OTL amp owners

[Aries Cerat]

In an abstract way,it can resemble the TriodeFet circuits of our Ianus amplifiers. However i would not go that far call it tube OTL amplifier

 

[Ron Resnick]

I interpret the translation to mean that the load (i.e., the loudspeaker) is seeing current delivered to it directly by the transistor power supply.

I think I would characterise this design as a "semi-hybrid" design in that the voltage side is tube driven with a tube output, and the current side is tube driven but with a solid-state output. I think this topology is somewhere between a hybrid amp (or a reverse hybrid amp) and an all-tube amp.

But if the current side is solid-state then is that why no output transformer is required? If so, can we consider this to be an OTL amplifier?

But I am practicing electronics without a license here. (I could not help proffering an interpretation!)

Would the real EEs please step up and solve this mystery for us?

 

[GMKF]

I interpret this to mean that the load (i.e., the loudspeaker) is seeing current delivered to it directly by the transistor power supply.

Yes

I think I would characterise this design as a "semi-hybrid" design in that the voltage side is tube driven with a tube output, and the current side is tube driven but with a solid-state output. I think this topology is somewhere between a hybrid amp (or a reverse hybrid amp) and an all-tube amp.

I'll call this amp a hybrid+ amp. tube voltage, ss current. so T-OTL-SS would be the terminus here.

But if the current side is solid-state then is that why no output transformer is required? If so, can we consider this to be an OTL amplifier?

The powersupply is "acting" as the transformer here. Its generating the current, which is needed to drive a speaker. The transformer would normaly turn high voltage low current into medium voltage medium current. With this setup the powersupply is "generating" the current which is fed into the output stage. Thats how you get around the transformer. The tube is high voltage and the power-supply is high current. Thats how you get the quoted high power output of the amp.

But what the heck is the KT150 doing then ? It doesnt even show up in the diagramm..

Would the real EEs please step up and solve this mystery for us?

Measurements would clear up this mystery propper.
I suspect the engineers of this amp aren't willing us to tell how this amp is working exactly... which is fair.

 

[Aries Cerat]

A hybrid amplifier would use tubes as voltage gain,driving a SS output stage.
Ron, SS are "OTL" (well there are exceptions) but the term is not used for Solid state designs.They do not use OPT as it is not needed.

This design,though it is not wise to comment on other designer's work,does not work like usual hybrid designs.

In hybrid designs,the tube stage only delivers drive to the output stage and not to the load.
In this design,the 300B does deliver current to the load(although it is limited to it's max 50-70mA ) which only is enough for a few milliwats for a normal speaker.The SS section of the output stage ,"sample" the current of the 300B which drives the load,and "boosts" the current.Now,the SS section, does in a way,mimics the transfer curves of the tube,so the distortion and sonic footprint of the amp is leaning more towards the 300B not the SS.

We have experiment in this technology,and a (distant though) member of this technology is the Ianus.

Having designed and lived with many examples of similar tech,it is not a surprise if the Alieno does indeed sound like a tube amp and performs marvelously if.

Cheers

Stavros

 

[Ron Resnick]

A hybrid amplifier would use tubes as voltage gain,driving a SS output stage.
Ron, SS are "OTL" (well there are exceptions) but the term is not used for Solid state designs.They do not use OPT as it is not needed.

This design,though it is not wise to comment on other designer's work,does not work like usual hybrid designs.

In hybrid designs,the tube stage only delivers drive to the output stage and not to the load.
In this design,the 300B does deliver current to the load(although it is limited to it's max 50-70mA ) which only is enough for a few milliwats for a normal speaker.The SS section of the output stage ,"sample" the current of the 300B which drives the load,and "boosts" the current.Now,the SS section, does in a way,mimics the transfer curves of the tube,so the distortion and sonic footprint of the amp is leaning more towards the 300B not the SS.

We have experiment in this technology,and a (distant though) member of this technology is the Ianus.

Having designed and lived with many examples of similar tech,it is not a surprise if the Alieno does indeed sound like a tube amp and performs marvelously if.

Cheers

Stavros

Dear Stavros,

Thank you very much for your analysis and description. I appreciate it! I understand on all points.

Putting to one side how we would describe and characterise the design (I do not think that I, personally, would consider it to be an OTL) it appears that Fulvio Chiappetta has created a truly unique and innovative amplifier design -- and that alone, I think, is a fantastic and laudatory accomplishment! The fact that it also sounds amazing -- and is not merely some design exercise for intellectual curiosity -- makes the accomplishment all the more impressive.

 

[microstrip]

I just spent about four hours using Google translate to translate from Italian to English dozens of posts about the Alieno 250 LTD amplifier on an Italian audio forum. The translated posts did not tell me anything.

