what is an ideal tube amp

John also has an Atma-Sphere setup, with a fairly well-optioned MP-1. He has had Atma-Sphere amps since 1989 if memory serves.

I bought a set of T-3s in 1998. There has been no looking back. They have been updated with the dual woofers, new crossover and the beryllium field coil drivers. My room is 17' x 21'. The efficiency of the speaker is more like 98 db, not 105 (that is more typical of his Hartsfield reproduction). Any amp we make works very effectively on the speaker- keep in mind that such amps are what the speaker was designed with.

The S-30 does very well on the T-3!. I tend to play the system louder than most audiophiles as my front end is relatively impervious to room-borne vibration. What this means is that it is very relaxed even when driven well above 100 db, so I can occasionally clip the S-30. Not so with the M-60 and MA-1 :)

If a tube amp with an output transformer has a 16 ohm tap, it will have slightly more power and certainly less distortion when driving 16 ohms. This is because the output transformer is more efficient. In fact it will also exhibit more bandwidth- often with greater bass extension.

With transistors you lose power, but you also lose distortion- causing the transistor amp to sound smoother and more musical. This is because there is a non-linear capacitive aspect to almost any semiconductor junction, which is magnified by current. This aspect contributes to odd-ordered harmonic distortion, which is unpleasant to the human ear. By reducing the current (i.e. going to a higher impedance load) odd-ordered harmonics are reduced. This is evident in the distortion curves of all solid state amplifiers, even class D.

With OTLs, depending on the size of the amp, you may gain some power- in particular with smaller OTLs. For example the S-30 makes 45 watts into 16 ohms. With larger OTLs this may not be true- for example with our MA-1 the output power is unchanged. What is more important is what happens to distortion and efficiency of the OTL! In all cases it will operate with less distortion, meaning that it will sound smoother and more detailed. The increased efficiency also means that you will get less heat with less power drawn from the wall, as more of the power generated by the amp will be dissipated in the load rather than the output section.

IMO there is every argument for 16 ohms **if your goal is sound quality**. There is no argument for 4 ohms regardless of the amp, unless your goal is **increased sound pressure where sound quality is less important**, *and* you have a solid state amplifier. IOW 16 ohms is a simple way of making the speaker seem more transparent, smoother and more musical, as opposed to 4 or 8 ohms.

Charles1dad, take a look at this article, it will help in understanding my explanation that follows:

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

Four ohms came in with the transistor- we all know that. However, along with the transistor came the new Voltage design rule. The rule has been in place since the mid 1970s. IMO the rule has its problems and one of them is this: with it has come the idea that if you parallel two speakers, the efficiency increases.

Now, if that were really the case, it would violate some pretty fundamental rules of physics- and clearly falls into the same area as perpetual motion machines. That is to say, you don't get that increased energy for nothing. What is really happening is that there is confusion between the idea of Efficiency (how much power it takes to drive a speaker) and Senstivity (which is sort of the Voltage rule equivalent, IOW how much voltage a speaker has applied to it for a certain amount of output).

Now into an 8 ohm load the two are the same. For example if you have 1 watt into 8 ohms the voltage is 2.83 volts. But if you apply 2.83 volts to a *4* ohm load you will have 2 watts in the load. People tend to use the two words interchangeably and that is where the confusion lies. Power (1 watt/1 meter) is the older Power Paradigm spec and Voltage (2.83V/1 meter) is the Voltage Paradigm spec.

Now tubes have always been kind of expensive to make power. So to deal with that, in the old days high efficiency speakers tended to be more common, as was 16 ohms. When transistors came in, one thing that was not lost on the industry was the fact that you could build a transistor amp that made the same power as a tube amp, charge about 90% of the tube amp price, but have about 10-50% of the cost of building the tube amp. IOW follow the dollars.

Now this was not lost on the speaker manufacturers. When it was seen that you could double the power of some transistor amps simply by cutting the impedance in half, some speaker manufacturers began making 4 ohm speakers to get that extra output from the amp. You see, building high efficiency speakers is expensive because you have to have precision voice coil gaps to get the efficiency. By going to reduced precision by increasing the voice coil gap, the cost of the speaker went down. But if you went to 4 ohms you got 3 db of Sensitivity "back". Again, for such speakers the industry found that they could charge slightly less than the high efficiency models, but the cost again was dramatically reduced- they made more money.

I think it would be great if this was simply my opinion (one would like to think that the changes we see in audio have to do with increased fidelity rather than just making more bucks) but you don't have to look very far to see that this is what went on. I can name plenty of examples beyond these two and I am sure others can too.

So- if we turn our backs on the money-making schemes and focus instead on how to make the equipment sound better, right away higher impedance in the speakers emerges as a simple solution.