Come on John; you understand well what I am saying. And you perfectly know that I've presented already on this forum several schematics that are really unusual. The bad thing is, you could not call the name who did something similar 40 years ago, in each other case no matter who presented what every time you commented that it was done by some name many years ago, adding "Learn and grow!"
What I am getting at, I am getting at a creative design process on this forum, and at support of each other, encouraging those who does something new (even if it is new only for him/her), to share knowledge, to share information. It is a diametrically opposite to what you are doing here: putting everyone down and trying to promote yourself as a designer of things that noone here can design, and as an associate of other designers who even don't want to communicate with us.
Well MJ, I am glad you can understand him, I have more difficulty, and I don't know why he is commenting to me about everything that I say.
For example, if I mention the Audible Illusions single triode line stage, it is because I have both listened to it, measured it extensively, and know what many reviewers think of it. There are others, of course, but I have not listened to, measured extensively, or seen reviews of, except, perhaps the CJ line stage that has been well reviewed and is similar in concept.
Should I include virtually everyone who as made an open loop triode line stage? How about a 12ax7 tube open loop output? Would that qualify? After all, the 12ax7 is the most popular tube that I know.
Why not a single bipolar transistor line amp? Should that qualify?
For example, if I mention the Audible Illusions single triode line stage, it is because I have both listened to it, measured it extensively, and know what many reviewers think of it. There are others, of course, but I have not listened to, measured extensively, or seen reviews of, except, perhaps the CJ line stage that has been well reviewed and is similar in concept.
Should I include virtually everyone who as made an open loop triode line stage? How about a 12ax7 tube open loop output? Would that qualify? After all, the 12ax7 is the most popular tube that I know.
Why not a single bipolar transistor line amp? Should that qualify?
Last edited:
Nobody asked you to include any names; but here we have a Tube Forum where DIYers design and build things that you've never saw in any available equipment for sale. You call yourself a teacher, but suppose a teacher every time says, "Good, everyone did a homework, like my nephew did", "Good for you, I give you A, like my nephew got", "You did this math problem; well; my nephew did this problem 40 years ago before you was not born", and so on -- is it promotion of the teacher's nephew, or putting down a class that did homework and got A grade, but no more than a teacher's nephew did? How long such a teacher would stay in the school?
Do you mean that microphone that used Gu-50 tube you've asked Zen Mod about?
I would ask him, may be he has one.
Edit: here you go: http://www.diyaudio.com/forums/showthread.php?p=1930552
Last edited:
I can't seem to get answers to most of my questions, so I will try to answer a few myself. First, if the current sources are NOT low noise gain, they will inject a certain amount of extra noise into the circuit, so while the mosfet noise (which is mostly 1/f in character) will be reduced, the current source could add midrange noise due to the inherent noise of the current source generator and its transconductance.
John.
To put not too fine a point on this and all else you preach, if you claim to be something, or claim to have done something, It's really up to you to support your claims. I've said this consistently. For some reason you seem top have life turned around in your head. You expect others to dig up proof to counter your claims.
You scanned, did you read? I'll give you that, but I am only showing that your statement really does not support your claims. I receive (and read) EDN', 'Electronic Engineering' and a host of others. I almost joined AES in Toronto, I attended a number of meetings, but Mr. Stanley Lipschitz came on the scene about then. AES became a debating club for the math sector - useless expect for their excellent Journel. I was an IEEE member as well. At the time, it did not match my interests well and so I dropped it. So, what's your point? I have witnessed complete morons who belong to each organization, so membership is not limited to intelligent people only. When the SMPTE group began attending AES, the collective IQ dropped to an all time low. This simply due to a couple of their members who were pretty good at halting the exchange of information. It became an elitist waste of time for me and a larger number of other people as well.
What you can do is claim that you are a designer and a consultant, audio being your specialty. If this isn't good enough for you, then looking in your mirror in the morning must be pretty difficult for you.
Okay, now for some interesting viewpoints you have ...
This is the thing that will have me stand against you John. Your complete lack of professionalism and simple manners. But this is the reason you and I disagree.
In closing, I had really hoped you could stay on track this time. It really looked like you might. I said so to people we both know, if you want to check that out. Then, after you decided to take a shot in my direction, I simply said "that all crap broke loose with your response". That, you can ask Stuart if you want. Yes John, I was happy that you seemed to be more responsible to the people in this thread. I wished you well and you flushed it down the drain. I'm going to bet that this may have been a common theme with your relationships with other people over the years. A possible reason you have had "bad luck", your personality.
I'm outta here. Anyone seeing John here, or anyone, abuse a member, just report the post. This type of verbal garbage doesn't belong here. Now John, I don't intend to check back unless called. Just focus your attention on the subject of the Blowtorch and related technologies.
