Soundstage!
Interviews


May 1997

Interview with Bruce Rozenblit of
Transcendent Sound, Inc.

by Richard Seah


Part I - Introduction

OTL. These three letters, which stand for "Output-Transformer-Less", did not have any meaning for me until mid-1996, when I read a review of two OTL tube amplifiers. I didn't remember most of what the reviewer wrote, mainly because I tend not to pay much attention to things that are beyond my budget, in this case US$6,295 for one model and US$11,950 for the other.

I did remember one thing though. The reviewer talked about how heavenly the amps sounded and he mentioned something about "Tube God"?

I was intrigued. And the term OTL stuck at the back of my mind. Shortly after, I came across another OTL amplifier, by Transcendent Sound Inc. Out of curiosity, I did an internet search and discovered that it was a lot more affordable, at US$2,995. (This model has since been replaced with an improved version at $3595). I had to find out more.

Thus began my e-mail acquaintance with Bruce Rozenblit, designer of the Transcendent Amplifier. Bruce tells me that OTL amps produce excellent sound because they do away with the output transformer, which is a major source of distortion.

The downside, however, is that conventional OTL amps are problematic. They generate a tremendous amount of heat. This, in turn, causes the tubes to burn out and sometimes blow up! Such amps draw lots of current and you may have to specially install a dedicated high current circuit before you can even plug it in.

This was confirmed when I re-read the reviewers article where he states: "OTLs tend to be space heaters/power hogs. You should certainly consider, as a minimum, a dedicated circuit, such as the one I have installed in my reference room, where a separate 20-amp circuit services the power amps."

These are by no means minor inconveniences. But that's not all. While OTLs are regarded as ideal for electrostatic loudspeakers, they didn't usually don't do a good job at driving dynamic loudspeakers. Moreover, the bass response tended to be weak.

In short, OTLs were excellent sounding, but problematic amplifiers with severe limitations. Which is why they never became popular - except among the most dedicated audiophiles - although OTL technology has been around some 45 years, since the early 1950s.

No problems now, says Bruce, who claims to have built the world's first practical OTL amp, one that you can simply plug and play. I think the best word to describe it is "user-friendly".

For starters, his Transcendent amp generates only 275 watts of heat for the stereo version, which is less than 100 watts warmer than a conventional tube amp. In comparison, other OTLs generate between 1,200 and 1,800 watts of heat!

You don't need a special high-current circuit to use the Transcendents. Best of all, it will drive most loudspeakers - electrostatic, dynamic, ribbon, horn, whatever - provided the impedance does not drop below 4 ohms. Bruce has even managed to put considerable punch and power into the bass.

Bruce has made OTL technology accessible to those of us who will not tolerate a whole lot of inconveniences - plus high expenses - just to enjoy music.

His achievements have been recognized by the US patents office, which awarded him a patent at the end of last year. This was the first OTL patent awarded since 1964.

Unfortunately, the audiophile community is still largely in the dark about his work. That's why we like to tell you more about the Transcendent OTL amplifier, and about the man who designed it, Bruce Rozenblit.


Part II - The Interview

[BRUCE ROZENBLIT](RS) Bruce, can you begin by telling us a bit about yourself. What got you started in building audio equipment? How old were you then and what was the first thing you built?

(BR) I first became involved with electronics when I was no more than 10 years old. An uncle of mine bought me a crystal radio kit and I was hooked. Throughout my whole childhood, I was fascinated with science of all kinds. Mostly, I was very curious about how things worked. I couldn't wait for appliances around the house to break so I could take them apart. Sometimes, I helped them break so I could have an excuse to get inside them.
 
I became involved with audio when I was about 13. There were some older boys across the street that had some audio equipment and I thought that this was the neatest stuff in the world. I use to dream of owning my own Hi-Fi gear but it wasn't until I was 18 before I had enough money to buy any. I bought a Heathkit amp and tuner, then a turntable and a pair of headphones.
 
