Designing and Building
Audio Amplifiers.

How Much Power is Enough?

I should note that this discussion is about consumer grade home entertainment amplifiers. Public address and musical instrument amplifiers require much higher power levels and must have the capability to sustain high output levels for long periods of time without being dammaged. In short, they are a horse of a different color.

The question of how much power is enough is one that can be kicked back and forth and never really solved. Perhaps a little history might be of help.

Some of the first amplifiers that called themselves high fidelity used 6V6s, 6L6s, or other comparable tubes that delivered 10 to 25 watts. This was before the popularity of stereo so the qualifier "per channel" is unnecessary. When tube technology was at its pinnacle, stereo was becoming quite popular, and transistors were just beginning to be heard from, The amplifier many audiophiles aspired to was a 50 watt per channel using EL34s.

Most equipment reviewers agreed, at the time, that the Heathkit AR-15 was the breakthrough product for solid state amplifiers/receivers. For those too young to remember and who think of the AR-15 as a gun, The AR-15, and later the integrated amplifier version AA-15, gave 50 watts per channel and previously unheard of low distortion figures.

The sound seemed good but something sent the serious listener in search of higher and higher power levels. That something was transient distortion. It can't be measured and quantified in the laboratory so the only way to test for it is to listen. Listening tests are so subjective that you would be hard put to find two people who can arrive at the same conclusion about a given amplifier.

In the section titled "Overall Feedback, Pros and Cons" I describe in detail what is wrong with solid state amplifiers. A brief summary is that the solid state amplifiers used too much feedback. The attitude of most design engineers is "if it can be done, let's do it." The elimination of the output transformer with its large inductance and magnetic non-linearity opened up the possibility of almost limitless amounts of negative feedback. So, they pored it on. The result was very low steady state distortion values but in listening test most agreed that the amplifier had a somewhat harsh sound as if it were being driven very close to clipping. The statement was often heard that "tubes overload more gracefully than transistors."

Fans of tubes stayed with them while fans of transistors, myself included in those years, sought higher power. For me it culminated in a 200 watt per channel amplifier of my own design. Its sound was effortless but I was only using about 1% of its capability. Unfortunately my method of construction has proven to be unreliable and it no longer works.

My next step was to pickup a basket case Harman-Kardon A-300 on ebay and restore it to operation. I was quite surprised by the sound. After all it had been 35 years since I had heard a tube amplifier. The H-K A-300 gives about 10 or 12 watts per channel. I don't have any of the original literature on it but it uses a ruggedized version of the 6V6 so it can't be much more than that. It has the same effortless sound as the 200 watt per channel solid state had.

My Conclusion.

At the peak of tube technology 50 watts per channel was plenty for most listeners. In the solid state age listeners pushed manufacturers to higher and higher power in an attempt to get away from the transient distortion that plagued those amplifiers. Evidently they succeeded somewhere above 200 watts per channel. The inherent cleanliness of the tube sound makes it possible to return to sanity when it comes to power levels.

It is true that there have been some super power tube amplifiers built and sold but frankly I don't see the need for them unless some very inefficient speaker systems are being used.

Speaker Efficiency.

The statement is often made that if you are using air suspension speakers, once known as infinite baffle, you need a higher power amplifier. I think this case is overstated. While there may be one or two models of speaker systems that need a proverbial sledge hammer to drive them most are not all that less efficient than a ducted port system. I am using a pair of SPL 99 air suspension speakers with my A-300 and it sounds just fine at normal living room levels. I have run into some clipping when listening to CDs at high volume. I will likely go up to 30 watts per channel and be happy with that.

sNext; A Low Distortion Gain Block.


Or use your "Back" button to return to where you were.

Thank you for visiting my page at Angelfire.
Please come back and visit again!

This site begun March 14, 2001

This page last updated June 8, 2005.