-Chris
I don't speak in code, as in "where the bear sits". If you want to communicate, speak clearly please.
Says you. I have not looked it up. It would be so easy for you to back up what you are saying by simply citing the patent number. And no, you are not an engineer. I haven't a clue why this is so very difficult for you to understand. Unless you have the paperwork that says you are an engineer, you are not. I think it's comical that this means so much to you, because I don't care. It's just that you are claiming to be something you are not. This devalues the title of "Engineer". So, if you are so much on the ball there John, why haven't you ever taken the formality of becoming a real, actual engineer?
To put not too fine a point on this and all else you preach, if you claim to be something, or claim to have done something, It's really up to you to support your claims. I've said this consistently. For some reason you seem top have life turned around in your head. You expect others to dig up proof to counter your claims.
Yipee!
You scanned, did you read? I'll give you that, but I am only showing that your statement really does not support your claims. I receive (and read) EDN', 'Electronic Engineering' and a host of others. I almost joined AES in Toronto, I attended a number of meetings, but Mr. Stanley Lipschitz came on the scene about then. AES became a debating club for the math sector - useless expect for their excellent Journel. I was an IEEE member as well. At the time, it did not match my interests well and so I dropped it. So, what's your point? I have witnessed complete morons who belong to each organization, so membership is not limited to intelligent people only. When the SMPTE group began attending AES, the collective IQ dropped to an all time low. This simply due to a couple of their members who were pretty good at halting the exchange of information. It became an elitist waste of time for me and a larger number of other people as well.
Well, at least you understand that point. What you do not understand is that this in no way diminishes anything you have done in your life. It's just that you are hung up on attempting to elevate yourself above your peers by falsely claiming that you are something you are not. It isn't very pretty John.
What you can do is claim that you are a designer and a consultant, audio being your specialty. If this isn't good enough for you, then looking in your mirror in the morning must be pretty difficult for you.
Okay, now for some interesting viewpoints you have ...
Okay John, I just have to know ... What in the world does one statement have to do with the other? What does 40 years in electronics have to do with you being able to understand what Anatoliy says? Or, was it just an opportunity for you to pump up that fragile ego of yours?
And here we see the quick return to the old John. You were very insulting there, for no reason what so ever. Why John? Did Anatoliy say anything like this to you? Unless it was in the form of a PM, no he did not.
This is the thing that will have me stand against you John. Your complete lack of professionalism and simple manners. But this is the reason you and I disagree.
If you don't understand why that is, try putting yourself in someone else's shoes and imagine being on the receiving end of your own comments. If anyone else treated you as you treat many around you, you'd go stark raving mad with anger. But, you do not consider anyone but yourself I don't think. You honestly don't see what you are doing, do you?
So, was it the reviews that turned you on to it? Being that this is a tube based design, and that you normally do not design with tubes, what point are you attempting to make? Not that many have heard this preamp, I have and my impression is that it's okay. Not the equipment for my tastes, by then it's not anyone's musical taste we are talking about here. I was under the mistaken impression that you'd be talking about the Blowtorch and related designs. You have gone off topic all by your little lonesome.
That's nonsensical for starters. You have exaggerated, I guess to make a point. I mean, a 12AX7 running with not feedback is not the intended use of that tube. Like any part, you are supposed to apply it correctly. A real engineer would know that.
I don't understand where you are going with this. Are you asking for proof to go up against your claim? I mean, that is your job John. It's up to you to cite the references that prove you did what you say you did. It sure isn't up to us to prove you didn't, you have the roles people play reversed a little bit there John.
In closing, I had really hoped you could stay on track this time. It really looked like you might. I said so to people we both know, if you want to check that out. Then, after you decided to take a shot in my direction, I simply said "that all crap broke loose with your response". That, you can ask Stuart if you want. Yes John, I was happy that you seemed to be more responsible to the people in this thread. I wished you well and you flushed it down the drain. I'm going to bet that this may have been a common theme with your relationships with other people over the years. A possible reason you have had "bad luck", your personality.
I'm outta here. Anyone seeing John here, or anyone, abuse a member, just report the post. This type of verbal garbage doesn't belong here. Now John, I don't intend to check back unless called. Just focus your attention on the subject of the Blowtorch and related technologies.
-Chris
Where I live, you know, Silicon Valley California, the title 'engineer' is given to people who can do engineering. It is true that in some states, and even other countries, this might not be so, but here, if you can do the work, you are given the title and the salary that goes with it. 25 years ago, when I worked with Humphrey Instruments, in San Learndro, Ca, a company with about 300 employees, my title on my business card supplied by the company was: Senior Design Engineer
This should not surprise anyone, as that is what I did at Humphrey Instruments. I did not repair, or assist other engineers, I did my own projects as part of a team making a precision laser system for medical use. Now that was 25 years ago, since then, I became VP Engineering for a startup company, then Senior partner in my own Company, Vendetta Research. I am not trying to make anything special out of the definition 'engineer' except that you have to be able to engineer new designs, often with tech support, who can often solder better, yet they do not get how the circuit really works. That is the difference between a technician and an engineer, here in California. I hope that this is clear to everyone.
