Wow, the more manufacturers create high quality mosfets, the more people leave bipolars 😀
Vladimir, I think you haven't posted your comparison between this as a preamp for your modified follower (I will use the light version) versus your high voltage SIT. I hope this one will win. I think it will.
Why not Bonsai? 😉
I know that you don't need technical answer and explanation from Vladimir as I know you know electronics very very well. Better than him I guess.
But can you be more open minded? I'm saying this because I have found typical issue while reading your posts since many years ago. Can you believe that one person can have superior "hearing" than others?
If you believe that everyone have the same hearing ability, I believe I will find again what I have found here in the future.
If you believe that everyone does not have the same hearing ability, now I'm confused 😕
Hello, Jay
Indeed, I still have not expressed my opinion about comparison of high-voltage preamp and the present circuit.
As for my friends, quite experienced ones, it seems that they prefer present schematics. Its not a strange, No NFB, very fast, Zout less than 2 Ohms (the High-Voltage preamp has Zout nearly 300 Ohms). But the last also sounds perfect, may be its a matter of personal preference.
Exactly. In my assumption I actually have applied my understanding of "the preference of the majority" and the most probable system setup.
From my experience PF99 is almost impossible to drive. The low Zout of the preamp will ensure better driving (This might fulfill my needs or "preference"). To be honest I have been confused, how Goran can be satisfied with the PF99 and his tube amp 😱 But like you said, its a matter of personal preference.
Hello Jay,
I see you are surprised how im satisfied with my tube preamp and PF99 🙂 well, Jay first its so subjective so it doesnt worth to talk about. I said i like how PF99 sounds without any preamp. My tube preamp is far from perfect, can you tell me what preamp will be good for PF99? Ill tell you, Pf 99 doesnt need any preamp. It just need efficient high quality speakers. I have tried many topologies, solid state push-pull, tube based push-pull, tube based single ended, im in this for 25 years, so i have some experience. PF99 is the best topology i have heard ever. The modified PF which Vladimir designed could be better.
Jay all is subjective.
What you prefer? What topology? Maybe i can learn something about good sound from you.
Best regards
Goran
Hello Goran,
I can understand if you like the PF99 very much. It has something that is not available in any other amps I have built. The best in one aspect. When Vladimir mentioned about "naturalness" I think he was talking about this very aspect. Tho with my very limited electronics knowledge I prefer to call it "low distortion/noise".
When you prefer to listen the PF99 direct from your CD player, that's because this "strength" of the PF-99 is not masked or ruined by the distortion from the preamp. Lesson learned here is, in my believe, to get an ultimate system one must be very careful to choose the preamp, as distortion/noise may come from active devices, power supply, external interference. I have been looking for a simple class-A preamp with no success. Tube preamp will provide to high output impedance to drive solid state, single ended mosfet will provide too high Ciss.
But the power supply for this preamp must be perfect. Choke, followed by shunt regulation, that's my intuition for the best possible scheme. Unfortunately, I don't know how to choose the proper inductance. The master said (if I understand it correctly), little is better than nothing. But that is not true. There is a proper value. I have tried to search the theory but found none.
You are right, if speaker is easy to drive, then it will solve the problem. In my case, I have tried a sensitive crossoverless speaker with no success. And this kind of speaker have their own problems (I will not opt to use them).
Thanks to Vladimir who come up with solution. Not only to this particular good sounding amplifier, but for a system as a whole.
I have seen one possible criterion in the literature for choosing proper choke inductance value, it sounds like "criterion of uninterrupted current through the choke". Qulitatively, one can imagine the following: with simple CRC filter, diodes are conducting short periods only. If one put R or L just after the bridge, before the first C, the relative period when diodes are conducting becomes longer. It is believed, that the L value must be increased until the diodes become conducting all the time (one pair of diodes, then another pair).
Full range speakers are not solution for PF99. The best full-range driver always need a tweeter too, and extremly complicated enclosure to reach acceptable frequency responce.
You said simple aproach is not the right solution, and in the same time, you said that implementing an preamp (means more active components) is not good also. So, what is that "golden" aproach?