I attended the University and got a Bachelor of Science degree in Electrical Engineering. I got B's mostly. I was more interested in building projects than studying. When I got out of school, I couldn't find work locally in electronics, so I worked in the power engineering industry. This caused me a lot of frustration. I didn't work with electronics for 7 or 8 years. Then at around age 29 or 30, I started getting involved again. I discovered Audio Amateur Publications who put out a series of magazines devoted solely to home construction of high quality audio equipment. I started reading and dreaming again. I ordered some kits and got back into the hobby.
 
Up until this time, I had no interest in vacuum tubes. In fact, I use to laugh at them. Who needs such an antiquated technology. Then it happened. I was shopping in a thrift store looking for bargains and I came across an old, rusty, Eico 35 watt per channel tube amp. I had to have it.
 
At this time (1987) I knew absolutely nothing about vacuum tubes and I mean NOTHING. In order to analyze the circuit, I made a pictorial diagram of how everything was physically connected. From this pictorial, I developed a schematic. I still didn't know what it meant. I went to the library to get some basic information just so I could even interpret the symbols properly.
 
After some study and I had a minimal idea what was going on, I cleaned up the chassis, repainted it and the transformers, got new parts and put the thing together. I worked feverishly into the night for days. I had to know how this thing was going to sound. I hooked it up to my speakers, and I was in hi-fi heaven. I was awe struck at the sonic quality of this little amp. That day was a major a turning point in my life. Incidentally, the amp is still in operation at a friends house. It has performed flawlessly for over 10 years.
 
I had to learn everything that I could about tubes. Fortunately, less than one mile from my home stands the Linda Hall Library. This is one of the greatest and most comprehensive scientific/technical libraries in the world - no exaggeration. I spent many Saturday afternoons pouring over the engineering records looking for articles and books that would teach me about tubes. The library was my teacher, my mentor. Slowly, after much study, things started falling into place and I started to learn.
 
When I read those old articles from the 30s, 40s and 50s, I felt like I was young man back in that era learning about tubes as if it was new technology at the time. I was so enchanted with tubes, that I felt like I was experiencing the breakthroughs that were published as if I were reading them as a contemporary of the period, as if it was my time then. Maybe it was.
 
My job(s) continued to give me much frustration. I have a lot of creative energy and I have to do something with it. Audio Amateur started a magazine called Glass Audio then. That was the answer. I would design tube projects that would be published in the magazine. This would give me a creative outlet. Something I could focus my energy on. I didn't make any money doing it, I did it for fun. Some of us are born to create. I have no talents in the fine arts, but I can work with technology. I am a born design engineer. I must create new things. That's what motivates me.

(RS) What attracted you to OTL amplifiers? Can you recall your first experience of an OTL amp?

(BR) During my studies at the library, I came across designs for OTL amplifiers. The concept fascinated me and I thought that it would make an interesting project for Glass Audio magazine. So I designed one in 1989 and it was published in late 1990. The sound was unbelievably good. I knew I was on to something. I developed another design in 1990 which was even better. My technology was in its infancy but I knew I was heading on a path to somewhere.
 
In Jan 1991, I decided to go into business for myself and formed my company. I kept developing better and better OTLs but it wasn't until the summer of 1994 that I developed the breakthrough that really catapulted the design and formed the basis of my patent. Up to that point, I had achieved good reliability and low heat but the amps all had weak bass. The new circuitry solved that problem and I new I had an amp that could do battle with the best of them.

(RS) You mean you built your first OTL amp without first having heard one?

(BR) Yes, I was attracted by the technology. I figured it would sound good. Then, when I built it and heard it, I was hooked on the sound.

[TRANSCENDENT AMP](RS) How would you describe the sound of an OTL amp?

(BR) OTLs have a distinctive sound. They do not sound like a conventional tube amp. They do not sound like a transistor amp. They capture the best of both types along with the ability to develop incredible three dimensional imaging that no other amp can produce. People that are expecting a traditional tube sound from an OTL will be disappointed. People that are the most happy with an OTL are looking for extreme accuracy and realism. They do not want any coloration of any kind.

