Part 2: Interview with Bruno Putzeys (the perfect power amplifier series)
Вставка
- Опубліковано 2 чер 2024
- It was an honour to spend some time talking with one of the worlds leading audio engineering developers. Bruno Putzeys' approach to design and his desire to ensure that his technology stays ahead of his colleagues in the industry, is very reassuring for the future of innovative design. In sport we applaud competition and the drive for excellence - I believe this same competitive spirit is also extremely important in science in general and in audio engineering in particular.
Chapters:
00:00 Introduction
01:38 Why design a new module?
05:09 Expected differences should going from UcD and Hypex to Purifi?
07:28 Why use negative feedback?
12:30 Can you tell us about your leadership team?
17:54 What is your view on GaN FET technology?
21:07 What is next for you and your team?
As always, if you would like to find out more about Pearl Acoustics and our Sibelius loudspeakers, please visit our website at:
www.pearlacoustics.com
or email us with your questions:
enquiries@pearlacoustics.com
Social media links:
/ pearlacoustics
/ pearl.acoustics
“If you’re the most intelligent person in the room, you’re in the wrong room” what a great expression!
👍
If everyone stuck to that motto, there wouldn't be a human in any room.
After watching part one and 2 the VTV AMPLIFIER Stereo Purifi Audio 1ET400A Amplifier will be my end game home theater amp. Thanks for making this content.
You’re very welcome. Thanks for taking the time to comment
Probably the most respectful person in the industry. Thanks a lot for the interview!
Thank you very high praise indeed. Much appreciated
An audio legend.... So soft spoken too !!! Excellent interview..
Thanks. Glad you enjoyed it
Class D is the future, and it sounds amazing.
I nominate Bruno for Nobel Prize in scientific development of class D! I find it an amazing pattern how soft and often considered by many not well spoken people can be such hidden geniuses, in fact I know of a few people like him. This man revolutionized the audio industry.
It amazes me how shallow minded stage monkeys always get the recognition in life, but thats another subject.... Thank you!
You’re very welcome, I’ll let him know he’s been nominated by you, and I will second your nomination! He is indeed a great guy!
And here I was, ready to take the plunge on a NAD M23. I guess I will wait for these new modules.
😀👍
Bruno is right about negative feedback which is the same as servo control. This is a very powerful tool and it's also a very complicated one to work with. If you don't understand it, it is very easy to do much more harm than good. As an electrical engineering major in a very fine and demanding engineering school the textbook I used was written by two US Air Force generals. It was targeted at electrical engineering seniors and master's degree candidates. The math prerequisites were considerable. There was more calculus in that book than in a calculus textbook. Like any powerful tool you need the skill to use it properly. With a power saw you can build a house or cut your arm off. People who don't know how to use it should stay away.
Negative feedback is a distortion nulling process that compares part of the output with the input to see the difference and add a phase inverse distortion component to the input essentially predistorting the following signal by the same distortion the circuit is going to create but in the opposite sense. There is a time delay however between the input and output. If the loop gain is one or greater when the phase is 180 degrees out between them you have created an oscillator. Therefore any mismatch resulting from too much feedback signal at high frequencies can destroy the amplifier but at the very least the amount of feedback signal and its frequency response change from the input signal has to be carefully controlled. It is not arbitrary.
The ideal power amplifier will have infinite input impedance and zero output impedance. The power supply should be very well line and load regulated and have substantial overcapacity including adequate reserve charge in large capacitors to not need any kind of power conditioner and easily deliver full undistorted power into its design load within its load phase rating. Today's kludgy corssover networks have wild phase swings. There's one that has a phase of minus 90 degrees at 30 hz. And on paper is should have infinite power from DC to light. It should have unmeasurable noise and distortion. If a power amplifier met those specifications it would be a horrible choice for a hi fi system. However, a real world amplifier that sounds identical to this hypothetical amplifier would be good enough. It should not be able to deliver power well above and below the audible range. Engineers use Thevenin's theorem for the complex output of one stage, the telegrapher's equation for the connecting wire, and the complex input impedance of the load stage to compute the resulting frequency response of the section of a circuit. If the input signal came from a stage that had zero source impedance and the next stage had infinite load impedance the characteristics of the connecting wire would be irrelevant.
