated amplifiers (even many larger
manufacturers used this format to advertise output
power, manufacturers who should have known better).
Over time, amplifier power ratings became more and
more massaged until we got to the situation today
where the likes of tiny desktop speaker systems have claimed power
outputs reaching into the realm of kilowatts.
Total harmonic distortion (THD)
factors in too, usually power outputs are
quoted into a certain THD figure - basically, as far
as quoted power ratings are concerned, more THD in
the power rating means that they drove the amplifier
heavier into distortion to get that figure. There's
much more than this to THD (and distortion in
general) and there has been a lot of discussion in
the past about the obsession some manufacturers have
had with low THD figures, but
that's all outside of the scope of this article. All
you need to know here is that if you drive an
amplifier into heavier distortion, you can get more
measurable output power from it - at the expense of
linearity and audio quality, of course.
An example
I have a small
average 'consumer grade' amplifier, lets say it runs
both channels on a small single power supply. Let's
be fair to this amplifier and drive it with a 1kHz
sine, a relatively easy signal for an audio amp to
drive. If I drive both channels, I can get around 25
watts RMS into an 8 ohm load before it begins to
noticeably clip the signal. I then drive one channel
only, this gives the amp a much easier time and it
can then dedicate its entire power supply to driving
just that one channel, hence it now gives me 35
watts RMS. Next I drive this little amp so hard into
distortion until my nice sine signal is now running
at 25% THD and I can barely recognise it. I now see
an output of 45 watts RMS, no matter how distorted
it looks on a scope or distortion analyser, no
matter how bad it would sound if you played music
with it running that hard. This still isn't a great
amount of power, to seriously to compete our product
against brand X's product, we need some further
massaging of these power figures:
45 watts RMS x 2 channels = 90 watts
RMS
90 watts RMS x 2 = 180 watts MPO
(Music Power Output)
180 watts MPO x 2 = 360 watts PMPO
(Peak Music Power Output)
Suddenly our humble little 25 watts
amplifier has a staggering 360 watts of power, sure
to please our advertising department no end - that's
some good advertising, eh? That's not all, finally
we can try another trick, we can take our figure
which we originally got by driving into an 8 ohms
resistive load and multiply that into 4 or even 2
ohms, regardless of the fact that our amplifier
couldn't ever hope to drive loads that low.
Ohm's law states that if you halve
the value of a resistor, you double the amount of
current (I) flowing through it - and as an
audio amplifier is a voltage amplifier, a
loudspeaker is (basically) a resistive load and the calculation
for power relating to voltage and resistance is P = V² / R, you can see that in theory
you'll get double the power output if you drive your
amplifier into 4 ohm speakers instead of 8 ohm. Not
so in the real world, as an amplifier's power supply
can't supply infinite amounts of power and will
begin to sag if asked for more current, and as the
amplifier's internal circuitry contains resistances
which will themselves cause voltage drops, you'll
hit a limiting factor which means that in reality
you'll rarely see anything near the amounts of power
that you'd get in theory. But, that doesn't stop our
advertising department trying, hence:
360 watts PMPO into 8Ω = 720 watts
PMPO into 4Ω
360 watts PMPO into 8Ω = 1,440 watts
PMPO into 2Ω
Yes, we've now transformed our
little amplifier into a 1.5 kilowatt behemoth. The
last two calculations are the extreme and not many
manufacturers would dare to go this far but it
sometimes happens - just look at the quoted power
outputs of some desktop PC speaker systems.
Some manufacturers continued to be
reasonably honest, quoting more than one power
output figure for their amplifiers. Usually, if you
look at the specifications at the back of the
handbook and there is more than one power output
quoted, the lowest power rating is the most truthful
- the others are for marketing purposes and should
be ignored. If you see figures quoted into 1% THD or
beyond, you can discount them as a lie and subtract
10% at the very least from that power output - the
exception here are valve (tube) amplifiers which
often measure badly but sound very good - if you own
valve equipment you shouldn't be worried about power
outputs anyway.
 |
|
Pioneer SA-9800 |
Also, there are other ways of
greatly increasing the power output from amplifiers,
most to do with measuring the power of very short
pulses combined with some math - indeed, some
manufacturers do calculate some of these terms in
different ways. From what I've seen over the past
few years, 'Music Power' was
originally devised in the 1970's as the quoted peak
output power driven from a burst signal for a short
period (typically a fraction of a second), however nowadays
it often seems to be merely a way to multiply the
RMS power of an amplifier and bears no relation to
it (at least not as peak voltage or current does to
RMS) - although I'd be happy to
be corrected on this.
In the image to the left is a Pioneer
SA-9800, this gives you some idea of the amount of size and
weight you need for a genuine 100 watt per channel power
output. Each channel has a huge power transformer and a
brace of 15,000µF reservoir capacitors, the whole thing
weighs nearly 30Kg.
There ARE ways to get more power for a
smaller size and weight, for instance the likes of the
Carver Cube which used high frequency switched power
supplies (higher frequency = more efficiency and hence
smaller components and lighter weight). But this technology
is rare in home audio and a traditional amplifier truly
capable of a hundred watts or more of clean power is going
to be big and heavy.
A good rule of thumb if you're
shopping for an amplifier (and it doesn't matter
whether you're looking for a cheapo desktop stereo
or a £2,000 power amplifier), ignore any power value
that isn't specified as RMS or 'Continuous'.
Power amplifier module from well known
brand of mini hi-fi system, using an STK IC.
Quoted power is 390w PMPO, the datasheet for the
chip only rates it at 35w RMS per channel...with
the tiny power supply you're probably only
seeing 25-30w RMS per channel once it's in the
mini system.
One of the early integrated power
amplifier ICs, made in the late 1970's by
Sanken. No datasheets for this but from memory
the power was around 35w RMS per channel. This
came from a Sharp SG-460E Music Center, one of
the first pieces of audio I ever owned. It was
effectively a regular power amplifier circuit
shrunk and fitted to a ceramic base, not as good
as a regular amplifier but still a feat of
engineering back then. This one has a blown left
channel, you can see damage on one of the
transistors and a singed resistor track nearby.
Power consumption vs. power
output
Another thing to mention, sometimes
people take the actual power rating from the rear
panel of the amplifier and use that to advertise the
power output, this seems to happen a lot on a
certain internet auction website. This is misleading
for several reasons.
Firstly, this figure is merely the
maximum amount of power which the amplifier will
consume from the mains supply. Depending on the
configuration of the amplifier, it might have a lot
of non-audio circuitry inside which also consumes a
large portion of that power, in the case of a modern
stereo receiver the radio and LCD displays and
various processors will sink quite a lot of power on
their own and none of this will be used as power
output through the speakers. Also, some
manufacturers fit power take-off sockets on the rear
panel to power tuners or turntables etc, in those
cases the total available to these will be added to
whatever is on the rating label.
As a very rough rule of thumb (and I
stress that this is a rough guide, don't rely on
it), if you have an average stereo power amplifier
with no extra power sockets on its rear panel and
other no bells or whistles to consume power, at best
you'll get 60-70% of that rated power through your
speakers, substantially less if it's running class A
or similar and creating lots of heat. You definitely
won't get more power than it consumes from the mains
supply; as the Law of Conservation of Energy states:
"Energy cannot be created nor destroyed".
Back to amplifiers page
