Soundlab speakers with sub woofers?

The right amp is what makes for bass on a Sound Lab. Generally speaking, transistors are not a good match. The reason is the Sound Lab, like most other ESLs, has a 10:1 change in impedance over its range, and is over 30 ohms in the bass, making it hard for transistor amps to make power, and it does need the power.

So if you have a transistor amplifier of 600 watts, it will make about 150 watts in the bass region, which means a 150 watt tube amp will be able to keep up with it no problem.

Set up correctly I've heard these speakers shake the walls and I do wonder if a sub is needed, since they can go to 20Hz all by themselves!

I thought I was pretty clear with my point, but maybe not. Here it is, referring to the math I posted earlier:

1) many solid state amps will not get flat frequency response- if so they will be bass shy and too bright.

Some people with transistor amps have the speaker placed closer to the rear wall than is ideal; in this way they get some bas reinforcement, but its usually a 'one note' sort of bass as the reinforcement from the wall only happens at one frequency. The speaker should be 5-6 feet from the wall for best results.

2) A solid state amp will manifest approximately 1/4 of its rated 8 ohm power specification.

Its not voltage that drives a Sound Lab. Its power. That is why I provided the link at the end of my last post.

In my feeble mind I imagine a relatively fixed voltage across the speaker terminals. As the impedance changes with frequency so will the current drawn from the amp and so will the power supplied by the amp. When the impedance rises in the bass, the current drawn from the amp will decrease as will the power supplied by the amp. Likewise when the impedance decreases in the midrange and treble, the current drawn from the amp will increase as will the power supplied by the amp. I think that's based on Ohm's law.

Hi Bob- your statement here is correct. So the impedance curve of the Sound Lab goes up to just over 30 ohms in the bass, and is about 1.5-3 ohms at 20KHz depending on the setting of the Brilliance control.

So let's do the math. We will assume a constant voltage, and for fun a 600 watt transistor amplifier driving 8 ohms.

Power = Current (I) x Resistance squared. So we need to solve for current.

600 = I x 64, 600/64= 9.375 is the current of 600 watts into 8 ohms. Using Ohms law: 8 Ohms =V/9.375 Amps, we see that the voltage is 75 volts.

Now we change the Resistance to 30 ohms, keeping the voltage constant. So: 30=75/I, solving for I we get 2.5 Amps.

Power is Voltage x Current, in this case the 600 watts is now 187 watts.

At the other end of the frequency range, the amp can put out over 1200 watts, as there is a 10:1 difference in impedance.

Basically what the math shows is that a 600 watt amp can't make that kind of power- and so a 150 watt tube amp can easily keep up with it, as the difference between 150 and 187 watts is not even 1db. If you have a 200-watt tube amp, you would need a transistor amp with about 800 watts in order to keep up.

The math also shows that there is a good chance that the transistor amplifier will be bright on the Sound Lab.

Now the application of negative feedback will cause the amplifier to reign in its power somewhat at higher frequencies. But since the feedback voltage is really not correct on this load, there will be an error that causes the amp to make too much power at the higher frequencies. Its is easily audible.

Now the speaker curve is not based on a driver in a box as we all know. Its based on a capacitor. Its efficiency is thus not a function of its impedance- its about the same at high frequencies as it is at low frequencies. You can see that a constant voltage characteristic in the amplifier is not really all that desirable. You can read more about this phenomena at this link:

http://www.atma-sphere.com/Resources/Paradigms_in_Amplifier_Design.php

Hi Bob, You got it almost completely right in your last paragraph- right up to the last sentence... so, you said

The power dissipated by the speaker will fluctuate throughout the audio band inversely proportional to the impedance.

Which is correct. Now all you have to do is understand that the efficiency of the speaker does not also change- it is the nearly the same at all frequencies.

The Sound Lab has an impedance curve that varies by about 10:1 from bass to ultrasonic (30 ohms down to 3 ohms or less). So if the amp makes more power than it should in the highs and less than it should in the bass, it can't help but to impart a coloration.

On most tube amplifiers its a good idea to try the Sound Lab on the various taps of the output transformer to see which works best.

Some transformers can ring if not properly loaded- so if you are on the 8 ohm tap and the amp is playing into a 30 ohm load on a bass note, you may get increased distortion (lower ordered harmonics) that might cause the amp to sound warmer than is actually correct. It depends on the design of the amp as to whether there will be a significant loss of power with such a mismatch, but IME usually the loss of power is slight. If the amp has negative feedback most of the distortion will be controlled but that can be a double edged sword as with more feedback comes a greater tendency to behave as a voltage source.

With our MA-2 (about 70% of our production in the last 20 years are driving Sound Labs) there is also a slight loss of power into 30 ohms (about 10 watts), but distortion is actually reduced into the higher impedances.

In short, this is a speaker that benefits more than many box designs from a slightly higher output impedance in the amplifier.