Electrical engineers are taught the concept of transfer function. In this concept the output is compared to the input usually by a differential equation. In the real world it's not so simple. A part of this concept involves the Fourier Transform equation and Inverse transform equation. It allows the the engineer to switch back and forth between the time domain of a signal and the frequency domain. In the frequency domain the output is simply the signal in the frequency domain times the gain in del operator notation.
The only advantage of a vacuum tube as an input stage I can see is its very high input impedance. Other than that attribute it appears to be a horrible choice having many faults. The reason the different op amps sound differently is that they alter the overall frequency response between the input and output of the amplifier.
Among the many basic flaws in audiophile-think is that if you had the perfect playback system meaning in audiophile-speak the perfect signal, the perfect wires, the perfect preamplifier, the perfect amplifier, the perfect speakers, the perfect utility power you'd ge perfect results. That is as far from the truth as you can get. You also don't want to hear what the recording engineer heard in his booth because what he hears is not what the audience hears. Unless and until the problem of high fidelity sound reproduction is viewed in an entirely different way, all systems will fail. But it clearly doesn't matter because for most of the market, the intrinsic quality of what they call music doesn't merit better than what is already available to them. For others, the only solution is live concerts.
Dear Mark thank you for your contribution to the topic. It is much appreciated
Mark, I think Bruno Putzeys is using MathWorks software to analyse his class D designs. He published several papers on the subject in the Philips or Hypex days. You can get an electronics module in SimScape that allows you to model electronics circuits in addition to Simulink to analyse feedback transfer functions. Handy for acoustic transmission line speakers too. Honestly, I can't imagine tackling feedback with that number of poles and zeros for the gain factors mentioned without a computational tool, unless you want Laplace transforms pouring out of your ears. 😵💫
@Martin Garrish It's been over 50 years for me but the math never changes. I recall pole zero diagrams and Nyquist diagrams as well. I think the very similar Fourier transform and inverse transform functions are more commonly used. I recall endless tables of them. I lived in the world of sliderules. I had a K&E myself. It's still somewhere in my house lost in a thousand unopened boxes.
The first time I saw a PLC I understood it implicitly and how powerful it is. I've been involved in many projects where they were used. I also arranged for hands on seminars for myself and another engineer at GE FANUC and Modicon when it was Siemens. Feedback or as it would be more commonly called servo control is essential in all modern industrial control systems. I don't mind doing the ladder logic diagrams but I hate doing the programming and specifying every last nut and bolt. Last one I did myself was with an Allen Bradley Powerflex 70 must be about 10 years ago.
@@markfischer3626 , I inherited a K&E sliderule. Still use it when I can't be bothered to find my trusty Texas Instruments calculator. Right up there with the AVO Mark 5 for reliability. But for S/N of 116dB then definitely circuit simulation is the order of the day. Mostly because you can quickly optimise topology and investigate new designs.
@Martin Garrish I also have a beautiful set of K&E drafting instruments. I went to college before the four function calculator existed. The school's computer was the IBM 1620 with 40 K of magnetic core memory. There was also a Univac 1105 vacuum tube computer used by the US Navy. I programmed in machine language and Fortran 2. However flow charting remains the same. I've seen advances and participated in them I never thought would be possible in my lifetime in every field of technology except one, two channel stereo. No conceptual change, just the details of execution. I laugh at how primitive it is and how emotionally worked up hobbyists get about it. It reminds me of the Lilliputians and Belfescuans going to war over which end of an egg should be opened. Great fun watching a clown show. Sorry to insult everybody but it provides me with a lot of amusement. I just can't take this seriously
Great interview. Always love hearing from Bruno. Thoroughly enjoyed it.
Thanks for listening
A great interview. Love these videos! Thanks Harley!
Thank you Alan.
I own a Purifi-based power amp, and it's too good for this world. Amps, like DACs, have in a way come to an evolutionary endpoint.
Absolutely
Excellent once again Harley, thank you so much! All the best, Martin
Thanks Martin!