(RS) From my experience, it seems that OTL amps are better known for their problems than for their excellent sound. For example, they generate a tremendous amount of heat which causes the tubes to blow or burn out. Some even require that you install a special high-current circuit in your house before you can plug it in. Can you comment on this.

(BR) I have been fighting this image since I started. I tell people I have an OTL that doesn't blow up and they laugh at me. I'm sorry to say this and it is probably going to make some people mad but the caliber of modern day tube designers isn't any where near where it was in the 50s. Instead of degreed engineers, most are experimenters with no formal training who don't understand the theory behind the circuitry. When you don't have the proper training, all you can do is take an established design and modify it. It is almost impossible to come up with something new. It's like changing your appearance with clothes or make up. You are still the same person underneath.
 
The reason of all the other OTLs suffer from these same problems is because they are all basically the same design. There is more to being a circuit designer than selecting a 3-inch piece of wire or using a 1/4-inch thick chassis. All of my designs (there have been 8 over a 5 year period) have been reliable.
 
I had some problems finding the best output tube but the circuitry always held up. Now with the introduction of the new Svetlana EL509 and some further improvements that I have made, the reliability has been pushed to levels as great as tube gear of the past generation.
 
I have heard many stories from dealers and audiophiles about the low reliability of modern day tube equipment. I have repaired many pieces of old tube gear that was 30 years old or more that finally broke. Why is it that the old fashioned stuff could last for decades and the new stuff breaks all the time? It is all in the design and implementation. That is called engineering and that is what I do.

(RS) You have managed to overcome those problems. I believe your Transcendent amp is the first, or one of the first, OTL amps that is practical and user-friendly. What made you think those problems could be solved in the first place? And what made you believe that you are the person who could solve them?

(BR) I embarked on a journey and I didn't know where the journey would take me. When I began the project, I approached it from a completely practical viewpoint. I knew that just about any tube amp sounds good. The challenge with the OTL was to get the heat down and the reliability up. I knew good sound would follow. Many contemporary designers of audio equipment don't even consider practical matters and they produce gear that you can't even fit into a room.
 
I didn't know what I would ultimately come up with. I had already designed many excellent projects so I had learned to trust my intuition and instincts. I let whatever ideas that inspired me direct my work. The biggest battle in harnessing creativity is to trust your intuition. Most engineers can't do that, let alone admit that such a thing as intuition exists. Tubes are my element, my medium. I just let myself go, being convinced that wherever I ended up would be a good place.

(RS) Were there times when you felt like giving up?

(BR) Giving up was not an option. I didn't treat the work as trying to reach some absolute goal. When you are on a journey, the fun of it is the travelling. I treated the work as an evolutionary process that takes me wherever it takes me. By maintaining this attitude, failed ideas caused little frustration and just increased my knowledge of what worked and what didn't.

(RS) I understand you were recently awarded all 20 claims in your patent. Does this mean you have solved ALL the problems of the OTL?

(BR) The only trade off that I can't solve is that the OTL doesn't do well with low impedance speakers. This is because tubes have a maximum current that they can pass. Transistors can pass tremendous amounts of current to the point that they can easily burn themselves up. With the OTL, power output start dropping very quickly when speaker impedance falls below 4 ohms. It is best to use 8 ohm speakers. Unfortunately, many speakers are 4 ohm because solid state amps can double their power when impedance drops from 8 to 4 ohms.

(RS) What is your design philosophy? Many people believe, for example, that a good amplifier needs pure class A circuitry, or expensive components, or point-to-point wiring without circuit boards. Some believe in single-ended design, some believe in push-pull. And so on. What do you believe in? Do you believe that OTL design is absolutely necessary? Or can you build an equally good - or better - conventional amplifier with an output transformer?

(BR) This is where the rest of the world is way off track. The answer is: "There is no best way, there are just ways." There is no ultimate circuitry, or resistor, or wire or technique or anything. There are just successful applications of technology and unsuccessful ones. It is all in the implementation. Many people involved with audio for some reason are on a quest to find a "holy grail." They are searching for THE answer. It doesn't exist. I use whatever provides outstanding performance at minimal cost.
 
