Atmasphere amps trustable?

"Retro" would be a much more apt term for the look of Ralph’s amplifiers than "weird." And they are intentionally designed with a retro appearance, partly as a matter of aesthetics, and partly (as I recall Ralph stating here in a past thread) because it keeps the costs of the amps lower than they would be with other approaches to their physical design.

As others have alluded to his designs are highly regarded world-class performers, when paired with appropriate speakers. They can also be used with speakers having impedance characteristics that are low or otherwise difficult by using a "Zero autoformer" at their outputs. (Which contrary to misconceptions that are sometimes expressed is not likely to negate much if any of the benefit provided by their output transformerless design, IMO).

As others have alluded to above Ralph is a regular contributor here, and is a completely class act.

Wolfie, as is often the case your post provided the best laugh of the day :-)

Regards,
-- Al

Stevecham 8-4-2016 10:34 pm edt
... the ~4 Ohm output impedance still seems to a limiting factor; it's difficult to apply the ten-to-one minimal nominal speaker to amplifier impedance ratio often cited in order to maintain flat frequency response.

Many and I believe the majority of tube amps of all kinds will not meet that criterion.  For an 8 ohm speaker it would imply a damping factor of 10 or more.  Most high quality tube amps seem to have damping factors somewhere between 8 and 2 (with some being even lower than that), corresponding to effective output impedances of between 1 and 4 ohms or so, respectively.  (Audio Research and McIntosh tube amps are among the minority having DFs greater than 10, just a bit more in the case of most ARC designs and as high as the low to mid 20s in some McIntosh designs).

As you recognize, the 10:1 ratio you referred to will assure minimal variation of the amp's output voltage as a function of variations in speaker impedance over the frequency range.  However meeting that criterion is neither necessary nor desirable in many cases.  Assuming that the particular speaker does not require a higher damping factor for woofer control, meeting that criterion will be unnecessary and/or undesirable in the following cases:

1)The impedance of the speaker varies relatively little over the frequency range.  In that situation the amp's output voltage will vary relatively little as a function of frequency even if the ratio is much lower.   

2)The impedance of the speaker is relatively high at all frequencies.

3)The design of the speaker is such that the variations of the POWER it receives as a function of speaker impedance variations over the frequency range should be minimized for best sonics, rather than variations of the VOLTAGE it receives.  See Ralph's paper on "Paradigms in Amplifier Design."  While such speakers are in the minority, there are many such choices that are highly regarded.

4)Cases where the load impedance seen by the amp can be raised to a suitable level via a Zero autoformer.

Regards,
-- Al
 

Thanks for providing that input, Roger (Rlawry).

It wouldn’t surprise me if a contributor to the problems you had with the Panthere amps was indeed an inability to deal with the heat produced by the 6C33C tubes, which as you alluded to run extremely hot. They consume something like 40 watts per tube just to heat the filaments, and considerably more when bias currents and plate dissipations are taken into account. The 6AS7G power tubes Ralph’s designs use consume about 16 watts per tube to heat the filaments.

Also, for the record, a minor correction: The Panthere was produced by Fourier Components. Fourier Systems was a speaker company that existed during the 1980s, with Peter Aczel of "The Audio Critic" publication as its president. I’m pretty certain that there was no relation between the two companies.

Best regards,
-- Al