Efficient speakers v. less efficient speakers

I agree with the posts pointing out that a high efficiency system usually has better dynamic contrast than a low efficiency system.

Let me try to explain why. It's a nasty little secret almost nobody talks about.

Theoretically, a loudspeaker's output will increase by 3 dB for a doubling of input power. In practice, this is ALMOST NEVER true. The reason is power compression (also often called thermal compression), and its primary cause is voice coil heating. As you increase the power going into a voice coil, it heats up. As it heats up, its resistance increases. As its resistance increases, more of the power going into it goes into overcoming that resistance (heating it up still more) and less goes into actually producing sound.

Let me give a few numbers as an example (drawing on measurements posted by Bill Roberts on Audio Asylum). At normal volume levels, the typical 86 dB efficient speaker may well only give you an average of 2.5 dB increase in loudness for a doubling of input power. So let's say you have an 86 dB efficient speaker playing at 80 dB average volume level, and along comes a +20 dB peak (quite common). This speaker will compress the peak and you'll only get about +17 dB. On the other hand, a high-efficiency system (say 96 dB efficient or higher) usually has negligible power compression at normal listening levels, and will more than likely give you the full +20 dB that the peak calls for.

Once again, this is a generalization - I'm sure there are exceptions, but unfortunately this is something nobody measures and includes in their specifications.

Differences in the power compression characteristics of the various drivers within a speaker often cause the tonal balance to change with volume level, with woofers typically suffering from more power compression than tweeters and therefor many multiway systems sound dull at low volume levels and bright at high volume levels as they've been optimized to sound right at medium to medium-high volume levels.

On another note, bass reflex loading only increases efficiency in the region of the port tuning. It does nothing for midband efficiency. The reason why bass reflex speakers are usually more efficient than sealed box speakers is that the driver parameters most suitable for reflex loading include a more powerful magnet system, which is what raises the midband efficiency.

Hope this helps some.

Duke

Audiokinesis,

Very good points about thermal compression issues and how they affect dynamic range and tonal balance. I agree with you fully.

A highly efficient speaker should certainly be better in these regards when listened to at moderate to higher levels.

Do you have a view on how distortion tends to be affected as you increase or decrease efficiency?

thank you for the comments/feedbacks

Here are a couple of links that clearly explain some of the complexity and compromise involved in speaker design.

http://www.soundonsound.com/sos/oct00/articles/monitor.htm

http://www.soundonsound.com/sos/nov00/articles/ustandingmons.htm

Rlew,

I am only concerned that someone reading this thread will think that a high sensitivity speaker is necessarily better than a low sensitivity speaker.

Unlike the general rule about a bigger box having better bandwidth and therefore being a better speaker.....there is no simple rule of thumb for efficiency.

There are indeed advantages in high sensitiivity speakers, as myself and others have pointed out, such as a better dynamic range (less compression), however speaker design requires a balance of compromises and high sensitivity is not always better.

In very general terms, ultra-efficient speakers should be avoided just as ultra-inefficient speakers should be avoided. Both will have strengths but extremes are generally achieved with large compromises in other areas instead of an overall balance in performance.

Let me give a couple of examples of how a manufacturer can achieve high efficiency at the expense of distortion;

Long coil operating in short magnetic gap gives a low cost and highly efficient driver but it increases harmonic distortion as the voice coil operates outside the linear area of the short magnet gap. Also the heat dissapation is poor in these designs....so while they are highly sensitive they do not dissipate heat as well as a shorter coil in a longer magnet gap.

Very light rigid cone diaphragms made from hard/stiff materials (magnesium,ceramic,polymers). These efficient rigid low mass cones have low internal damping and tend to have high Q resonances. This efficient choice of cone leads to higher harmonic distortion than more critically damped designs.