Other question, with all respect to Vladimir, what is the soultion he came up? Shunt PS or cap-cluster modification?Or HF power transistors? Did you build his PF or his SE preamp? Did you heard how it sound?
I don't know electronics theory but I know what might work and what not. You may not believe me if I say that I know how this circuit will "sound" before I build it. Actually the circuit doesn't create sound, just like our tongue doesn't create voices (by itself).
I hear sound (and sound differences) and always look for the cause of that sound. I hear "bad" sound from IRF610, I change to IRF510, I change to IRF530 and many others, I checked at the datasheet, oh, the smallest Ciss sound the best, the highest Ciss sound the worst, I generate "temporary" conclusion. People say, transconductance per Ciss! Okay, I found 2 candidates with equal fT/Ciss, the one with lower Ciss "won", I drew temporary conclusion. And many many other conclusions.
--->> Experience of being able to hear something, and being able to relate it to something.
I understand all the debates. No one is right or wrong. It's just... how can I say it? If you want to open your mind, answer this: what is your best experience with sound? When? Get this first, then find out what causes such experience.
All of us has heard multi billion dollar systems. We have enough experience, no doubt. Just define one situation where you have the most enjoyable experience. When you heard the sound of the PF99 direct from your CD player. Is this really better than when you heard burmester system in a hi-end show?
When you have the answer to that question, believe me, you will have PRIORITY on which is more important than which. And if you can relate further with the "cause", you will know.
I'm reluctant to say what I know. Do you have any idea why?
Because we already have too many information around us. Problem is, some are "false" and some are "true" but no one knows for sure which ones are the "true"/"false" ones. People will weight the correctness of the information based on the credibility of the person who say it. Do I have some credibility to you? If not, then why do you need to listen to my opinion?
Even the masters make mistakes, and they still are. Every masters have something in their hands. You just need to collect the "true" ones from each of them. But you need to have the ability to draw conclusion. It will be easier after you can answer: what is it you are looking for?
My friend said "...still looking for the holy grail..." How can he find it if he cannot define it. Today he build Dayton system, next week he build Vifa system, next month he build Scan-Speak system. And suddenly he refer to the $$$ for his understanding of good sound??
Okay, thanks Vladimir, that explains something. Now, what happen when the choke is put before the bridge? Does the same thing apply? Can an arbitrary common value of L be chosen then iterate on the value of C?
I believe that the contribution of the power supply, even with the shunt regulator, to the sound is still too high. If this can be isolated, I believe it would be like pulling a blanket from a speaker. Only with such a strong believe, one would try to explore.
I can accept the fact that you can hear the difference between changeing different components as you mentioned, of an circuit. But i think that circuit does create the sound. Will You agree that Push-pull topology doesnt create his own sound with nasty cross-talk distortion? Or that SE topology sound doesnt suffer from influence of output cap and higher output impedance? I think that circuit has influence on the sound. But i agree that components may have even bigger influence.
My best experience with sound is always live performance of acoustic instruments 🙂.
Jay, as I said, PF99 gaves me the more natural sound, closest to live performance (this is my personal opinion).
What is more interesting..it sounds good with every speaker i have tried, from 500 to 5000 euro. So, circuit have the influence on the sound.
I have to admit that PF99 gaves the best results with acoustic music, jazz and..so....
Not the best choice for heavy metal, or techno.
I would really like if you answer to my question, whats the topology you prefer? As you said how we can recognize a "Holy grail"?
BTW, i respect your opinion and im trying to learn something from this conversation, and in that manner im also asking what makes you surprised that im satisfied with PF99? 🙂
About Vladimir's designs i can only say that i like them a lot, just i have to build them first and listen how they sound. I like a lot his approach and according me its worth to follow his threads. Maybe the best threads on this forum.
Best regards
Goran
I can ten times say AGREE with your conclusion. In my vision, important is to use most perfect Current Source part of the shunt PS schematics, with low parasitic capacitances, in order to stop HF noise propagation to the main schematics. Maybe some fine tuned filtering could be added. But I would not put an emphasis on chokes (its my personal preference), they are difficult to build and are not free from parasitic capacitances.
