Pre-amp suggestions for Thiel 2.4/Pass Labs XA30.5

Gentlemen, I believe that what the autoformers in the McIntosh amplifiers essentially do is that they allow the amp to maintain the very low output impedance of a solid state amp, while relieving the transistorized output circuits of the need to deliver progressively larger amounts of current into speakers having low nominal impedances.

The autoformers apparently do that by causing the solid state output circuits of the amp (ahead of the autoformer) to see the same load impedance when an 8 ohm speaker is connected to the 8 ohm tap as when a 4 ohm load is connected to the 4 ohm tap, and as when a 2 ohm load is connected to the 2 ohm tap. Presumably that is accomplished by stepping down the voltage that is produced by the output circuits (ahead of the autoformer) such that the voltage to the 4 ohm tap is nominally 0.707 times the voltage at the 8 ohm tap, and the voltage at the 2 ohm tap is nominally 0.5 times the voltage at the 8 ohm tap. Based on P = Esquared/R, that will result in the same MAXIMUM power capability in all three situations, while at the same time eliminating the need for the output transistors to have to supply progressively larger amounts of current as the nominal load impedance is reduced. Thus the amps have the same maximum power ratings for 2, 4, and 8 ohm speakers, based on the presumption that the speaker will be connected to the corresponding output tap.

For a GIVEN output tap, though, the variation of the amount of power that will be delivered into varying load impedances is, as usual, dependent on the output impedance of the amp, until the limits imposed by the amp's maximum power capability are reached. The output impedance of the circuit ahead of the autoformer is negligibly small, since it is solid state, and the voltage step down provided by the autoformer will make it even smaller, at least on the 4 and 2 ohm taps. (Transformers transform voltages in proportion to their turns ratio, while at the same time they transform impedance in proportion to the square of that turns ratio).

Note that the MC402 and MC501 have specified damping factors of 40 or more, that number corresponding to an output impedance of 8/40 = 0.2 ohms, which is in solid state territory.

So if a nominally 4 ohm speaker is connected to the 4 ohm tap, and its impedance drops to 2 ohms at some frequencies, for a given input voltage to the amplifier twice the amount of power will be delivered to the speaker at frequencies for which the speaker's impedance is 2 ohms, compared to frequencies for which it is 4 ohms. That does NOT, however, mean that that the amp's MAXIMUM power capability will double down. Current limitations and/or thermal limitations and/or limitations imposed by self-protection mechanisms will limit the MAXIMUM amount of power into 2 ohms to much less than twice the 4 ohm maximum power capability. Otherwise there would be no need for the autoformer.

I note this comment in John Atkinson's measurements of the MC501:

The 8 ohm tap was limited to 225W into 2 ohms (17.5dBW), for example, while the 4 ohm tap delivered 1000W (24dBW), the 2 ohm tap 630W (22dBW).

So as long as the amp is operated within the limits of its ability to deliver power into given load impedances, it will behave like a solid state amplifier. Meaning that the tonal effects resulting from the interaction of its output impedance with variations of speaker impedance as a function of frequency will be similar to those of a typical solid state amp. However, the amp's MAXIMUM power capability will not double down, as it will in the case of many high quality solid state amps.

Hope that clarifies more than it confuses. Best regards,
-- Al

Al, do you think that McIntosh is on too something that other SS folks are not addressing? Obviously they have been doing this for a long time - is it a difference that makes a difference and all towards the positive, or a just a tradeoff like so many other things in audio design?

Hi Paul,

I have no particular insight into that. But I note that the MC402 and MC501 are both described as double-balanced, meaning that each channel consists of two fully balanced amplifiers that are bridged together via the autoformer. I'd expect that a key factor in the choice of design approach was that the hardware required to implement a fully balanced amplifier that could provide the current necessary to support those kinds of high power levels into low impedance loads, using conventional approaches that don't use autoformers, would result in a large increase in their already very hefty size and weight, with cost increasing commensurately.

Best regards,
-- Al

Thanks Al. Interesting how in the example you provided the dBW's go up and then down with impedance halving. Some how I get the sense that McIntosh just fell back on some of the technology they previously used with their tube amps, and that it's a bit of a cost cutting approach, to avoid spending on expensive heat sinks, etc. Interesting that in the decades following Mac's introduction of ss amps, very few (any?) seem to have followed this with approach.

It occurred to me that back when Mac first started building ss amps, the transistors of the time were not nearly as durable as they are today, and perhaps to maintain reliability they used autoformers to keep them from getting stressed. Speakers of the time were probably higher impedance loads designed for the tube amps that were prevalent at the time. Using the autoformers with the those high impedance speakers also permitted them run more power into them.