YES, "racing yourself" IS the secret to bringing out the best in each and everyone of us . . . just as we inherit certain traits from our parents, we ALL inherit from the universe (uni-verse) THE most obvious trait of all - creativity - from the "CREATIVE FORCE" that brought our world and ourselves into existence . . . we all have this "creative force" within us, in abundance . . . it is up to all of us to use it and exploit it to the max, which IMHO is our primary reason and purpose for even BEING here :-) . . . there is another reason to race our selves . . . it keeps us modest . . . for as we try to outdo our SELVES and improve every day, we cease to compare ourselves to others as either superior or inferior because we will naturally always be "faster" or "slower" than the next guy which could, in the end, affect our creative spirit in some negative way. Happy listening to all!
Thanks David
Fantastic - I learnt a lot from this interview. Thanks so much for doing it!
You’re very welcome
Fascinating. Thank you.
Glad you enjoyed it
Stumbled across your channel a few days ago and you didn't take issue with my choice of Naim CDS-II/XPS as a missing part of your 'CD top 20' - so you're obviously a top bloke. This interview was really interesting. My background is large-scale live touring sound (I'm talking Wembley Arena large-scale), where the virtually-universal pro PA amplifiers of choice are made by a Swedish firm called Lab Gruppen. Up to 7kW per channel into 2 ohms - and quite comfortable with that 'men from the boys' impedance. And class D obviously, hence my interest in this interview. For all their colossal grunt, they sound fantastic. Subscribed.
Thanks Kevin for your kind appreciation and very interesting insight into the world of arena amplification.
Another excellent and enjoyable episode. Very technical indeed, but even more insightful. And nice to see references to my home country as a common thread in the series of videos. Thanks, Harley.
Thanks Yves!
Thx for the interview…I know we are talking amps, but just wanted to say I’m ano proud owner of Bruno’s dac design in mola mola Tambaqui!
You’re very welcome
Interesting and instructive!
Glad you liked it!
Hi Harley.
Great video.
A good friend of mine have a nad m33 and it is a great amplifier.
Gr André de vries.
Indeed it is! Thanks for your kind appreciation
Great interview!
Thanks!
Thank you. I really enjoyed the interview. You both looked like you did too.
Glad you enjoyed it! 😀
Harley as usual a wonderful video...!! Hey, have you notice how much your channel has grown? Congratulations..!!
Thanks Edgar, you’re very kind. Indeed, I find it rather flattering that so many people enjoy my talking about the things I am passionate about. And in doing so both they and me learn for the research I mostly have to do and also from the ensuing discussion in the comments.
Fantastic interview! Did I hear right that a next version of the Eigenakt is coming!? I was happy to be of the first jumping on the Purify bandwagon buying a pre-production test modules. I'm hoping the new version would be available soon!
Hi thanks Dimiter. yes, the new, more powerful version is out now (I believe)
@@PearlAcoustics I saw it somewhere but not the official website. Were you discussing it or it was about a proprietary PSU Bruno is working on?
@@dkokalanov Bruno is going to work on a new PSU but I think that is still some time off…
@@PearlAcoustics Thanks. I'm looking forward to it!
@@dkokalanov me too!
Fascinating. I currently run an Allnic 300 b tube amp for acoustic music and jazz and a high power Kinki studio Class AB amp for the rest but I often wonder what using the Allnic as a preamp with an Eigentakt power amp would sound like.
Might be fun purchasing one of the new modules and doing a bit of DIY as a project.
Thanks. I think that combination could be very good indeed! Certainly headed in the right direction.
For the first time I easily understood class D speak great interview!
Thanks!
I wonder what Bruno thinks on the VTV option adding tubes to his modules. Valuable upgrade or unnecessary adjunct? Two good things in separate does not necessarily mean that they show additive effects in combination , or what you think?
It’s a great question. Thanks. I had the impression from my meeting with Bruno, that he is very open to different designers and manufacturers coming up with their own buffers. After all, the purify module needs them in most cases. + His biggest customers are original equipment manufacturers, such as NAD. It’s no big secret, but the Pearl Acoustics design team is also working on their own design application for the Purify module. It’s an extremely interesting technical challenge, let alone a music appreciation one!
Great interview. I wish they start sooner the power supply line of products, and the combo SMPS+amp modules, before refining even further the Eigentakt's already very low distortion.
Hi Matias, thanks for your kind comment. I guess the new amp was first in the list because of OEM demand and the fact that it is (more or less) the same but more powerful. I know the power supply is next. More, I know not…
Just came across this interview, very much appreciated! Just wonder, what’s status in terms of the work on power supply?
