Greetings Fellow Tubaholics,
My name is George, and I have been tube-free for...
well, I *haven't* been tube-free. ; )
As further introduction--I am a relative newbie
looking for a DIY amplifier project of about 50WPC and
fantastic sound. By "fantastic", I mean those
subjective tube euphonics of openess,sweetness,
and holography.
As I have searched about, my eye has come upon
the designs of one Menno Vanderveen (or van der Veen ),
a proponent of new topologies and torroidal OPTs.
I am intrigued by his new PP designs, particularly
his Super-Triode, as described in this article on the plitron site.
This design uses both Ultralinear and cathode feedback,
and recommends SV6550C tetrodes.
As far as I can see, the potentially controversial aspects
are the new topology, and torroidal transformers. I
have no insight into the topology, but it seems to me
that the torroid offers extreme linearity and bandwidth,
at the expense of sensitivity to DC current saturation.
However, the design offers DC adjustment, so is it true
that this would not be a problem to the conscientious
tweaker?
What little comment I have heard on his designs
is positive, but I would like to get your advice
before I leap.
So I would greatly appreciate your opinions, insights
and preferences. Do you think this is an excellent
design? Why isn't it more popular? What DIY 50WPC
DIY design would you recommend to a builder seeking
excellent sonics?
Although this is my first post, I have enjoyed
reading your friendly and insightful posts. Thanks
in advance for your help.
Happy Tweaking and Listening!
George Ferguson
My name is George, and I have been tube-free for...
well, I *haven't* been tube-free. ; )
As further introduction--I am a relative newbie
looking for a DIY amplifier project of about 50WPC and
fantastic sound. By "fantastic", I mean those
subjective tube euphonics of openess,sweetness,
and holography.
As I have searched about, my eye has come upon
the designs of one Menno Vanderveen (or van der Veen ),
a proponent of new topologies and torroidal OPTs.
I am intrigued by his new PP designs, particularly
his Super-Triode, as described in this article on the plitron site.
This design uses both Ultralinear and cathode feedback,
and recommends SV6550C tetrodes.
As far as I can see, the potentially controversial aspects
are the new topology, and torroidal transformers. I
have no insight into the topology, but it seems to me
that the torroid offers extreme linearity and bandwidth,
at the expense of sensitivity to DC current saturation.
However, the design offers DC adjustment, so is it true
that this would not be a problem to the conscientious
tweaker?
What little comment I have heard on his designs
is positive, but I would like to get your advice
before I leap.
So I would greatly appreciate your opinions, insights
and preferences. Do you think this is an excellent
design? Why isn't it more popular? What DIY 50WPC
DIY design would you recommend to a builder seeking
excellent sonics?
Although this is my first post, I have enjoyed
reading your friendly and insightful posts. Thanks
in advance for your help.
Happy Tweaking and Listening!
George Ferguson
What new topology, I can't see anything that hasn't been done before. In fact most topologies using tubes has been tried before and for the ones that hasn't been used much there is a very simple reason they dont produce a very good result, something to have in mind wjhen seeing a brand "new" tube based circuit. For the topology in this case it isn't new it has been tried before and you can ask why it hasn't been more successful?
I wouldn't, that is a wouldn't recommend any 50W amplifier but instead something simpler with lower output power. For 50W I probably would recommend to build something based on KT88 or 6550 in ultralinear coupling but not before the buillder had gathered a lot of experience building several other amplifiers and got an understanding how choice of components and circuit influence the end result. It doesn't matter how good the the concept is if the final result is not base on a thorough understanding how the circuit works in all detail.
Regards Hans
Hi Hans!
Thanks for taking the time to respond, and
share your insights.
Yes, that's exactly the question I asked.
Vanderveen asserted that tube design
innovation was cut short by the advent of the transistor,
and its interesting to hear another point of view.
Can you name any ultralinear-with-cathode-feedback
designs, either DIY or commercial? It would be interesting
to compare the designs to Vanderveen's, and see what
people thought of them.