This should not surprise anyone, as that is what I did at Humphrey Instruments. I did not repair, or assist other engineers, I did my own projects as part of a team making a precision laser system for medical use. Now that was 25 years ago, since then, I became VP Engineering for a startup company, then Senior partner in my own Company, Vendetta Research. I am not trying to make anything special out of the definition 'engineer' except that you have to be able to engineer new designs, often with tech support, who can often solder better, yet they do not get how the circuit really works. That is the difference between a technician and an engineer, here in California. I hope that this is clear to everyone.
Attachments
Is this a voice of reason, in the chaos? Thanks a lot for contributing, Pavel.
Of course, one of the real advantages of jfets is the low noise, due to the fact that you do not have to degenerate with resistors.
The complementary jfet design was possible back about 1970 or even earlier, by using the first relatively complementary jfets that were made by Motorola. Unfortunately, these parts were long gate devices, and had relatively low Gm. Still, they could have been used, and I used them, myself, at first. Then, the 'breakthrough' came when Siliconix introduced the J series of fets in 1973 or so. It was easier to get complements that were both higher Gm and lower noise. These parts were used with the Mark Levinson JC-2 design.
About 1978, the first complementary jfets came out of Japan, along with V-fets that were, in some ways, really amazing. Hitachi came first, then Toshiba, and others. Toshiba completely dominated the market until recently, when they just pulled the plug, for the most part. I hope that serious experimenters loaded up on Toshiba jfets for future designs.
Fortunately, Linear Systems is trying to make equivalent fets to the discontinued Toshiba series. They are aware of the potential demand, once they get their processes completely in order. In fact, they might have complementary devices available today, but they are relatively expensive compared to Toshiba of a just a few years ago.
Unfortunately, the 3 designs posted here, that are associated with some design that I have made, either in the past, or in the present, are completely dependent on complementary fets, so many will find it difficult to copy them, in future, except for personal prototypes for home use.
Of course, one of the real advantages of jfets is the low noise, due to the fact that you do not have to degenerate with resistors.
The complementary jfet design was possible back about 1970 or even earlier, by using the first relatively complementary jfets that were made by Motorola. Unfortunately, these parts were long gate devices, and had relatively low Gm. Still, they could have been used, and I used them, myself, at first. Then, the 'breakthrough' came when Siliconix introduced the J series of fets in 1973 or so. It was easier to get complements that were both higher Gm and lower noise. These parts were used with the Mark Levinson JC-2 design.
About 1978, the first complementary jfets came out of Japan, along with V-fets that were, in some ways, really amazing. Hitachi came first, then Toshiba, and others. Toshiba completely dominated the market until recently, when they just pulled the plug, for the most part. I hope that serious experimenters loaded up on Toshiba jfets for future designs.
Fortunately, Linear Systems is trying to make equivalent fets to the discontinued Toshiba series. They are aware of the potential demand, once they get their processes completely in order. In fact, they might have complementary devices available today, but they are relatively expensive compared to Toshiba of a just a few years ago.
Unfortunately, the 3 designs posted here, that are associated with some design that I have made, either in the past, or in the present, are completely dependent on complementary fets, so many will find it difficult to copy them, in future, except for personal prototypes for home use.
Last edited:
This would qualify as being pretty close to it, I think, isn't it?
It's the Linear LT1363, nice piece.
Ciao, Tino
Attachments
Yes, a nice picture. I would call it, a peace of an art. Somebody had both drawing talent and love to Diamond Buffers. However, some additional phase lag on an inverting input does not look nice...
I don't understand what you're saying. Where should the additional phase lag come from, compared to a "traditional" input diff pair? From cob of the "biasing transistors" in the diamond buffer? But they are there also in the noninverting input.
Last edited:
Flat wrong. The one and only true advantage of JFETs when it comes to noise is that they lack current noise, being therefore excellent for high impedance sources. Otherwise, JFETs are in any respect worse (in terms of noise) than bipolars.
It is also flat wrong that JFETs don't need source degeneration for decent performance. From an AC perspective, both JFETs and bipolars need degeneration for any decent performances. Yes, JFETs are self biasing, and have a parabolic transfer function (vs. exponential) but this is already another story.
For same range of usable input differential voltage, you will get lower voltage noise wit JFETs like 2SK170 or better 2SK369, compared to bipolar degenerated to same Gm. Another advantage is that you can use much higher Id with JFET. Last and not least, parabolic transfer function, as you have already mentioned, and low current noise as well.
