John Curl's Blowtorch preamplifier

[beppe61]

I doubt that even if the right topology is presented here by someone it will be named as such, but I could be wrong of course.

Dear Sir,

thank you for your kind reply.
Of course I did not mean the actual schematic.
I understand perfectly the impossibility of disclosing it publicly.
Anyway I seem to remember as well that Mr. Curl has switched to more simple (basic?) topologies than those used in the past because better sounding.
I find this extremely, and I stress extremely, interesting.
So I was searching for some confirmation here.

Kind regards,

bg

 

[courage]

bg,

This thread is aimed at discussing, what can be considered, John's best design, the Blowtorch, according to me. The dream of most DIY-ers is to try and get as close possible to the original design and that in itself is understandable. It all depends on to what extend John encourages this.

 

[Ultima Thule]

Dear Sir,

thank you for your kind reply.
Of course I did not mean the actual schematic.
I understand perfectly the impossibility of disclosing it publicly.
Anyway I seem to remember as well that Mr. Curl has switched to more simple (basic?) topologies than those used in the past because better sounding.
I find this extremely, and I stress extremely, interesting.
So I was searching for some confirmation here.

Kind regards,

bg

Hi Beppe,

if you wan't to discover the "mysterious" about simple circuits sounding well I suggest you to look up Pass Labs and Mr NP's DIY homepage www.passdiy.com, lot of interesting topic's to read about.
I myself enjoy sometimes reading he's stuff!

And please don't be so formal and thank me, it's not the Gentlemen's Tee Club! just go there...

Cheers Michael

 

[Panelhead]

Parallel fets

Dear Sirs,

I seem to remember some rumors about a single amplification stage made out of more fets in parallel, no feedback.
Something like the Conrad Johnson ART preamp but made instead with fets.

Any info about this.

Kind regards,

beppe

There is a lot to be said for this simple type circuitry. My fave tube linestage uses a sinlge 6080 in a cap coupled voltage amp circuit. Each channel runs at 80 ma. No feedback, except the cathode bypass caps.
This sounded better than any other types I heard using 9 pin tubes. But now I have moved on to play with solid state.
The comment about having power supplies that are more complex than the circuitry show there are two camps here. All my junk has power supplies that are more complex and robust than the actual circuitry.
Maybe someday John will present his actual circuitry for the Blowtorch. Hopefully years after production ends. If not, there will be Blowtorch clones and kits coming from the East before Easter.

George

 

[beppe61]

Re: Parallel fets

1) There is a lot to be said for this simple type circuitry.
My fave tube linestage uses a sinlge 6080 in a cap coupled voltage amp circuit.
Each channel runs at 80 ma.
No feedback, except the cathode bypass caps.
This sounded better than any other types I heard using 9 pin tubes.
2) But now I have moved on to play with solid state.
3) The comment about having power supplies that are more complex than the circuitry show there are two camps here.
All my junk has power supplies that are more complex and robust than the actual circuitry.
4) Maybe someday John will present his actual circuitry for the Blowtorch. Hopefully years after production ends.
5) If not, there will be Blowtorch clones and kits coming from the East before Easter.
George

Thanks a lot for your kind and interesting reply.
1) Very interesting indeed.
One thing that worries me with tubed single stage is the high output impedance, even if a friend of mine has driven a 5K input power amp with a 3,5k output tube preamp with very nice results.
2) I cannot help asking you why.
3) One thing is sure: with a single stage tube all the noise from the PS is "available" at the output to be amplified by the following stages.
So a high quality power supply is of paramount importance.
4) We honestly cannot ask for the actual circuitry.
At maximum we could ask for generic hints like: no feedback, single gain stage, etc.
5) If I were He, I would cancel every mark on chips and potted everythink in not accessible black boxes.
Like the old JC2 I think.

Kind regards,

beppe

 

[jbaudiophile]

Hello, Beppe,

Potting can't prevent any secret (just see oooooold Elso Kwak's experiences on it... When you are BOTH an engineer and a chemist, it helps (eh? Elso ?)... He just used X-rays (clever an approach no ? LOL !!!)).