But then I found this:

(...)




* * *

Would the electronic engineers here please get to work reading and understanding this description, and explaining the circuit design of the Alieno to the rest of us? Thank you!

IMHO, we are still playing with semantics. The controlled current source will have a sound signature, it is not an OTL in its usual sense used during decades. Probably an interesting design, I can not comment any more without too much guess - we would need technical details.

We had similar issues about the Berning ZOTL's - hundreds of posts debating if they were or not OTLs, the subject is not new.

Perhaps someone can design a tube amplifier block and couple it to the booster transistors using Peter Walker current dumping technique. Then he can call it an OTL! See https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=13&cad=rja&uact=8&ved=0ahUKEwivjPyA3v3YAhUGchQKHVIYAawQFghoMAw&url=http%3A%2F%2Fwww.keith-snook.info%2Fwireless-world-magazine%2FWireless-World-1978%2FCurrent%2520dumping%2520-%2520does%2520it%2520really%2520work%2520DCD.pdf&usg=AOvVaw0g8lMNEb1Sbiw6MaGmaxTQ )Current dumping—does it really work? Theory and practice by J. Vanderkooy and S. P. Lipshitz University of Waterloo, Ontario)

Unfortunately OTL is becoming a marketing key word - all these amplifers want to live on the tradition of great sounding true OTLs.

BTW, please remember that the Aries Cerat Collatio looks like a true OTL!

Please note, it is as far as I know, I can be wrong because information is very scarce and can not be confirmed.

 

[DaveC]

IMO, it's not possible to know exactly as the schematic is not known. But CCS on the output is a type of hybrid design with SS output section even if the tube provides voltage gain. This is NOT an OTL in the conventional meaning, although the term "Output TransformerLess" can apply to any SS amp in a generic way that disregards it's intended meaning and is therefore misleading.

 

[Aries Cerat]

IMHO, we are still playing with semantics. The controlled current source will have a sound signature, it is not an OTL in its usual sense used during decades. Probably an interesting design, I can not comment any more without too much guess - we would need technical details.

We had similar issues about the Berning ZOTL's - hundreds of posts debating if they were or not OTLs, the subject is not new.

Perhaps someone can design a tube amplifier block and couple it to the booster transistors using Peter Walker current dumping technique. Then he can call it an OTL! See https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=13&cad=rja&uact=8&ved=0ahUKEwivjPyA3v3YAhUGchQKHVIYAawQFghoMAw&url=http%3A%2F%2Fwww.keith-snook.info%2Fwireless-world-magazine%2FWireless-World-1978%2FCurrent%2520dumping%2520-%2520does%2520it%2520really%2520work%2520DCD.pdf&usg=AOvVaw0g8lMNEb1Sbiw6MaGmaxTQ )Current dumping—does it really work? Theory and practice by J. Vanderkooy and S. P. Lipshitz University of Waterloo, Ontario)

Unfortunately OTL is becoming a marketing key word - all these amplifers want to live on the tradition of great sounding true OTLs.

BTW, please remember that the Aries Cerat Collatio looks like a true OTL!

Please note, it is as far as I know, I can be wrong because information is very scarce and can not be confirmed.

Hello
The Collatio is a true OTL, and a Single ended one.A DHt tube driving a big bank of tubes running in Class A.
The Ianus us vastly different. It uses our Triodefet tech, which uses a solid state apparatus in parallel with a tube, which mimics the tube transfer curves. I would not call the Ianus OTL amp though. Although the tube does provide drive current to the load.and, even though it does not have a OPT.
Best
Stavros

 

[Argonaut]

Another consideration of the AC Collation that appeals to me is the AC coupled output stage, meaning that in the unlikely event of a catastrophic fault condition no DC can be present at any time at the speaker terminal, not a feature of most OTL amplifier topology.

 

[Ron Resnick]

The Collatio is a true, purebred OTL. There is absolutely no misunderstanding or question about that!

 

[microstrip]

The Collatio is a true, purebred OTL. There is absolutely no misunderstanding or question about that!

Purebred OTL, but not OTL/OCL - the supreme class where the same blessed electrons that flow through the tubes flow through your speaker.

Futterman's also had a capacitor in the output, but were push pull type, not SE.

 

[Aries Cerat]

Purebred OTL, but not OTL/OCL - the supreme class where the same blessed electrons that flow through the tubes flow through your speaker.

Futterman's also had a capacitor in the output, but were push pull type, not SE.

Hello micro.

The capacitor used at the exit is as transparent as a single wire (1mohm ! ESR) ,mind you this cap has with a max surge spec of 12.000A.Pretty much as single wire as can bewith a size to much..It presents much less internal resistance than the contact resistance of a good binding post

It can protect the user's valuable and fragile drivers flying through their grills in case of bad tube(i have seen this happening with an otl amplifier and a ceramic driver speaker,with fine ceramic particles flying through the grill).