No news, I am afraid…. Sorry
Great interview. Talking about class D amps: are Nuprime (ex-NuForce) amps under your radar too?
I listened to some amazing sound coming out of them (Evolution One, Evolution STA, ...).
Thanks José. I know I heard some a while back but need to re-visit. Thanks for the tip
Very interesting interview
Thank you
I just got used to how astonishingly good Hypex NCore DIY amplifiers sound with full range drivers! But the -116dB distortion spec for the Eigentakt 1ET400 is amazing. Reading the purifi-audio information on Magnetic Hysteresis Distortion (MHD) then I'm slightly puzzled why not just go with air core inductors in the resonant feedback circuit? Also, the old audiophile tricks of replacing electrolytic capacitors with polypropylene and replacing steel terminals with non-ferrous brass are worth a look too? I'd certainly pay the extra cost for top spec components.
Hi Martin, thanks for your comment
@@PearlAcoustics , you're quite welcome. You might seriously consider active versions of your excellent mass loaded V-horn Sibelius design. Maybe Bruno would agree to a Pearl acoustics Eigentakt special using air core inductors and polypropylene (or better) capacitors in the resonant feedback filter circuit? Feedback corrects the non-linearities, but can you hear a difference?
For example, I'm driving my own low distortion paper FR designs directly with Hypex NC400 to avoid the speaker cable (and horrid ferrous connectors) and it is definitely worth doing. Like being in the same room as the musicians (breath-taking).
@@martingarrish4082 interesting. I’ll talk about your findings with the team and with Bruno
Is the higher power version of the Eigentakt that he mentions the 1ET7040SA, or is it something newer than that?
It’s the 1ET7040SA.
Pearl Acoustics entrevista a la mas grande mente maestra de la amplificacion del siglo XXI.
Con UcD vencio a la clase AB, con Ncore igualo a la clase A y con Purifi vencio al single ended clsse A. Brillante.
Que deteriorado esta de salud Bruno☹️🥺quizas demasiadas pantallas en su vida demasiadas/contaminantes/vida sedentaria....le deseo todo lo mejor que se mejore.
Thanks for your comment. I am not aware of the fact Bruno is ill… is this a recent thing?
NOT TO BE TECHNICAL,HARLEY …THANKS FOR DOING THIS INTERVIEW 🤗👍💚💚💚
😀👍🙏
@@PearlAcoustics 😁🌞 I hope you have a great day
The Man !
A very good interview with a major contributor, thanks. Is Eigentakt/Hypex/Class D the endgame? Let us ask, why can Bruno so easily describe the sound difference between NCore and Eigentakt, when the NCore already has vanishingly low distortion, impedance and high wattage? For me it points to the inferior measurements, which are based on steady-state simple test tones as far from real dynamic music as it gets. The Class D style amplifiers are very un-linear, and need extremely high feedback ratio for self-correction, and sharp digital filters to avoid aliasing. My own NCore 250 has tremendous power compared to my Sugden, but there is no match when it come to realism and dynamic sound to my ears, the simple topology is better, just as a filterless full range speaker sounds so much more real than a complex multiway speaker.
Thanks for your kind words and comment. I have to agree about the Sugden (A21, A21SE, SPA-4) they have something very special indeed.
How would you describe an un-linear sound with aliasing? I'm also using a filterless single driver speaker (EMS LB5), but I'm too cheap to buy a serious amp. I just use a vintage A/B amp with 40W.
@@user-xg6zz8qs3q Your A/B amp is likely great, most are. A larger amp is not better, it just louder. Typical load is not more than 0.1Watt @ 75-80db normal listening levels. Your A/B amp use much less feedback, and no digital filters, which might be the reason why KEF also use A/B for tweeter and class D for bass in their active designs. Class D need massive feedback to behave linearly, and filters to remove noise.
@@alfovebraseth It’s a Continental Edison PA9109. Not exactly high end. It was probably built by Sanyo in 1979. I’m shocked by how analytical it sounds (after the servicing). There are schematics for that amp online. It’s like the engineers who designed that product knew what they were doing. Everything is bog standard/generic. And I can’t fault it.
I spent much time studying sigma-delta conversion twenty years ago. I have a question for Bruno: which of the TI chips does he most admire?