With time, some grinding preventers were included with epoxy : sand, silex,... All grinding means to grinding tools... Was a real war.... !!! But nowadays, means employed to discover the real parts that are used is so much sophisticated that sometimes engineering is just....reversing another guy clever idea !!!

I was targeting John Curl's work from the very beginning.... And I know fairly well what is his curent thinking... I am on the same wavelength he is !

Sorrry for, guys, but I am now running my own way on John's designs. Some words though : just study the MAN's story, and then come back to his PUBLISHED designs... And at last, just INTENSELY think of what he has got as knowledge with years... He shared with us much more thatn others did ! Nothing that difficult... Just a search and seek game, no ?


Jbaudiophile

 

[beppe61]

Hello, Beppe,
1) Potting can't prevent any secret (just see oooooold Elso Kwak's experiences on it...
2) I was targeting John Curl's work from the very beginning.... And I know fairly well what is his curent thinking...
I am on the same wavelength he is !
3) Sorrry for, guys, but I am now running my own way on John's designs.
4) Some words though : just study the MAN's story, and then come back to his PUBLISHED designs...
And at last, just INTENSELY think of what he has got as knowledge with years...
He shared with us much more thatn others did !
Nothing that difficult... Just a search and seek game, no ?
Jbaudiophile

Dear Sir,

thanks a lot for your kind and valuable advice.
1) Honestly I did not know that.
So it is very hard to hide audio secrets nowadays.
2) I have seen the line stage available on the net of his JC2.
I am not an expert but it seems to me a very nice example of discrete op-amp with a feedback loop, but maybe I am missing something.
3) Good luck then!
4) How true were your words !
Just googling I found the following his words that say a lot indeed about his current philosophy:
" Negative feedback should be reduced or avoided if practical. Preamps sound better without global negative feedback, at least in my experience ".
Very valuable words indeed.
Quite different approach from that of the old JC2.

Thank you very much for your valuable advice.
Kind regards,

beppe

 

[john curl]

Quality discrete solid state design is getting more difficult each year. This is because quality parts are becoming more difficult to buy. I don't really try to hide my circuit designs anymore, but I just don't publish my latest schematics.
One reason is that many people, when they try to copy something, will blame the designer, if they get into trouble.
My circuits are simple, but they require selection and matching of complementary fets. It is difficult to get a successful match, unless you have 100's of fets to chose from. This is not only expensive, but it is almost impossible to get this many parts from scratch.
My Parasound designs are made with similar parts, but the circuits do not require the matching or the Idss range, that my CTC design demands. This is why they can be made in Taiwan by the 100's.
I have found that the most simple designs, that are also optimized for lowest distortion, are best, sonically. This includes triodes, as well as solid state, especially fets. Fets are significantly less linear than triodes, so I have to make complementary circuit paths that help each other, to obtain the same distortion as a good open loop triode. The only real advantage that I have is that I can direct couple both in and out, without caps. Also, my fets don't seem to age (get tired) over the years.

 

[wrenchone]

I have seen the 2SA1316 and 2SC3329 (Toshiba) mentioned as replacements for the Rohm 2SB737 and 2SD786 for things like cascode transistors, since the Rohm devices are now pretty much unobtanium. Finding the Toshiba parts for less than a king's ransom promises to be a real pain. The Toshiba parts have the advantage of higher breakdown voltage (80V vs 40V). Until I get my hands on some, I'll continue to use good old 2N4401s and 2N4403s, as I have a lot of NOS stock. I find the Toshiba 2SA965 and 2SC2235 useful for cascoding when the purpose is mainly to protect some lower voltage devices (like a JFET differential stage, for instance) down below. They have 120V breakdown and are in the TO-92L package, so they can dissipate a little power.

 

[courage]

My circuits are simple, but they require selection and matching of complementary fets. It is difficult to get a successful match, unless you have 100's of fets to chose from. This is not only expensive, but it is almost impossible to get this many parts from scratch.

