Valve circuits.... Manley Labs takes a different approach to valve circuits than most pro audio companies. Starting with a little history, co-founder David Manley's background was EMI classical recording, owning his own 7 story recording studio complex, owning a large stereo store, and manufacturing hi end HI-FI equipment. David was credited in American HI-FI journals such as "The Absolute Sound" as the man responsible for re-introducing tubes into the top rank of HI-FI. Its worth pointing out that distortions and colourations are frowned upon in these "schools". Perfection is the goal.
As valve circuits began to become popular again in pro audio various journalists struggled to explain why people like tubes. Some of them had backgrounds on the test bench and others were guitar players and others were familiar with a few antique pieces. Understandably, the only conclusion they arrived at was it must be distortion that people like. Someone just as brilliantly coined the term "warm" to describe this effect and it caught on. Today, there are many manufacturers led by marketing departments armed with the catchy phase "warm", attempting to design in a technology they are not experienced with to aim for more distortion. Manley Labs is different. To us, "warm" means "bikini effect" - droopy top and soggy bottom. We don't like the overused word "warm". We go for words like "rich", "natural", "musical", "dynamic" and "involving" to describe our tube sound.
We learned different lessons and learned them many years ago. As transistor gear made its entry into studios in the early 60's, we heard new distortions that disappointed us in this new amazing technology. Then IC's came into popularity and audio suffered again. Digital CD's became the rage but they often sounded harsh. It was time for tubes again. Early valve gear often had "personality", but this was more the result of the parts that were available then more than the tubes themselves. The early designs used noisy carbon resistors, puny power supplies and the transformers were far from flat. Some colourations crept in with time. Capacitors dried out, connectors and switches corroded and pots became scratchy. Much of what people attributed to tubes was actually everything but the tubes.
From Hi-Fi we knew the methods to pursue. The first goal was to use valves. Simple tube amplifiers are inherently more linear than transistor amplifiers. We knew the common transistor design practice using large amounts of negative feedback to correct linearity damaged the transients. We did not have to resort to feedback to produce a good sounding circuit, but instead we were able to use much less feedback and in better ways. The second goal was to minimize the circuitry. Simpler circuits always tended to sound better. The third goal was to use better than common parts. Better from the standpoint of reproducing music. This required some research, listening tests and willingness to spend more money because the good parts cost more. In the case of transformers, we had to wind our own. We even assemble our own RCA jacks and have wire and tubes custom made. Details like this rarely cost less.
The result was a very respected line of power amplifiers, preamps and digital converters using valves. Vacuum Tube Logic became the equipment of choice among many of the hi-fi reviewers personal reference systems. They applauded a "new" sense of 3 dimensional reproduction, far more accurate tonalities, exciting transients and emotional impact. As more hi-fi companies jumped towards tubes, VTL held onto its reputation because its amps were still far better, particularly in the low end and reliability. VTL is now owned by David's son Luke Manley and the reputation continues.
In the midst of VTL's leadership in hi-fi, David was getting approached by studios to create his own versions of vintage pro gear. This was a field David could enjoy even more than hi-fi. His background as a recording engineer, studio owner and audio manufacturer made it impossible to resist. In very little time, he had designed a number of new products, pulled out a few of his old designs and made arrangements to legitimately re-issue some of the best classic designs. VTL arrived at its first AES convention in LA in 1990 with over 20 products. Shortly after, David put his family name on the new professional products and MANLEY was born. In 1993, David sold VTL to Luke, and started a new 11,000 square foot factory with 30 employees to focus on the pro market. Since then, MANLEY LABS has become a significant force in pro audio. (ed. note: David Manley left Manley Labs 1996. EveAnna Manley took over the company then.)
There are two main types of tube circuits Manley employs in his designs - if we ignore power amps. The first can be called single-ended. Variations of this are used in the MANLEY PULTEC EQs, ELOP® LIMITER, MIC-PREAMPS, D to A CONVERTERS, and DEESSERS. The basic design uses two tubes to amplify both voltage and current. The two tubes are actually 4 separate triodes or gain stages. The first 2 triodes provide voltage gain and the last 2 are arranged as a "totem pole" for current gain to drive low impedance loads. Our line driver has a typical output impedance of 50 Ohms and can reach +30 dBu easily. Generally, we use transformers to provide high performance balanced inputs because single-ended circuits are un-balanced in and out. For many years we have only provided un-balanced outputs and considering the calls we get perhaps some explanation is in order. Balanced inputs provide the hum rejection through common mode rejection. They do it just as well when they are fed balanced or un-balanced outputs. For the final output capacitor we use very expensive RELCAP MultiCaps®. These parts alone costs more than many of our competitors unit's entire parts budget. To make the outputs balanced we have to add an output transformer. We have avoided going this route because of the "sonic personality" virtually all transformers add. Finally, after many attempts, we have made one that meets our specs in every way. In 1997 we will begin to update the products that can use them. The first product to be updated will be the ELOP LIMITER which will get balanced inputs and outputs (4 transformers) for the stereo unit.