I agree that would make things easier, but it is not
an option in this case--this amp will power some
inefficient speakers, and 50WPC is about the lowest
acceptible power. Otherwise I would gladly go SET.
We're thinking along the same lines. What are
your favorites?
That approach has its merits, though I would rather build
something that is actually useful to me, apart from the learning.
In my defense, I have sucessfully constructed a preamp project,
and am in the process of building a Marchand tube crossover.
Yes, in terms of design I am a Newbie, but have a technical background, and am studying.
Actually, I do not think that at this point I need a deep
understanding of components and topology, though I
certainly hope to learn. That's why I am asking for recommendations of a proven, great sounding design.
I've heard good things about Vanderveen's amps,
and am intrigued by their ability to educate me--the
design I referred to is configurable in 8 topologies,
which could be quite entertaining and instructive.
Also, the Ultimate KT88 design that Tone Loc mentioned
seems a very proven design, and could be instructional.
So can you Hans, or anyone else suggest a great sounding,
proven design of 50WPC or more?
Thanks for taking the time, and happy building
and listening!
Best,
George Ferguson
Thanks for taking the time to respond, and
share your insights.
Yes, that's exactly the question I asked.
Vanderveen asserted that tube design
innovation was cut short by the advent of the transistor,
and its interesting to hear another point of view.
Can you name any ultralinear-with-cathode-feedback
designs, either DIY or commercial? It would be interesting
to compare the designs to Vanderveen's, and see what
people thought of them.
I agree that would make things easier, but it is not
an option in this case--this amp will power some
inefficient speakers, and 50WPC is about the lowest
acceptible power. Otherwise I would gladly go SET.
We're thinking along the same lines. What are
your favorites?
That approach has its merits, though I would rather build
something that is actually useful to me, apart from the learning.
In my defense, I have sucessfully constructed a preamp project,
and am in the process of building a Marchand tube crossover.
Yes, in terms of design I am a Newbie, but have a technical background, and am studying.
Actually, I do not think that at this point I need a deep
understanding of components and topology, though I
certainly hope to learn. That's why I am asking for recommendations of a proven, great sounding design.
I've heard good things about Vanderveen's amps,
and am intrigued by their ability to educate me--the
design I referred to is configurable in 8 topologies,
which could be quite entertaining and instructive.
Also, the Ultimate KT88 design that Tone Loc mentioned
seems a very proven design, and could be instructional.
So can you Hans, or anyone else suggest a great sounding,
proven design of 50WPC or more?
Thanks for taking the time, and happy building
and listening!
Best,
George Ferguson
Hi,
Well, it's not a big name brand but it has an excellent reputation amongst the " I don't care about AR, give me music" crowd. The name is Audiomat and in the Solfege amplifier they used both UL and CF feedback windings.
Maybe in some other models aswell, I am not sure....
So far, I've never, ever seen a DIY amp commercially available using CFB windings.
This is probably due to the fact that most of these kits use popular, off the shelf OPTs that are readlily available from the major winders.
In the range from 30 to 60 Watt the Quicksilvers are still hard to beat.
While it's not hard to copy as such, the endresult will be much dependent on the iron used.
With a little luck it could possibly surpass the original QS iron, who knows?
Has anyone tried yet?
Cheers,
Well, it's not a big name brand but it has an excellent reputation amongst the " I don't care about AR, give me music" crowd. The name is Audiomat and in the Solfege amplifier they used both UL and CF feedback windings.
Maybe in some other models aswell, I am not sure....
So far, I've never, ever seen a DIY amp commercially available using CFB windings.
This is probably due to the fact that most of these kits use popular, off the shelf OPTs that are readlily available from the major winders.
In the range from 30 to 60 Watt the Quicksilvers are still hard to beat.
While it's not hard to copy as such, the endresult will be much dependent on the iron used.
With a little luck it could possibly surpass the original QS iron, who knows?
Has anyone tried yet?
Cheers,
I'll have to dig up Crowhurst's book to see if he showed it there, but I think he did. And I think that the Luxman 3045 might have used it, too.