Cheers

Stavros

 

[microstrip]

Hello micro.

The capacitor used at the exit is as transparent as a single wire (1mohm ! ESR) ,mind you this cap has with a max surge spec of 12.000A.Pretty much as single wire as can bewith a size to much..It presents much less internal resistance than the contact resistance of a good binding post

It can protect the user's valuable and fragile drivers flying through their grills in case of bad tube(i have seen this happening with an otl amplifier and a ceramic driver speaker,with fine ceramic particles flying through the grill).

Cheers

Stavros

Yes, I knew about it, but probably the tube electrons do not!

Although I think that what matters is sound quality, not the microfarads, we must remember that a 50 000 uF capacitor has an impedance around of 150 mhoms at 20 Hz. Can I ask what is the capacitance of your output capacitor?

We should consider than many audiophiles consider that the best sounding solid state Quad amplifier was the capacitor coupled 303, not the later DC coupled ones.

 

[Aries Cerat]

Yes, I knew about it, but probably the tube electrons do not!

Although I think that what matters is sound quality, not the microfarads, we must remember that a 50 000 uF capacitor has an impedance around of 150 mhoms at 20 Hz. Can I ask what is the capacitance of your output capacitor?

We should consider than many audiophiles consider that the best sounding solid state Quad amplifier was the capacitor coupled 303, not the later DC coupled ones.

Hello micro

Thanks for the talk.
I only stated the specs of the cap to get a glimpse of it's tech.Of course sound is what is sets the criteria,and sonic wise it is worlds apart from the back to back electrolytics used in AC coupled amplifiers.

We use much smaller value than the 50000,in the 6000 ball park.However,this in not a voltage-source amplifier,where the hundred of mohm would waste a good output impedance.The Collatio is a current-source amplifier,making the elevated cap impedance in LF somehow unimportant.

Best
Stavros

 

[analogsa]

Very interesting. Did not see a mention of the output impedance in the specs...are the 6?19? connected as followers? How would a customer determine if this would work with a particular set of horns?

As for the cap coupling, it is a compromise, but then what isn't? Coil design and core material also appear super critical, perhaps close to the importance of an output transformer.

 

[Aries Cerat]

Very interesting. Did not see a mention of the output impedance in the specs...are the 6?19? connected as followers? How would a customer determine if this would work with a particular set of horns?

As for the cap coupling, it is a compromise, but then what isn't? Coil design and core material also appear super critical, perhaps close to the importance of an output transformer.

The cap quality is very important ,though not to the same extend as the importance of the OPT in a SE design. Though cap quality/tech is important,it is really a detail in comparison to other choices. SE vs PP, Class A vs Class AB, feedback vs non, driver circuit, PSU topology, power tubes used. The choice of this super spec cap was not necessary,but we do take the extra mile often.Call it design OCD

An ideal voltage source amplifier would have zero output impedance.That is the reason of the race for lower output impedance in amplifiers,to get as close to the ideal voltage source amp.
An ideal current source amplifier,would have infinite output impedance.However,as you know zero output impedance dos not exists,so infinite output impedance also does not exist.A rule good rule of thumb would be to have the output impedance at least 20x the load impedance.The Collatio is close to this number.However it can be configured in voltage mode when in monoblock mode (about 1ohm output impedance).


Generally speaking,compression drivers do like current amps.Same for some very low Q cone drivers,usually used to drive mid and bass horns.Some transformer driven ribbons also like current drive.
Sealed/BR loaded bass units, and normal Q mid drivers do not like current amplifiers,as they depend on the amplifier for proper dampening.

If you want to give examples,for example the Tune audio Anima,our Symphonia speaker,most of Cessaro horns, AG,are great match. A big Wilson is not


Best
Stavros

 

[analogsa]

Thanks for the explanation! So, a proper current drive, which can be reconfigured as a very acceptable voltage drive if needed.

 

[microstrip]

(...) The Collatio is close to this number.However it can be configured in voltage mode when in monoblock mode (about 1ohm output impedance). (...)

Thanks for the details. Do you use local feedback in the monoblocks?

 

[Aries Cerat]

Thanks for the details. Do you use local feedback in the monoblocks?

Only in voltage out mode,switchable.It is our only design using local feedback anywhere.I never used it with our horns,not needed as non-f sounds better.

 

[Mrmb]

I'm using Atma-Sphere MA-1's with my Soundlab M1-PX stats. It's a superb combination! BTW, Ralph is a consummate gentleman and helpful to a fault. One of the good guys in this hobby.