Hello Vladimir,
What do you think about regulated power supply implemented in the PF99 circuit?
Bad and good sides?
Hello, Goran
I use it without changing, it plays a role of "ripple eater", nothing to do with shunt PS idea. But, in the original PF99 schematics, the need in shunt PS is replaced by the current source and shunt-like load connection to active transistor, this gives similar effect.
But, if I put a standard common-source front stage with a resistor load, it will not sound without shunt PS, which I implemented in the schematics of No NFB Headphone/Pre Amp.
More precisely, with 4-diodes rectifying bridge, and LC filter, it is recommended to take L (Henry) = U(V) / [3 x 3,14 x f(Hz) x I(A)]
where U(V) - voltage to the load, I(A) - load current. If we have to provide 40V and 2A to class A schematics, the choke will be
40 / [9,42x50x2] = 40 mH (for air core choke, it will come to several kg copper wire and hell of work, if not the air core, one must be experienced to keep the right gap width in the steel core).
At this, we also must avoid possible resonance and ensure that 1/sqrt(LC) < 2*3,14*f
Last edited:
Even resistors create their own "sound". Also if you remove one diode from your system, you may not be able to hear any sound. All components work together in a complex way to create sound. How, in what way exactly each component may affect the sound is almost a mistery. Knowing the effect only but not knowing the relative severity is often not usefull.
There is almost no reasonable way to say that component has bigger influence than circuit, it always depends on many things. Usually it depends on the WEAKEST LINK in the system. If we want the most improvement in our system, find the weakest link, fix it, then go find the next weakest link.
I don't really understand theory. But I tried to relate what I hear/experience with these terminologies.
Impedance, is just about how to transfer the energy effectively from one stage to another stage. In any given situation, it can be critical or it can be not. For high-Zout amps, mostly it depends on the load/speaker. For "ultimate" system where speakers must be prime, these amps can be a problem.
Crosstalk, I'm not sure which crosstalk or even what it is. If what you mean is crossover distortion in class-B (push pull), it is no mistery. It can be seen in oscilloscope and can be heard easily. Can the "push" and the "pull" be joined together perfectly? With great care may be, but lets say 99% perfect, who care? I know that my experience told me that class-B never achieve what class-A can. But I also know that this small difference may be insignificant.
Output Cap, does influence sound. No doubt. Does a certain supply cap being in signal path or not, I don't care, that's for theorist, I just know when it affects the sound or not. Single ended with output cap may (and usually) sound better than push-pull without output cap. But this has nothing to do with the cap. The cap itself does give bad effect to sound. Whether serious or not it depends on the system. When the cap become a bottleneck, you have to fix it, by changing the circuit, or do what Vladimir has done. He invented NOTHING. He just know that the extra $$ spent for the caps may/will worth the effort.
Yes, that's partly because this kind of sound is very easily be reproduced by any amp-speaker system. Some system (such as many tube systems) are very close to the "holy grail". But many of them fail miserably when reproducing certain music. Will you prefer only listening to country songs, patricia barber, etc etc? Me, no.
Yes, circuits influence the sound, to be precise, they RUIN the sound. But we need the circuit to translate small signal information. If we can do it with minimal impact, then we're done. But how much we asks from these circuits will determine the level of the impact. Everyone have their own preference/priority.
Running a follower direct from a source, you will get a very pure sound, no doubt. Whether it is "sufficient" or not, that's the question. We have to set the minimum requirement that we want to have in our system. Getting the purest first 3 watts and feeding the signal to the most efficient high quality horn driver is probably your plan to get to the "holy grail". But who's going to design the horn system? Many ways to go there, but you must know what you are after (and how to do it is the biggest challenge).
There's little update in circuit topology. There are topologies used by successful amps, there are topologies that have never been successfull in implementation. Some can "hear" the difference, some cannot. Some knows which one he like, some confuses.
For me, amps are so advanced (in this internet era) that they are usually not a bottleneck (weakest link) in a system. It is just about simple preference, not critical at all. Any good amps can bring me to the "holy grail".