Back in the late '70s, a popular mod for the Dynaco equipment was to use the secondary as a cathode feedback winding, like in the Audio Research pentode amps (cathodes connected to 16 ohm and common, 4 ohm tap grounded), but retaining the UL connection. Tertiary windings weren't really an option for the amateur in those days for lack of availability. But in any case, it was certainly UL-wit-cathode-feedback.
For What its worth, I have concerns with the circuit listed in an "super triode" configuration.
Simply stated, the "super triode" uses local feedback to obtain tube gain simular to an equivelent circuit using triodes rather than pentodes. Ok.
Triodes need AB2 in order to put out maximum power that the transformer / power supply / tubes can deliver. Since the "super triode" has simular gain, it would need simular drive.
The circuit is clearly AB1, due to the lack of direct coupled driver or interstage transformers that would allow for positive grid current.
Bottom line, I would not choose this design to copy unless you
are willing to live with 25 w or can re-design the amp for ab2 if necessary. However, within its limitations, it may sound wonderful. It just doesn't look like a mature design, more like an engineering concept to loosely follow.
Just my opinion, and I have not done the math.
Cheers;
Doug
Simply stated, the "super triode" uses local feedback to obtain tube gain simular to an equivelent circuit using triodes rather than pentodes. Ok.
Triodes need AB2 in order to put out maximum power that the transformer / power supply / tubes can deliver. Since the "super triode" has simular gain, it would need simular drive.
The circuit is clearly AB1, due to the lack of direct coupled driver or interstage transformers that would allow for positive grid current.
Bottom line, I would not choose this design to copy unless you
are willing to live with 25 w or can re-design the amp for ab2 if necessary. However, within its limitations, it may sound wonderful. It just doesn't look like a mature design, more like an engineering concept to loosely follow.
Just my opinion, and I have not done the math.
Cheers;
Doug
Hey Fdegrove,
Thanks for the tip!
I followed up on it, and it received at least one glowing review.
Here it is , in case someone is interested.
I suspected as much, which is why I am intrigued by advances in technology that allow DIYers to build designs with components that weren't available off the shelf in the 50s.
Well, you may be have a treat in store, for a minor winder makes OPTs for these designs--Plitron in Canada, Amplimo in the Netherlands.
It's not an off-the-shelf kit, but
here are Plitron/Vanderveen's build instructions, and here
is one of the transformers the design can use. Note the
"Cathode feedback tap"
Amplimo sells the same OPTs in your neck of the woods, under different model numbers.
(Although I must look like a salesman for these products, I have no relationship with these companies. I am just an excitable geek.
)Thanks for the info Fdegrove. If you, or anyone else, happen to check out the Plitron or Amplimo site, I'd be very interested in your assessment.
By clicking the "site map" link on the Plitron site, one can see a bunch of articles on how to use these transformers.
Again, many thanks, and Happy Tweaking and Listening!
Best,
George Ferguson
Hey DougL,
Your concerns are very worthwhile to me, for it is "sanity checks" from experts (certainly more expert than me
) that I am seeking.Does it make any difference that, to my point of view, "Super-Triode" is something of a misnomer? It seems to me that the circuit in question is in fact a pentode in Ultralinear configuration, with cathode feedback? Does that ease your concerns in any way?
If not, what about the fact that this design can be reconfigured as a traditional Ultralinear (circuit 2)? It seems to me that, if the Super-Triode circuit does not work, as a fallback I could configure the amp as a traditional Ultralinear. Looking at circuit 2, are you comfortable with that design?
Thanks for your message, and for any other insights you can provide.
Best,
George Ferguson
The first time I saw this was in the Swedish DIY magazine "Populär Elektronik" from the early sixties where they described inventions and ideas together with DIY projects.
I am not very positive towards authors who claim that they invented something that was actually tried long before, it seems also to be common to invent a flashy new name together with the presentation of the new invention. There are quite many examples of these "inventors" around in the tube community today which is unfortunate because beginners can easily be tempted trying something that not only isn't new but has been rejected by many other designers earlier due to several valid reasons.