It is highly unfortunate that a culture of terribly expensive parts has evolved over the last 20 years in audio. People think that if you don't have $10 resistors in your circuit then it can't possibly be high end equipment. Baloney!
 
I feel the main reason this "parts culture" has evolved is that there has been so little innovation in circuit design that the only way one manufacturer can distinguish his product his competitor is to say that he uses brand XYZ parts while the other guy uses those awful brand ABC parts. The circuitry used in most contemporary tube equipment is about the same. This wasn't the case in the 50's and 60's. There was wonderful diversity in design during that period.
 
I have a customer who has owned many very expensive amplifiers and was dissatisfied by all. He bought my amp and loves it. He says its the best amp he has ever had. He took the bottom off, and didn't see a chassis full of $10 resistors, $30 capacitors and $20-a-foot hook up wire and therefore is convinced that the amp is not performing anywhere near its capability. As good as the amp is, he can't accept it for what it is. Even though he freely admits that it is the best he has ever owned, he has to modify it. It must have those exotic parts!
 
This is an example of the psychology that 20 years of successful marketing can produce. Many times these highly expensive parts do alter the character of the sound, but is the alteration an improvement or just a change? If you just paid $200 for a new toy and it changed the performance of your equipment, it is a very small step to believe that the change was for the better to justify the $200 you just spent. If I were to use these items on a blanket level, my products would cost 2 to 3 times as much. Perhaps in the future I will offer such a product to satisfy customers who want to spend that much but right now, I am trying to emphasize value.
 
If I have a philosophy, it is to make the circuit as simple as possible. I try to strip out all unnecessary circuits and components. Long ago a friend used to always tell me, "simplicity is the keynote of design." I have never forgotten and he was very, very, right.
 
It is possible to create high performance audio equipment using a variety of techniques. The end result is determined by what parameters and characteristics you want to maximize at the expense of others and at what cost? There are always compromises. This is the challenge of engineering.
 
There are certain advantages that eliminating the output transformer can produce. It creates a sonic signature that is unique to the genre. I prefer it. Some people prefer single ended sound. That is their choice.

(RS) Your earlier amp used the Russian 6C33C-B tubes, which, as you saw at the CES in Las Vegas, are starting to become quite popular. But now you have switched to the EL509. Why?

(BR) The Svetlana EL509 is a vastly superior tube. I used the 6C33CB because it was the only tube at the time that could take the pounding. The 6AS7 (which I used in my first OTL design) sounded a lot better but it would not hold up. The 6C33CB is a very imprecise tube. The plate current in a random sample of eight tubes can vary 10 to one. That's a 1000 % variation. This is intolerable for a manufacturer. Fully 20% of the tubes that I buy are unusable because they are so far out of spec. There are two separate triodes in each tube with common electrical connections. Each triode's specs. are usually dramatically different. The problem, I think, is in the grid circuit. After all of the trouble I go through to weed them out, about 5 to 7% will fail in three or four months. When you get a set of good ones, however, they will last for years, and are very stable. The problem is the consistency. The 6C33CB is also a rather hard sounding tube. In order to make the amp sound smooth, I had to restrict the bandwidth to 20 kHz.
 
The EL509 does not suffer from these problems. I don't have to pretest and sort them out. They don't have to be purchased in matched pairs. They have a much more stable cathode. The tube is more musical sounding so I can let the bandwidth extend to beyond 50 kHz. The power is reduced by 20% which is minimal but the waste heat is reduced by almost 100 watts per cabinet. The tube is in ample supply. In short, it sounds better, is easier to work with, much more stable and easier for the consumer to live with.

(RS) How well have your amps been selling so far?

(BR) Sales have been ramping up. Most of my sales come from customer referrals. Someone buys it, tells all of his friends, they buy it, etc. I would like to see sales go higher for obvious reasons.
 