Imagine why you always go back to the PF99. "Life performance of accoustic performance" you said? We have to learn from the tube guys. They are very fanatic. Even after listening a multi dollar SS system they will happily go back to their system and be sure with what they want.
In my observasion, and importantly my own experience, nothing can beat the experience of listening to a system where you can feel like you are standing in front of the singer, can see where the drum is, where the clarinet is, etc. etc (but not the singer on top of your face 😉). This kind of experience (with some variations) can happen to many tube guys, but so rare to the ss guys because tube guys use different speakers. There is real problem with using big/complex speakers as done by SS guys.
The good news is, this "holy grail" has nothing to do with the price of the system. Not less than often an expensive solid state system (that uses complex speaker system) fail to recreate this soundstage. Speaker design is the one to blame.
If we don't know what we are after (and be focus to it) we may be drifted from the track. We want more transparency, we want more deep bass, we want more resolution, the more we chase these parameters (which usually require $$$) the farther we are from the ability to recreate a proper "soundstage".
Always start with the speaker, or the room, because your listening room size will determine the drivers etc etc. This is very complex. But you have to have a speaker that can make themselves dissapeare in order to create this soundstage.
Speaker design is still a "mistery" (difficult). But once everyone knows how to do it, the weakest link is the driver. You have to have a low distortion driver. It doesn't have to be expensive. Expensive drivers give you resolution, but if the distortion is high, it will be difficult to make it dissapeare. You will have a "hi-end" sound, but not the "holy grail" you are after. It's different.
Fiva XT25 is known to be able to make itself dissappeare. That imho should be prioritized than the edginess etc etc ascociated with it. Many expensive tweeter have to be put in special baffle to be able to make it dissappeare. Not in wide rectangle baffle like most speakers.
Every cone material has sound character. Different sound character between drivers will not blend as easy as drivers with similar character.
That is just a few things that doesn't require much thinking to spot. There are a lot more. The science behind speaker design is even more complex than the science behind amplifier design.
Yes, I like it also because he addressed issue that should be addressed. I mean, not everyone is doing what I THINK should be done. Everybody know the bad effect of Ciss, but not everyone know the LEVEL of the severity. Addressing this issue with better mosfet is something that should have been done since the launch of the Aleph series. And many more, not only this particular issue. Vladimir doesn't invent new things here, he just know what should have been done and how to do it.
Those "experts" who know enough about electronics, have good hearing skill, intelligent, willing to share his knowledge, are treasures. Especially if we know who they are.
About this new PF99 from Vladimir (along with the preamp), may be it will be the last class-A design that I will consider for my "ultimate" system. For simple amp like this, you have to pay extra attention to detail, and can be very costly. It is possible that my choice of mosfet will not make it. I have never had a "courage" to try expensive caps for the output cap, but now I remember that I can use caps from computer motherboard. IBM computer motherboards usually use good caps. I have a few modules that I took from a server MB that uses a few OSCONs, but still not enough for stereo.
Edit: Oh no! They are small voltage ones 🙁
Last edited:
Thanks Vladimir! I don't understand where the first formula come from, but doesn't it dictat that only one single value of L should be used? That's difficult to implement.
But for such a high current preamp, I think I'm agree that improving the shunt regulator, even with expensive transistor, may be more feasible. Two amperes, eh? Let me think if it is too big 😀
Jay, 🙂 if you carefully read your posts, you;ll find out how contradictory they are. First you are claiming that circuits doesnt affect sound, and in the next post you are writting the oposite.
No matter, Yes the output caps are important, i just ordered some caps to implements cap-cluster on my follower. And Yes, i have feeling that speakers, amp, preamp, all dessapears when im listening the vocals. You can believe me or not, your choice. In this moment im using 4700uF/63V Spragues for the output caps, bypassed with WIMA 16uF and some no name big polipropilens 22uF/400V, yesterday i bought about 40 peaces of 1000/uF/35V to try cap claster. Those caps are very cheap and i will use them just to see if there will be some difference.