No I dont think we are actually, I recommended to start with something easier then a 50W UL amp. Anyway as Frank says iron is very important, I would start from that end and then decide what amplifier to build. The circuit that Tone Loc mentioned seems to be very standard and unspectacular and could be a good starting point even if dont agree on some of the component choices, also as I live in Japan and have experience of Tango and Tamura transformers I would probably choose another transformer instead of the Plitron but that is my own personal choice.
Regards Hans
Thanks for crushing my unrealistic expectations.
Speaking only for myself, MV appeals to this newbie because he has a design he likes, and is willing to explain why he likes it and how to build it. I'd love to get similar information on a better 50+WPC design, but maybe that's another thread?
Best,
George Ferguson
Hi,
About a year ago I did dig also in this circuit and it intrigues me also. Local cathode feedback in the output stage has the possibility of doing without overall feedback. Overall feedback has the disadvantage that driving stages run out of steam (saturate) near clipping and you need extra loop gain.
Why UL-CFB is not done IMO for DIY is the extremely critical OP trannie. The cathode windings need to be very tight coupled to the anode windings otherwise you will run for sure into stability problems. That is not an easy task to solve and asks for well thought out winding arrangement. It looks MvdV has solved this by using a toroidal core which allows much tighter coupling than EI cores.
What keeps me off building it actually is in the first place the extremely high price of these trannies from Plitron/Amplimo. There are also other pitfalls. Toroidal trannies are extremely sensitive to unbalance. A few mA unbalance will rise distortion dramatically. Also the UL-CFB topology will need huge driving voltage for the output tubes although this doesn’t need to be a real problem. Btw Biasing penthodes in class AB2 is not an option with the tubes at hand (6550‘s / KT88’s).
As others already said, I wouldn’t recommend this for a first major power amp project. But look for a more proven concept like straight UL/triode configuration. It will be much cheaper too...
Cheers
About a year ago I did dig also in this circuit and it intrigues me also. Local cathode feedback in the output stage has the possibility of doing without overall feedback. Overall feedback has the disadvantage that driving stages run out of steam (saturate) near clipping and you need extra loop gain.
Why UL-CFB is not done IMO for DIY is the extremely critical OP trannie. The cathode windings need to be very tight coupled to the anode windings otherwise you will run for sure into stability problems. That is not an easy task to solve and asks for well thought out winding arrangement. It looks MvdV has solved this by using a toroidal core which allows much tighter coupling than EI cores.
What keeps me off building it actually is in the first place the extremely high price of these trannies from Plitron/Amplimo. There are also other pitfalls. Toroidal trannies are extremely sensitive to unbalance. A few mA unbalance will rise distortion dramatically. Also the UL-CFB topology will need huge driving voltage for the output tubes although this doesn’t need to be a real problem. Btw Biasing penthodes in class AB2 is not an option with the tubes at hand (6550‘s / KT88’s).
As others already said, I wouldn’t recommend this for a first major power amp project. But look for a more proven concept like straight UL/triode configuration. It will be much cheaper too...
Cheers
My suspicions confirmed- this "invention" is shown in Crowhurst's "Understanding HiFi Circuits" from 1958. Crowhurst shows UL cathode followers, too. The biggest disadvantage, besides the O/P transformer requirements, appears to be the astonishingly high level of drive required. The AR-style UL-cathode-feedback (the first time I saw this was in Ike Eisenson's "Audio Modification Manual" from 1978), which only has about 6 dB of feedback, still requires a doubling of drive voltage. So anything you gain in the output stage with respect to distortion, you lose from the driver.
It does seem to be a fashion these days to take an old circuit which had a niche, dust it off, give it a fancy name like "Super-Triode" or "Enhanced Triode" or "Parafeed" or whatever, then flog it as the latest and greatest discovery. Whatever happened to the idea of giving credit to the original inventors, or at the very least labelling something a revival of an old idea?
It does seem to be a fashion these days to take an old circuit which had a niche, dust it off, give it a fancy name like "Super-Triode" or "Enhanced Triode" or "Parafeed" or whatever, then flog it as the latest and greatest discovery. Whatever happened to the idea of giving credit to the original inventors, or at the very least labelling something a revival of an old idea?
