General Asylum

Posted by Ralph (M) on February 20, 2007 at 09:32:35

In Reply to: I typically agree with Ralph... posted by Sean on February 19, 2007 at 15:21:33:

When the accepted story falls apart, I look for explanations.

Speakers are driven by power, not voltage. All speakers I can think of have a finite impedance which results in a substantial amount of current in the load. Voltage is a component of power.

Now there is the story about speakers being voltage driven, but what we find is that story is part of a larger story called the 'Voltage Paradigm'. In it, we see that all measurements relate only to voltage and rarely to anything else. This paradigm of design and measurement also contains the story that the ideal amplifier can make constant voltage into any load. An example of this is a transistor amplifier that makes 100 watts into 8 ohms and 200 watts into 4 ohms- doubling power as load is cut in half.

What is important to get is that the Voltage Paradigm is only a story and is not universally supported as fact. Here are some examples:

Horns, ESLs and magnetic planars have impedance curves that have nothing to do with box resonance. In fact for such speakers they are expecting the power to be constant regardless of the impedance that they have. In the case of ESLs, which have high impedance in the bass region and low impedance in the highs (often varying by more than 5:1), 'constant voltage' amplifiers will play too much highs and not enough bass. Try a 'constant power' amplifier (tubes- Power Paradigm) on there and suddenly the speaker is making flat frequency response.

In the 45 years since they were introduced to audiophiles, transistors failed to supplant the prior art, the 'obsolete' technology. Why? The first thing to look at is the language, or story, surrounding transistors. Perhaps the idea of constant voltage is incorrect? Indeed, anyone successfully running tube amps knows that you don't need that characteristic to make things work and in fact you are better off without it; that just because a tube amp is not designed with intention to drive loads less than 4 ohms is not a bad thing at all. Conversely transistors do not drive high impedances, and what we see from history is that speaker designers and manufacturers have always gone where the money is. In the 50s there were 32 ohm speakers and 16 ohms was common.

Back in the 1960s, transistors were becoming possible, but (due to being non-linear) needed (and still need) a lot of feedback to make them work. Realizing this, the industry made up the 'constant voltage' story and pitched it to the market, since transistors cost about 1/10th that of tubes to make the same power, but you get to charge just as much. There was a lot of money to be made! The public bought the story, and now there is 40 years of 'drift' away from what is real (BTW we are seeing the same thing happen now with class D, which costs 1/10 the money to make power as transistors cost...).

Some evidence for this is that there are more manufacturers of tube amplifiers in the US now then there was in 1956! Unlike the demise of the flathead engine to dual overhead cams in the same period, tubes have not been supplanted for the simple reason that they work.

So if the story about constant voltage is made up for dollar's sake, then of course the measurements we have for speakers and amplifiers today is not going to carry a lot of weight with our ears: we are measuring the wrong things! We have created a situation where we have made our measurements virtually irrelevant.