The second circuit Manley uses is called "fully differential". This circuit is similar to a "balanced line" where there are two wires carrying opposite polarity signals. Fully Differential designs require twice as many parts. In a balanced line the advantage is cancellation of hum. In a fully differential circuit there is cancellation of hum, distortion and power supply fluctuation. We use this in the VARIABLE MU® COMPRESSOR LIMIER and MIC EQ-500 COMBO. Each of these has transformer balanced inputs and outputs and the transformer provides much of the internal balancing. Absolute balancing of the tube circuits is very important or the two halves become unequal. Part of the key is tube selection and part is the trimmers we provide. The "BALANCE ADJUST" that we have on the front panel of the VARIABLE MU is one such trimmer. We also have "test points" on the front panel. Optimum performance is assured by adjusting this trim so that a null or zero DC volts occurs between the test points. Normally, a user will not need to worry about these or adjust the trim until a tube is replaced. If the trim is "out" one may expect more distortion or "roughness" than expected. There are also internal trimmers in each of these circuits so that a tech can optimize the performance of the unit as tubes are replaced.
It is difficult to quantify the differences sound-wise between these two types of tube circuits. The single-ended circuits are very quick, dynamic, and brutally honest (much like David was). The fully differential circuits are smoother and seem to enhance the music with space, width, and richness. Both are superb with acoustic and electronic sources. One engineer commented that "all Manley stuff has a boost at 5K" another said it boosts mids. We are proud that our stuff does sound different from solid state. It was not easy telling them that everything measures flat beyond 20K and that they were hearing transients accurately probably for the first time ever . One can demonstrate this quality with voice and latin percussion. Sibilance stays sweet and natural without getting harsh and difficult - percussion instruments stay real sounding and don't blur to where they sound like pink noise. Acoustic pianos sound like pianos not electronic. The tools you need to show this are a MANLEY MIC and MIC-PRE. The better the monitors, the more obvious the demo. . Take a few minutes to set up the mics and this should require a couple of trips back and forth between the studio. Record one mic into the left side of a DAT and the MANLEY rig into the right. Keep this tape. Play it for the skeptics. You could make DATs comparing MANLEY to vintage gear or the best in solid state.
The single-ended mic preamps have a variable feedback feature disguised as a gain switch. This is not a simple pad. It does change the "character" of the sound in interesting ways. In the low gain settings (40dB), (most feedback) noise is lower, distortion is lower, and the sound becomes more like solid state. It seems slower, smoother and perhaps appropriate for an old Nashville sound. The high gain settings (50 dB), are more aggressive and "in your face" and more of what many people expect of tubes. The middle settings are usually where the switch spends most of its time. Here, the preamps have the best of both worlds and seem the least coloured. It is interesting to note that the optimum amount of feedback is neither none nor plenty. We provide a feedback switch in many of our power amps too. In every product the amount of feedback is very carefully optimized to be reasonably low and less than 10 dB. Distortion is always below perceptible thresholds.
Our Tube Direct Interface is different. The desire was to provide some colours and distortion resembling the basic "clean" amp sound we are accustomed to. This is a different style for us. The "CONSOLE BOOST" and "UNITY" refer to power gain, not voltage gain. Each gets its signal from a different part of the circuit and has a different sound. UNITY is somewhat richer and rounder. The EQ switch rolls off a controlled amount of low end much like amps do.
The power amps are what made Manley famous. It is difficult to convince a studio person to give them a try. Several reasons - they haven't seen them advertised, the cost is higher, fear of tubes and they have no idea how much better their monitors can sound. The only way to sell them is to deliver the amps and let them use them for a week. The improvement nullifies most of the previous concerns. As far as maintenance - explain these are not like guitar amps - the tubes typically last 5 years or more. Better yet, explain that their speakers will last 3 to 4 times longer. This offsets the initial cost because they could easily be spending a few thousand in recones every year. Remind them that potential clients judge a studio often by listening to the speakers and that one or two new bookings covers the cost. These are amps to keep - not keep on trading in every five years like solid state.
ON THE TOPIC OF "TUBE SOUND" - TRUE OR FALSE ?
