Otl's with AvantGarde Duos?


I am driving my Duos with Wavelength's Venus. Not entirely satisfied ( to lean?) i am considering going for more power (15-30watts). Would like to have some feedback with OTLs & the AvantGarde.
wham1000
The comment that Ralph made in regards to me having admitted that the ZOTL is not an Output-Transformerless amplifier in PF magazine is incorrect. What I stated is something to the effect that the ZH270 is full of transformers, but none are output (or audio) transformers. In fact, except for a small DC-blocking capacitor at the input of the amplifier, the ZH270 is DC-coupled, an impossibility for any audio transformer-coupled amplifier.

David Berning
Hi David, could you then explain the presence of the air core transformer in the output section of your amplifier? Seems to me that a rose by any other name is still a rose. So I guess what I want to know is how the air core transformer shown is not in fact a transformer in the output section.
As I read the patent, it seems to play a crucial role in the design.

Also, you *did* mention in PF some years back when this issue first appeared (in a letter to the editor) that people did not seem to understand the idea of zero hysteresis loss, and found it easier to understand when you told them it was an otl instead. I can dig the quote up for you if you like. Anyone with back issues of PF can find it.

This was a bit after the time that Harvey Rosenberg was promoting your amplifier in his reviews, having negotiated a contract with you to sell your amps. So that would make it in the 1997-1999 period. Harvey told me that at the '98 CES and you told me that the contract had expired at the '98 Philadelphia Triode Show. So I know the timeframe is correct.
Ralph-
The transformers, of which there is a pair for push-pull in the ZH270, are power-conversion transformers that operate at fixed frequency. They have low-loss ferrite cores (not air core) optimized for 250 kHz. They are indeed critical components for the impedance matching in the ZH270. Audio output transformers in conventional tube amplifiers perform this impedance-matching function, and are major limiting factors in obtaining the wide bandwidth needed to prevent the amplifier from having the filtering properties that both you and I are trying to avoid by not using them.

Now to try to get to the bottom of this controversy as to you thinking that the ZOTL has output transformers and my claiming that it doesn't, we need to come to an understanding of what an audio output transformer is. To me, an audio output transformer is a transformer used to couple the audio-frequency signal from the output tube(s) to the speaker, and for high-fidelity applications it is designed to cover three to four decades of frequencies starting from at least 20 Hz. To do this they must be physically large. In the design of these devices, compromise must be made between high frequency and low-frequency performance because of parasitic capacitance of windings and number of turns needed to achieve sufficiently low magnetizing current to reduce core saturation and distortion at low frequencies.

All I have been saying all along is that the ZOTL does not have any such device. I see the term OTL as an acronym for output transformer less; I believe that that was the original intent when the early pioneers such as Futterman had the dream of finding a way around the main limiting factor in audio power amplifiers of the time. Any amplifier that by its design eliminates this problematic device could be called OTL for that matter, including most transistor amps. Amplifiers that are free of these devices avoid the filtering (frequency-dependent) type coloration associated with coupling output devices to the speaker, assuming no other audio-frequency dependent coupling network is used.

The ZOTL falls into a generic class of amplifier described by the acronym OTL both technically and in spirit. The power conversion transformers called out by the patent cannot and do not operate at audio frequencies. The semiconductor devices called out by the patent drive these transformers at a fixed 250 kHz and do not operate at audio frequencies or as amplifiers. The patented ZOTL circuit is a two-way coupling circuit that performs impedance matching like an audio output transformer, but does not have any of the limitations and distortions imposed by audio output transformers. This circuit is in no way a buffer, such as a transistor output stage in a hybrid tube-transistor amplifier. There is no power gain, and more importantly, the speaker “sees” the tube transfer characteristics and the tubes “see” the speaker’s dynamic characteristics. This two-way power flow separates the ZOTL from the hybrid amplifier in that sonically it is pure tube. Further technical details and graphical demonstration can be found on my web site davidberning.com in the form of white papers.

The circuit topology of the ZH270 is completely different as to how the transformers are connected. I simply fail to see how anyone familiar with audio-amplifier circuits can confuse these transformers with audio output transformers. At least one winding of an audio output transformer is connected to the tube(s) in the non-OTL amplifier, and a winding is connected to the speaker. Neither of these connections is made with the conversion transformers in the ZOTL.

While I realized that the earlier OTL amplifier designs, that is pre ZOTL, have wonderful sonic characteristics when used with compatible speakers, I did not want to make OTL amplifiers before I invented the ZOTL because of the high heat and reduced reliability that comes from having many output tubes running in parallel. I also realized that speaker selection was more critical because these tube OTLs are starving the speaker for current. Since the ZH270 was introduced in 1996, it has slowly gained respect; so much so that I can no longer keep up with demand and there is a waiting period that now extends into next year. Clearly more people are finding that the ZH270 can provide sonically what previously only older-type OTL designs could provide, and can do so for a wider selection of speakers and without the excessive heat and fuss of replacing tubes and chasing drifting bias settings.
Hi David, we may disagree on whether on not conventional OTLs have weakness that other arts do not; that is a matter for elsewhere.

You misinterpreted my request for an explanation of the air core transformer's function. I was asking if it is an OTL, how can there be a transformer essential to its function in the output section. The acronym is for "TransformerLess". In plain English, this means that there is no transformer, aircore or otherwise.

So we also disagree on semantics. Even though the air core device you use does indeed allow for DC response, it is nonetheless a defined as a transformer by the electronic texts (any physics prof will disagree with you also). Even though it is supplying the power for the output tube... well, with this simplified description, that happens with a lot of output transformers (although not in the same way). And, if you strip the 250KHz out of the signal that the air core transfers, there is an audio signal that is the resultant. Plus, if you trace from the speaker terminals back through the circuit, you do not find it connected to the elements of a tube (not even through a coupling cap)- it is connected to the tube via a DC to DC converter and an aircore transformer.

This is clearly not an OTL. Technical explanations to the contrary have, for the last 7 years, amounted to obfuscation.

Now the odd thing to me is this: Before you came along, there were two techniques for coupling output tubes to loudspeakers: the transformer coupled way and the OTL way. All tube amps made prior to this invention operated on some varient of the the above two techniques.

Now there is a third way and you invented it! Why, for heaven's sake and for your own, would you not take advantage of the tremendous marketing opportunity that represents??

Of course, I know that you are not referring to me in some of your comments directed at OTLs :) Controlling bias so that is does *not* drift, and providing as much current as a transistor amplifier are two of the things that we do. But I agree that those are problems with (other) current and past OTLs.
Good discourse going on here by two of the audio paragons. For what it's worth, the opinion of the village idiot...

I am the owner of a pair of AtmaSphere M60 OTLs, and have also had the pleasure of being around a Berning ZOTL just over one month ago.

While I will not get into the circuit designs of either, as I have not immersed myself in them. I don't know what is and what is not this or that. I will say that I was more than a bit surprised to find out exactly just how similar both amplifiers sound to each other. And, in the end, is it not really the sound? However close or not the designs are, I have NEVER encountered an amp that sounded so much like my Atmas as the Berning.