Audio is an extremely tough market. I liken it to being a baby turtle that has just hatched on the beach along with 1000 others. As soon as you start to move towards the water, every manner and type of predator will do everything possible to eat you. If you somehow make it to the water and begin to feed, no one will pay any attention to you until you are large enough to make a big pot of turtle soup. I'm at the stage where I have made it to the water and begun to feed but I am too small to throw into the pot.

(RS) I understand you are now working on a pre-amp. In what way will it be different from other existing pre-amps? When do you expect it to be ready?

(BR) Now that my book is done, I will have time to develop new products. I have no intention of producing another me too anything. My pre-amps will utilize new circuitry. That raises the question of another possible patent application which is a lot of work. The question will be if it is different enough from the prior art to justify the application and there are a lot more pre-amp designs to go through on a patent search than OTLs. I have several ideas to explore. I will soon start on that journey.

(RS) What about other products, like DACs? Or a budget amplifier or an absolutely state-of-the-art cost-no-object amplifier?

(BR) I am an analog circuit designer. I don't have much experience with digital circuits. It would take me at least 6 months and probably a year of study before I could come up with something. There are some very fine digital converters on the market. I like to work with tubes, not micro chips. If my heart isn't in it, I won't do the kind of work that I expect of myself.
 
As far as a super amp, that is still possible. I am motivated by efficiency and cost / performance. I don't see the sense in spending 500% more for 1% improvement in performance.

(RS) You mentioned having completed your book. I understand it's on amplifier design. What can we expect from it? Are you going to teach people how to build amplifiers? Will you be giving away any trade secrets? What's the title of the book? And when will it be out?

(BR) The book should be out by mid year or sooner, I hope. It is being published by Audio Amateur Corp, (publishers of Glass Audio). The title will be The Beginners Guide To Tube Audio Design.
 
What I have attempted to do is write a book for lay people that teaches them how to design tube circuits. It is not a cookbook of circuits. It is a teaching book. I have tried to relate the fundamentals of a college degree in a practical, intuitive way so people can understand the material. The material is generic but there is plenty there to get involved with. I give the reader the benefit of my experience in showing ways to solve design problems.
 
Even if you don't want to do your own designs, the book will yield a great deal of knowledge about circuitry in general. Hopefully, it will help audiophiles gain a solid understanding of electronics and tubes so they can better enjoy the hobby, make informed and intelligent purchases of audio equipment, and not be taken by the many snake oil salesmen and con artists that unfortunately proliferate this business. I really want people to get involved and build their own gear. It is a lot of fun, very rewarding, gets people away from the television set. The world needs more doing and less watching.

(RS) Where do you see yourself 5 or 10 years from now? Do you think you will radically influence the way future amplifiers are built? Do you see yourself becoming famous, perhaps like another Mark Levinson?

(BR) I don't know where I'll be in 10 years. In the US, success is measured by your financial accomplishments, not your technical or artistic achievements. I am already assured a place in audio history with the issuance of my patent. This validates the claim that I have significantly advanced the state of the art. Whether or not I can get the rest of the world to care is another matter.
 
The cruel reality of the marketplace is that I don't have the financial resources to buy myself an image as large as any of the established audio companies. Even if I had the money, I may not have the marketing savvy to pull it off. Luck and being in the right place at the right time has a lot to do with becoming a financial success. All I can do is try, give it my best and see where the journey takes me.

Part III - Loudspeaker/Amplifier Safety

(RS) Most of the claims Bruce makes about having solved the problems of OTL design can be easily verified. We can see that he uses few tubes, we can feel that the tubes aren't too hot, we can hear whether or not the amplifier can drive dynamic speakers, or whether it delivers bass punch. Another common worry, however, is that an tube fault might cause the loudspeakers to burn. How can we be sure this won't happen? Bruce explains:

(BR) There is absolutely no possibility of damage to the consumers loudspeakers. This will require a technical answer. First of all, the protection scheme that I am using is an improvement over the original used in 1989. The original amplifier used 6AS7G output tubes. It had an extremely smooth sound. But the tubes failed all over the place. They couldn't take the pounding. There was never any damage to any speaker when this occurred and it occurred dozens of times.