Im not looking for perfect sound, just sound what will satsfied my ears.
Im not missing the heavy metal or techno (that kind of music sounds awfull on PF99)
I also play acoustic guitar and its my reference how the setup should sound with acoustic instruments, i just like those kind of music, thatway im satisfied with PF99 - acoustic guitar, in fact all acoustic instruments sounds so natural, but the best point are the vocals.
You can notice that my english is awfull but im doing my best to keep conversation with you which i really enjoy.
Best regards
Goran
No matter, Yes the output caps are important, i just ordered some caps to implements cap-cluster on my follower. And Yes, i have feeling that speakers, amp, preamp, all dessapears when im listening the vocals. You can believe me or not, your choice. In this moment im using 4700uF/63V Spragues for the output caps, bypassed with WIMA 16uF and some no name big polipropilens 22uF/400V, yesterday i bought about 40 peaces of 1000/uF/35V to try cap claster. Those caps are very cheap and i will use them just to see if there will be some difference.
Im not looking for perfect sound, just sound what will satsfied my ears.
Im not missing the heavy metal or techno (that kind of music sounds awfull on PF99)
I also play acoustic guitar and its my reference how the setup should sound with acoustic instruments, i just like those kind of music, thatway im satisfied with PF99 - acoustic guitar, in fact all acoustic instruments sounds so natural, but the best point are the vocals.
You can notice that my english is awfull but im doing my best to keep conversation with you which i really enjoy.
Best regards
Goran
Grammar/language aside, there can never be contradictions 😉
Circuits do affect sound (even resistors). But with so many good circuit, the difference is not critical. There are good single ended amps, there are good push-pull amps, etc etc.
Of course I believe you. With vocal, PF99 is probably the best. That's why I have been looking for the "retouch" to this PF99, as is doing by Vladimir.
And acoustic guitar sound is really easy to reproduce. You just need to pay attention to the "ambiance" (not so sure with the proper term 😉). This is the micro detail that may easily be lost due to bad speaker crossover (phase error that kills the detail). I myself prefer to use piano recording as reference. Bad speaker cannot reproduce them well, as it is a blend of wide frequency sounds.
But don't forget, this is my observation, that in general, sound/music enjoyment comes from listening to vocal and bass. Bass is not only drums, but (often in low SPL) low frequencies associated with many instruments.
And sometimes you cannot avoid certain sound/frequency that will "distort" the music (not to mention bad recording).
My ears are so sensitive. I cannot stand to hear distortion of speakers commonly used by tube guys.
Tho the output cap effect to sound is annoying, they are not really big problem to the ears. They just take up some details from the music.
Try smaller amount of (electrolytic) caps first, and see if you miss the low frequencies.
About WIMA 16uF paralleled with MKP 22uF, I'm not so sure, I have to "listen" to it. Often it is not good to parallel caps that have the same high frequency "area of work".
Vladimir will tell you the minimum voltage rating for the output cap required for his amp (if haven't already).
I see,
About ambient, with my setup i can even notice what clothes the singer wear 😉, no seriosly, there are huge amount of details and space between instruments and arround them.
As i said the Follower what Vladimir designed can sound better, but for me its to expencive aproach..maybe in the future will be different.
Im very interested about his NO NFB preamp. It will be my next project. Yes he was so nice to answer to all of my questions. I dont want to bother him asking all the time about trivial things.
But if ill have any problem durring building the preamp i would feel free to ask him for the solution.
Are you going to build something now? I mean do you have a plans to build Vladimir's designs?
About ambient, with my setup i can even notice what clothes the singer wear 😉, no seriosly, there are huge amount of details and space between instruments and arround them.
As i said the Follower what Vladimir designed can sound better, but for me its to expencive aproach..maybe in the future will be different.
Im very interested about his NO NFB preamp. It will be my next project. Yes he was so nice to answer to all of my questions. I dont want to bother him asking all the time about trivial things.
But if ill have any problem durring building the preamp i would feel free to ask him for the solution.
Are you going to build something now? I mean do you have a plans to build Vladimir's designs?
- Status
- Not open for further replies.