Hi Trevor, thanks for your question. Not sure that Bruno will be available to answer questions via this channel. But maybe you might be able to share your view on this topic?
I'm still using Class AB amplifiers but I'm curious about Class D. Sir Clive Sinclair introduced Class D power amplifier modules in the 60s. Class D was then dormant for many years. Switch mode power supply technology became dominant in the 80s, first in high power applications then low power. Sigma-Delta signal conversion started in telecomms in the 70s, to reduce A/D & D/A cost using digital integrated circuits then migrated to Hi-Fi, with DSD & SACD being the ultimate expression of the art. I assume Bruno's work applies all previous art to power (rather than signal level) applications. I havn't followed the art but there's probably a good book out there detailing the history of all these advances?
No op amp can work without NFB, they can not perform any "operation" without it. All filters, all gainstages demand NFB. Only very low gain discrete amp stages can get by without feedback but even then, when you inspect the circuit there will always be some sort of current feedback even if there's no voltage FB
How would a Purifi 1ET400A implementation (e.g. D-Sonic, Buckeye, NuPrime) compare to a classic AB like my Mark Levinson 533H? I am interested in the musical performance first of all, not just power. Looking for a new 3-channel power amp to use as LCR in home cinema setup, as well to use as a power amp in a good stereo setup (currently have the Musical Fidelity Nu-Vista 600 as stereo amp).
Hi Lars, you have very fine equipment. You really will have to try one for yourself. We are working on a variant amplifier for really high quality applications. Not sure exactly when it will be ready, but if you write Toby’s via our website we can keep you up to date on its progress.
@@PearlAcoustics I sent an email. Thanks.
Excellent
Thanks!
What is the difference between the new amplifier mentioned and current one?
Mostly Power… it can deliver more power and up to 40A.
I have a Spectron musical MK lll class D amp designed John Ulrich who worked at Infinity and designed the switching/pulse-width-modulated/class-D amplifiers in the early 70's The amp weights 54lbs Wow would it be something special get these two gentlemen in the same room together Unfortunately John passed away back in 2015
Thanks for your comment Jack. Much appreciated
You should have asked about the rise in distortion with frequencies in class D.. most are just -70-80db from 7khz and up.. tho i guess tweeters need less power..
Not with the Purify modules
@@razisn yeah sure. Sad you and so many others lack the gnosis and you choose to fool yourself...feel free to enjoy your trip on your self induced fantasy island 🏝️
Legend
👍
How can I distinct older Eigentakt amps and the new generation that Bruno mentioned?
They are more powerful and some say more ‘musical’ I hope that helps?
I find it very interesting to see that people with very active minds seem to have trouble putting their thoughts into words...reminded me a lot of the way elon musk is talking...although elon sometimes is hard to listen to, I guess Bruno does get enough sleep while Elon sometimes is just a weird chaotic mess due to lack of sleep.
Bruno bus a brilliant mathematician and engineer. He wants to be precise with his words, which I fully understand. Patience rewards in these circumstances
@@PearlAcoustics my comment wasnt meant as criticism, I just try to imagine whats goin on with such brilliant minds and I can only assume that it is a bit similar to having a firework in the head and trying to describe one of those vivid explosions while the firework is still going wild...hard to focus on describing one thought when ideas, questions, solutions, datasheets and kermit the frog pop up in your head simultaniously and all the time:)
but again, no criticism intended, I have nothing but deepest respect for brilliant minds like bruno...I'm lightyears away from even fully understamding what he develops!
@@arnejensen9216 absolutely. It’s hard to imagine how great people think. I have been privileged to have work with quite a few in my career.
"The only advantage of class D is efficiency, everything else is a nightmare". Did I hear that correctly? Because the "experts" over the online forums are treating class D as the very best (Hypex NCORE).
I think he meant a nightmare in terms of complexity and difficulty regarding designing the higher order feedback loops.
@@jimfarrell4635 I believe you. But with complexity comes compromise. There's a case to be made for simpler topologies with reasonably small power (
@@user-xg6zz8qs3q I don't disagree. I'm a Class A and AB guy and still judge amps by weight but I have yet to hear a Purifi amp and am open to persuasion.