Secondly, the first design was published in Glass Audio and many people, possibly hundreds, from all over the world built the amp. I never received one letter from anyone about speaker damage. There are dozens of amplifiers in the field now, some for over a year, and there has never been any damage anyone's speakers.

In order to understand how the protection works, it is necessary to understand the mechanism of failure for a loudspeaker. There are two modes of failure, heat and mechanical. Excessive heat in the voice coil will melt it and cause it to short out or open up. A mechanical failure can occur if a sufficiently large pulse of current pushes the diaphragm so hard that there is a rupture. Voice coil burn out is more common.

Heat is the result of energy delivered into a substance, in this case the voice coil. There is a fundamental difference between power and energy.

Power is the RATE that energy is delivered. Example: If you were to try and boil one litre of water in a pan with a candle, it probably never would boil. One candle can't get that much water hot enough. Let's measure the temperature of the flame and see that it is 500 deg C.

Now, take 100 candles (assuming they can fit under the pan) and we find that water quickly boils. Interestingly enough, the temperature of the flames for 100 candles is still 500 deg C. The reason the water boils is that there is 100 times the rate of energy delivered to the pan to cause it to boil. The rate of delivery of energy with one candle is so slow that the pans cools off as soon as the heat is applied so it never boils. It will warm up a little, but it wont boil.

Obviously, a certain amount of time is required before the water will boil. If two hundred candles were used, the water would boil in half the time. With 200 candles, the same amount of heat is applied in half of the time because the rate of delivery of energy or power is double.

An electrical analogy of this would correspond with voltage replacing candle flame temperature and current replacing the number of candles. Time is still time. So two factors must be considered when examining overheating of a voice coil. One is how much power is applied and two is for how long. Both quantities, the rate of energy delivery and the time length of energy delivery are vital. The net total energy applied is always the product of rate of energy delivery multiplied by time.

To the novice, it sounds like a disaster if 150 volts is applied to a speaker. That depends for how long. You can safely pass your finger through a candle flame and not be injured. However, hold your finger in the flame long enough and you will receive a severe burn. The increase in time allowed more heat to be applied to your finger causing the damage.The same amount of power is present in both instances. This is the principle behind the speaker protection in my amplifier.

A fuse is the simplest, most reliable type of overcorrect protection available. Electronic circuits used for fault protection take time to recognize the fault and then to act it. The fuse will open faster than any protection circuit can operate. The power supplies are fused. In the extremely rare event of an output tube short, the surge of current causes the fuse to blow almost instantaneously. There isn't enough time for any damage to occur just as you can pass your finger through a flame and not be burned.

Furthermore, the original design had nothing to limit the current allowing a tremendous surge (probably in the range of more than 1000 amps) and still there has never been any speaker damage. The rise time of the fault is so fast that the reactive properties of the loudspeaker keep any current from flowing into it. The energy of the fault is dissipated across the tube not the speaker. By the time the reactive properties have had a chance to charge up and allow for steady current flow, the fuse has long since blown. The event is over with before it started.

The new design uses passive devices to limit the fault current to 75 amps at most. This a reduction in power of one hundred times. There is no way a sustained fault can be maintained. The fuse will always blow in an instant. Not only is the speaker fully protected, but the amplifier and output tubes are also protected from damage. The protection scheme is foolproof and has been thoroughly field tested.


SoundStage! wishes to thank Bruce Rozenblit and Richard Seah for this interview.
If you would like to contact Bruce he can be reached via e-mail at tubehifi@worldnet.att.net.
Richard Seah lives in Singapore and can be reached at seahss@singnet.com.sg

The interview also appears in the April/May edition of High End Journal (in Singapore). This interview is printed in SoundStage! with the permission of Richard and Inkwell Publications, the publishers of High End Journal.