This means that those attempting to build a - as good as original - copy of the Blowtorch will never reach the sound-quality an original Blowtorch delivers. On the other hand most of us here have never heard a Blowtorch and assume it's the best sounding line-stage there is.

My experience is to never blindly follow the achievements of one person as there may be other ways of achieving better sounding amps. I do not mean it disrespectful, because John encourages us all, through the help of his knowledge and experience, to find, discover and build a design which gives us individually the most listening pleasure. In the end that's what it's all about, right?

 

[Justcallmedad]

My experience is to never blindly follow the achievements of one person as there may be other ways of achieving better sounding amps. I do not mean it disrespectful, because John encourages us all, through the help of his knowledge and experience, to find, discover and build a design which gives us individually the most listening pleasure. In the end that's what it's all about, right?

Absolutely agree with you !

What is interesting in these forums and particularly here with the presence of M. John Curl it is the advisability of learning enormously, to progress and maybe one day to do as well or even better . Sharing the experiences appears to me a very important point for all of us to go further in the audio reproduction fidelity, it’s possible for a majority of us, but for people involved professionally and in commercial competition we can easily understand that they cannot reveal their latest designs. What also appears very significant to me is to leave an heritage to the future generations on these particular and specific techniques concerning the audio, a little different from traditional electronics (there is very little literature on the subject).
For those who just want to carry out a quality preamp I think that, on this forum and on all the others, there is plenty of excellent examples, the simplest one could be a simple fet follower (dual or complementary) preceded by a switch and an attenuator with a good PSU.

Ambiguity here comes from the name itself of the forum, some people here seek to simply build their hi-fi equipment as a hobby, what is very respectable, while the others are curious about sound reproduction quality criteria, want to know more, and to learn.

So M. Curl, for those who are eager to learn, and to understand don't be angry if sometimes we abused. I dare say that trying to know some more about your design from some photographs could be very exciting for some people around here as they look simple at first sight (or “stripped” rather than simple). They don’t know all that a particular design itself is nothing without the knowledge, or says differently, 2 extremely different circuits designs could be similar in qualities.

 

[beppe61]

Quality discrete solid state design is getting more difficult each year.
This is because quality parts are becoming more difficult to buy.
I don't really try to hide my circuit designs anymore, but I just don't publish my latest schematics.
One reason is that many people, when they try to copy something, will blame the designer, if they get into trouble.
My circuits are simple, but they require selection and matching of complementary fets.
It is difficult to get a successful match, unless you have 100's of fets to chose from.
This is not only expensive, but it is almost impossible to get this many parts from scratch.
My Parasound designs are made with similar parts, but the circuits do not require the matching or the Idss range, that my CTC design demands.
This is why they can be made in Taiwan by the 100's.
I have found that the most simple designs, that are also optimized for lowest distortion, are best, sonically.
This includes triodes, as well as solid state, especially fets.
Fets are significantly less linear than triodes, so I have to make complementary circuit paths that help each other, to obtain the same distortion as a good open loop triode.
The only real advantage that I have is that I can direct couple both in and out, without caps.
Also, my fets don't seem to age (get tired) over the years.

Dear Mr. Curl,

I would like to thank you sincerely for your extremely kind and valuable disclosing of your enormous expertise and knowledge.
I have always found your posts exceptionally interesting indeed.
If I understand well for a open loop single gain stage a triode should be the active device of choice for linearity and distortion (maybe it is the same thing but I am a beginner ad not even one of the smartest actually).
You go through all the hassle of using fets (that need to be matched) and much more complex topologies in order to avoid the coupling caps.
Then my question is: are caps (even the best example available) so bad and if so in which way? do they add distortion?
I ask you this because there are a lot of so called "ultimate" preamps around that are capacitor coupled actually.

Thank you so much again for spending your time and disclosing your truly great knowledge here with us.

Kind regards,

beppe

P.S. please excuse my poor English.

 

[jbaudiophile]

Hello,

Beppe, JustCallMeDad, Darry, Courage, and some others,...