You are shifting the discussion and introducing new elements.
JFET devices have much larger voltage noise compared to bipolars. JFETs are more convenient to use and are good for high impedance sources. Low gain (as much as you understand that what really matters is the S/N ratio) and parabolic transfer function are irrelevant to the noise performance.
The closest I am aware of is the multi-tanh principle:
J. C. Schmook, “An input stage transconductance reduction technique for high-slew-rate operational amplifiers,” IEEE J. Solid-State Circuits, vol. SC-10, pp. 407-411, Dec. 1975.
Gilbert, B., "The multi-tanh principle: a tutorial overview," Solid-State Circuits, IEEE Journal of , vol.33, no.1, pp.2-17, Jan 1998
Thanks Bob, for your input. Yes, the LT1362. in many ways, is the bipolar equivalent of the CTC Blowtorch topology. It sounds good too. Bob Crump, my deceased business partner, modified a number of designs using conventional IC's with the LT1362, or another member of the same family. It is considered by 'us' the hi fi gurus, as being one of the best parts in the business.
Now, how quiet is a jfet, compared to a low noise transistor? Well, in truth with any source impedance over 500 ohms, the fet wins the day.
What is the essential noise of a jfet compared to a quality bipolar transistor?
The jfet is shown to have a voltage noise of about 2/pi Gm
A bipolar transistor is shown to have a voltage noise of 1/2 Gm, due to second stage shot noise referred to the input.
Since the bipolar has a Gm of 40,000 umho's/ma, it would seem a clear winner, but it has other limitations, as we shall see.
Now, the better jfets, might have 40,000 umho's/10 ma, typical, and it can be seen that even a jfet running at 10 ma, would be noisier. However, there are other noise contributions that make the bipolar fit for optimum operation only at relatively low impedances, due to input stage shot noise, which is proportional to base current. This means that IF you put an input pot of any reasonable resistance in front of a bipolar input stage, you get into big trouble easily. For example a 10K input volume control, in front of a differential pair of bipolar transistors would made noiser, especially at 6 dB below maximum pot level, as the best possible maximum source impedance would be 10K/4 or 2.5K and this would tend to make the noise current contribute significantly to the noise equation. Think what happens with a 50K pot, like many Parasound power amps have?
Now, there are other advantages to using a discrete jfet input stage, one being that it is practical to have an input quiescent current of 2ma-10ma or more, with jfets, yet, bipolar devices get really compromised with more than 1ma quiescent current on the input stage. and this means that it is difficult to drive the second stage optimally, without additional active components to supply the necessary drive current, such as emitter followers. Simple, elegant (my definition), designs try to avoid added series parts, and jfets on the input, help to make this so.
The best comp-diff j-fet input stage operating at 10ma might have an intrinsic noise of about 1nV/rt Hz with some minor 1/f noise below 100 Hz or so.
Now, how quiet is a jfet, compared to a low noise transistor? Well, in truth with any source impedance over 500 ohms, the fet wins the day.
What is the essential noise of a jfet compared to a quality bipolar transistor?
The jfet is shown to have a voltage noise of about 2/pi Gm
A bipolar transistor is shown to have a voltage noise of 1/2 Gm, due to second stage shot noise referred to the input.
Since the bipolar has a Gm of 40,000 umho's/ma, it would seem a clear winner, but it has other limitations, as we shall see.
Now, the better jfets, might have 40,000 umho's/10 ma, typical, and it can be seen that even a jfet running at 10 ma, would be noisier. However, there are other noise contributions that make the bipolar fit for optimum operation only at relatively low impedances, due to input stage shot noise, which is proportional to base current. This means that IF you put an input pot of any reasonable resistance in front of a bipolar input stage, you get into big trouble easily. For example a 10K input volume control, in front of a differential pair of bipolar transistors would made noiser, especially at 6 dB below maximum pot level, as the best possible maximum source impedance would be 10K/4 or 2.5K and this would tend to make the noise current contribute significantly to the noise equation. Think what happens with a 50K pot, like many Parasound power amps have?
Now, there are other advantages to using a discrete jfet input stage, one being that it is practical to have an input quiescent current of 2ma-10ma or more, with jfets, yet, bipolar devices get really compromised with more than 1ma quiescent current on the input stage. and this means that it is difficult to drive the second stage optimally, without additional active components to supply the necessary drive current, such as emitter followers. Simple, elegant (my definition), designs try to avoid added series parts, and jfets on the input, help to make this so.
The best comp-diff j-fet input stage operating at 10ma might have an intrinsic noise of about 1nV/rt Hz with some minor 1/f noise below 100 Hz or so.
- Status
- Not open for further replies.