Which 300b amplifier for Devore O/96

Ralph, thanks for the extremely informative post.  I hadn't previously realized how much the sonic characteristics that are commonly attributed to the various directly heated triodes, in SET applications, are the result of the requirements each tube type imposes on the design of the transformer.

Best regards,

-- Al

@essrand Thanks for providing the update. Assuming a listening distance of say 10 or 12 feet, and a medium sized room, and given that the speakers are not planars or line sources, and factoring in a few db of "room gain," according to my calculations the 8 watts that can be provided to each of the 92 db/1W/1m speakers (as indicated in my earlier post) by your Franks can produce a maximum SPL at a centered listening position of about 97 to 100 db.

So given those assumptions your findings don’t seem to me to be surprising. But given the distortion vs. power level characteristics of SETs, which Ralph (Atmasphere) has described and which from a technical standpoint make sense to me, the bottom line would seem to be that when playing recordings having the kind of dynamic range you described (about 32 db) you are not hearing the Franks at their best.

I’ll add that while a dynamic range of 32 db is no doubt a good deal greater than the dynamic range of a considerable majority of recordings among the various genres, there are some that greatly exceed that amount. I referred earlier to the wide dynamic range that can often be found on classical symphonic recordings. In a few such cases, on labels such as Sheffield Labs and Telarc, by examining waveforms of the recordings on a computer using a professional audio editing program I have found dynamic ranges of as much as 55 db! Correspondingly, at my 12 foot listening distance SPLs produced by those recordings cover a range from about 50 db to about 105 db. I would feel safe in saying that the combination of an 8 watt SET and 92 db/1W/1m speakers could not handle such recordings at reasonable average volume levels, say 75 db or so.

Just some food for thought. It has long been my belief that a major reason for the divergent opinions we often see about power requirements is differences in the dynamic range of the recordings different listeners listen to. As well, of course, as differences in individual volume preferences, listening distance, room size, etc.

Good luck, however you decide to proceed. Regards,

-- Al

Atmasphere 5-15-2019
The O/96 is a great speaker; 2dB less efficient than my Classic Audio Loudspeakers.

Ralph, if I recall correctly your speakers are 98 db/1 watt/1 meter, and the O/96 is indeed spec’d 2 db less than that, at 96 db. However John Atkinson’s measurements in Stereophile showed the O/96 to have a sensitivity of only 91 db/2.83 volts/1 meter. For its 10 ohm nominal impedance that would correspond to about 92 db/1 watt/1 meter.

Which would seem to give further credence to your (and JSautter’s) advocacy of more power than most SETs can provide.

Essrand 5-15-2019
I feel like my Coincidents were powerful enough. But now you are making me doubt that. Will put those amps back in, listen more and get back.

To state a fairly obvious point, I would just suggest that you make a point of including recordings having particularly wide dynamic range in your assessment. Such as many recordings of classical symphonic music, if you listen to that type of music. For a given average volume, recordings of music having narrow dynamic range and/or recordings that have been engineered with a lot of dynamic compression are of course less likely to reveal the issues Ralph and JSautter have referred to.

Good luck. Regards,

-- Al