I was / am concerned that the driver stage will have to swing plus & minus 40 Volts or so to get the output you need with all the local feedback. That is getting close to the bias voltage.
Yes. Running the output in either cathode feedback (only) or ultralinear (only) would greatly ease the drive requirements.
AS long as you are awair that the circuit presented was somewhat generic and needed tweeking and perhaps more, nothing more needs be said.
I found a published design AB2 with 6550's. However, it may not be practical.
Norman Koren "The Emperors new Amplifier"
Cheers;
Doug
Hi Doug,
Did some calculations for the Plitron trannies and 2 x KT88. With fixed AB1 bias at –50V you need 120V – 140V pp swing to drive the output tubes for 45W output. Such a swing is not easy with low distortion.
To the bashers : Since these UL_CFB trannies are commercial available for DIY, I suppose MvdV did do his homework well. And in his book he is not claiming that this topology is entirely new.
Cheers
Did some calculations for the Plitron trannies and 2 x KT88. With fixed AB1 bias at –50V you need 120V – 140V pp swing to drive the output tubes for 45W output. Such a swing is not easy with low distortion.
To the bashers
: Since these UL_CFB trannies are commercial available for DIY, I suppose MvdV did do his homework well. And in his book he is not claiming that this topology is entirely new.Cheers
Pjotr said:Hi,
About a year ago I did dig also in this circuit and it intrigues me also. Local cathode feedback in the output stage has the possibility of doing without overall feedback. Overall feedback has the disadvantage that driving stages run out of steam (saturate) near clipping and you need extra loop gain.
Why UL-CFB is not done IMO for DIY is the extremely critical OP trannie. The cathode windings need to be very tight coupled to the anode windings otherwise you will run for sure into stability problems. That is not an easy task to solve and asks for well thought out winding arrangement. It looks MvdV has solved this by using a toroidal core which allows much tighter coupling than EI cores.
What keeps me off building it actually is in the first place the extremely high price of these trannies from Plitron/Amplimo. There are also other pitfalls. Toroidal trannies are extremely sensitive to unbalance. A few mA unbalance will rise distortion dramatically. Also the UL-CFB topology will need huge driving voltage for the output tubes although this doesn’t need to be a real problem. Btw Biasing penthodes in class AB2 is not an option with the tubes at hand (6550‘s / KT88’s).
As others already said, I wouldn’t recommend this for a first major power amp project. But look for a more proven concept like straight UL/triode configuration. It will be much cheaper too...
Cheers
As several others have said also, the criticality of the OPT is one of the reasons that these designs are rather rare. With the Plitron/Amplimo toroid core OPTs for these designs, they are becoming possible. These are very complex, state of the art xformers. It may be interesting to know that the senior audio transformer designer for both Plitron and Amplimo is no other than van der Veen. And you have to believe me on my word, but this guy KNOWS transformers. I think these designs, with the given transformer type, will perform well.
Jan Didden
/not associated with any of the names/companies
agreed
If this is one of your first projects, I would also recommend something simpler for starters. Gets rather complex.
___________________________________________________
I wouldn't jump to conclusions as to what designs are good and which are not, as all aspects of audio design haven't been fully explored and tested.
For instance, there are nearly an infinite number of OPT variations that can make the sound bass heavy or bright sounding, or in between. Finding one that is really flat sounding is a challenge. May need a custom made OPT for flatest response? This could change conclusions of what is a good design and what isn't?
If this is one of your first projects, I would also recommend something simpler for starters. Gets rather complex.
___________________________________________________
I wouldn't jump to conclusions as to what designs are good and which are not, as all aspects of audio design haven't been fully explored and tested.
For instance, there are nearly an infinite number of OPT variations that can make the sound bass heavy or bright sounding, or in between. Finding one that is really flat sounding is a challenge. May need a custom made OPT for flatest response? This could change conclusions of what is a good design and what isn't?