Thanks for your comment. Bruno was referring to the complexity in product development. I don’t see any correlation between complexity to design and performance. This applies to all science but especially software development, mathematics and signal processing
I don’t think you’ve heard the guy. Kinda typical trolling technique to take a few words out of context
What is the new amp module he is talking about.
Hi Jos, it’s the 1ET7040SA
@@PearlAcoustics ok thanks. That one is on my list.
No one is waiting for a better class A amplifier? I for one very much was looking for even one truly ultra high end class A amplifier.
Hi David, we are planning to release one in the Summer, maybe you can come and hear it and see what you think? 😉
@@PearlAcoustics Excellent...Love Ultra High Performance and Thank You 😊
That sofa looks very comfortable to sleep on 😀
😀 never tried. But bound to happen one day!
So I wonder what Amp Bruno listens to at home. Surely not a Purifi Kit.
😀 - probably an old single ended tube amplifier! 😂 (Actually, I believe he listens to his Klll loudspeakers so he only needs a pre-amplifier, so likely to be a Mola Mola.
Oh the irony! Talking very quietly about audio amplifiers! =o)
11:32 is what a designer thinks about Class D. " Absolutely terrible in terms of complexity!"
Hi Josef, what Bruno us talking about here is the scientific design. Meaning it’s very complex (hard) to design, to get right and that’s why too many don’t sound very good. The mathematics is extremely complex. But when you get it right, then you have something close to ‘perfection’ for want of a better word.
Really hope they would work on DAC and 8 channel dsp and Airplay all in one
Hi Anand, I am sure one of Bruno’s team will be reading these comments, so you never know!
His comment about ADC/DAC tech relating to AMP design is dead on, and how the whole "no feedback" thing is only promoted by amateurs...
Thanks for your comment
hypex are good, but how come some asian dudes managed to pull way way ahead in clean signal reproduction with just a ordinary lm3886 class-ab amp? beating everything ever measured
It has gain peaking , it is not as sweet as he say , it can't drive capacitive load 😅😅
The reason I use linear power supplies is simple. In over 50 years of making amplifier of one kind or another, I have NEVER had one fail. Should one fail, they can always be repaired. A SMPS on the other hand are not so reliable. I have had quite a few fail over the years, some of which cannot be repaired for one reason or another. I hope his power supply is not over clever for cleverness sake. After all we only need a smooth DC supply, it's not rocket science after all.
Hi Michael, thank you for your comment. I totally get where you are coming from, however the power supply in question is far from straight forward. You might enjoy reading about it? For a start it has to produce several different voltages, some parts are current driven. But the important part is, it needs to deliver exactly what is required and that is not a constant. Comparing a very sophisticated power supply for a purify amplifying module with a linear supply, is a bit like comparing petrol carburettors on an old car with modern, highly sophisticated fuel injection systems. The latter is way more complex but delivers more power, uses way less fuel and can adjust itself to the needs of the receiving components at any moment in time. Modern switching power supplies are very stable and should last decades. I hope that helps?
Honesty I didn't understand what you guys are saying I was scratching my testicle all the time like a baboon 🦧
Bottom line : I lost all respect for Bruno. seriously go and LISTEN to his designs. They are the thinnest, least musical and most clinical sounding gear. The Kii, Tambaqui, Purifi, etc. Well they can be modded to sound better but I strongly believe he does not LISTEN TO MUSIC, he designs with measurements.
That’s a shame. In my experience I cannot agree with you. I happen to know that Bruno listens to a lot of classical music and has a very broad taste in music. We spent an afternoon together listening to a very wide variety of music from his personal listening list.
@@PearlAcoustics Did you listen on his equipment? Have you compared his amplifier to a SET?
@@xfloodcasual8124 hi. I am very familiar with the sound of his gear and yes, I usually listen on SET class A amplifiers. I understand where you are coming from but one has to accept that his amplifiers are doing what they ‘should’ that’s why I say in my video that to make something ‘nice’ there is still some work todo on the buffer design.
@@xfloodcasual8124just stop it troll. You are judging without any knowledge or facts about Bruno, his setup, and musical preferences. Educate yourself first, do your due diligence before you start trolling and spreading baseless opinions
@@larsv6144 I'm very educated. I own a lot of his gear and have reversed engineered and modified it to understand the weaknesses. I've talked with him several times over email. Also, you can just listen and draw your own conclusions. Don't assume.