I fully agree with you all... And if I need a kit, even an audio one, best it ro go to the corner shop, just pay, and drag it home, no ? Next solder all things together.... Quick, good, but... We are NOT all satisfied with because it lacks some self-tailoring, and also some special soul put in it, no ?... Just the same kit the neighbour build some weeks ago... It works, but is nothing special...

On the forum, we can well range into 3 groups : the professionals ones, that take their livings from it (John Curl amongst them...) - the real DIYselfers that only want to discuss a choice or any improvment - and a third one : the audio "students", glued in a twilight zone between academic theories and the audio sages experience...

Well... Even if we cannot afford a Blowtorch (for most of us...), we get all of us a real proud to let others know that something of the best audio thinking is in the breadboarded thing on your workbench that is said to be a Blowtorch clone (hope that you all have a large smile here !), made with off the shelf vintage parts (don't laugh here ! I have tons !!!) bought at a time when they were the best ones or the best available...

Then, with time, and family life, most parts keep quiet in a corner of both your mind (and your garage, perhaps...)... And now you have much ideas and time to achieve them... Go ahead, guys ! I am also in this third group, like others, and no University courses still answer my questions....

We have some real teachers here, and most of us will NEVER go commercials... And I am happy when John Curl says not DO but still also DON'T... Still a student I am...

All the best, guys

Jbaudiophile

As a general remark : we can have access in our respective countries to papers not available elsewhere (most are just too old, or not reprinted, or having them is money group engaging (IEEE, AES,...)). I would suggest the DIYforum to have a new "collecting papers" new rubric.. Would help all of us, and cross our knowledge and findings too.... Not possible to purt copyrighted papers in it ? Yes ! I fully understand it... But to put anothert guy asking for Email coordinates is absolutely diferent, no ?


Just my 2 cents...

 

[john curl]

For the record, I don't like to use caps in the audio path. They are too darn expensive, and they still give problems. I can direct couple, so I do. That's called 'elegant' design.

 

[Justcallmedad]

Speaking about fets, and specially about the 2SK170/2SJ74 -2SK389/2SJ109 pairs, I notice than using them at low Vds, lower than 12V give the best results, even M. Borbely uses them in the saturation region, I guess around 1V in his fet/fet cascode configuration, can you enlighten me on this subject? Igsx?

 

[beppe61]

For the record, I don't like to use caps in the audio path.
They are too darn expensive, and they still give problems.
I can direct couple, so I do.
That's called 'elegant' design.

Dear Sir,

thank you very much indeed for your kind and extremely valuable reply.
I have no instruments at hand but I am pretty sure that "any" coupling cap causes some modification of the signal waveform.

Kind regards,

bg

 

[Upupa Epops]

...caps in audio path...

Strictly taken, caps in PS are also " on this way "... And are you using only fullrange speakers ?

 

[beppe61]

Re: ...caps in audio path...

1) Strictly taken, caps in PS are also " on this way "...
2) And are you using only fullrange speakers ?

Dear Sir,

thanks a lot for your kind and valuable reply.
1) I am aware of that.
I tested the influence of caps on sound playing with a DIY power supply for a little line preamp.
2) Yes. I decided to stay full-range after all.
Even renouncing to the very low bass (< 50Hz).

Thank you very much.
Regards,
bg

 

[courage]

John,

What is the actual voltage the circuits are fed with?

 

[Elso Kwak]

Enough!

Hi JBaudiophile, without trying various circuits and listening test you cannot say anything about the schematic!
I have built line-stages with various FETs and phonostages too. The latter with and without servo. Son of Ampzilla has no servo right from the factory! I don't like the sound of servos. Servos sound bad! Power supplies crazy!
Now what's so special about this? As output bipolar transistors sound better than those FETs. One Sony CDP had a buffer with MOS-FETs as outputs, my buffer with AD711 and bipolars sounded better!
You guys really have no idea how many circuits I have built and tested for sonic signature....THERE I got my knowledge from, NOT from cloning JC's circuits.
I hope you now understand why I find the Blow-Torch thread a little foolish?
Of course all my personal opinion and experience! No offence meant to John Curl.