1) Tubes produce more audible distortion than solid state. False!
2) Solid state is perfect, clean and uncoloured. False!
3) Digital audio is perfect, clean and uncoloured. False!
4) There is a certain "colour" vintage tube gear has and that colour is "warm". False!
5) Tube gear requires more maintenance. False!
6) Tubes tend to have a more useful distortion than solid state. True!
7) Tube gear is often useful in making digital sound better. True!
8) Musicians like the "sound" of tubes. True!
1&2 - Fact - Transistors are not as linear as tubes. Transistor circuits use massive amounts of feedback to "bend them straight" and to produce low "measured" distortions. The audibility of distortion has been tested many times and MANLEY tube circuits are well below the thresholds. Williamson published in 1947 "more than 20 dB of feedback damages transients". Solid state designers prefer to ignore this fact. After all, test benches can not measure transients. We say - test benches don't make records. Much of the musical signal is in the transients. There were several AES papers describing an effect called "Transient Intermodulation Distortion" in op-amp circuits that is now referred to as "The great TIM scare of the '80s". This distortion is still very real but many companies would prefer us to forget it. "TIM" sounds like harsh sibilance and cold nasty highs. It makes shakers sound like pink noise and tends to wipe out any 3D depth and imaging clues. It is possible to build a tube circuit to sound like this but who would? It is also possible to build a solid state circuit that is linear, has little feedback and has a musical sound but we only know of a few examples or less than 1% of all pro products. Another boogie man of solid state is crossover distortion. This problem is biggest when the signal is the smallest and crossing zero. Class A designs are typical in tube products but rare in solid state. Try padding down a line level music signal 30 dB and feeding it into a few tube and solid state mic -preamps. You should compare the source to the preamp output. Notice which passes the music best. Proof that tubes can be cleaner and a fascinating demo.
3) Digital audio is perfect, clean and uncoloured, False.
Digital audio is a good thing but not perfect. Digital storage has its advantages compared to analog tape and vinyl records. Audiophiles generally prefer analog over digital but the gap is slowly closing. There are a few big problems with digital. The biggest is the marketing departments that promote that it is flawless. Digital audio has only become reasonable because users have been demanding that it be improved and companies don't want to be left behind. The well known problems are sampling resolution and sample rate. There now exist converters that address these issues with 20, 22, 24 and even 27? bit conversion and with rates up to 96K. Now, all we need is machines that record and play it. Most digital is still 16 bit sampled at 44.1 or 48kHz. Dither has crept into our vocabulary lately. That is the "distortion" that reduces our 20 bit data to 12 or 14 effective bits as soon as you add a good length of AES cable. Another huge problem is that most converters are designed by digital engineers. They like solid state and are not aware it can sound bad. They have a tendency to rely on op-amps rather than discrete gain optimized circuits and design for the test bench (and ads) more than for the music. Digital audio will no doubt continue to get better as converter designs, storage mediums and DSP evolve. It is worth remembering that microphones, speakers, and human hearing is analog - only the purely electronic sections can be totally digital.
4) There is a certain "colour" vintage tube gear has and that colour is "warm", False.
If one has used various vintage gear, the obvious fact is that it all sounds different. Much of it is not very useful, some of it is truly wonderful, and some can be called "warm sounding". This usually due to a loss of highs, sometimes due to low frequency boosts and distortion from transformers and sometimes because it doesn't have solid state distortions. Another point here is that some vintage solid state gear like Neve EQs and UREI LA-3s can be called "warm". That is probably because they have simple class A discrete circuits and good transformers with plenty of line driver current available. Most of the sought after "warm sounding" products , both tube and transistor, were the result of solid engineering and extensive listening and session tests. These classic products were not designed to colour the signal or add distortion - quite the opposite. Not only does vintage gear sound different from each other - modern gear as well. We adhere to the design philosophies that made good gear classic.
5) Tube gear requires more maintenance, False.
Ever notice that most of the vintage tube gear in most studios is still operating 30+ years after being built. Notice how few pieces of early solid state gear is in control rooms. Which would you rather fix? Hint - you can still get the tubes. Hint - your father probably used to fix your TV and radio. Hint - you probably won't have to plug in the soldering iron. Reliability comes from good engineering but a tube is easier to replace than a transistor. OK, IC's in sockets are easy to replace but one has to still call in a good technician with schematics to do it.
6) True, tubes tend to have a more useful distortion than solid state.
Guitar players have never left valve amps. They rely on "musical" distortion. Why do tubes seem to distort musically? There have been explanations that suggest it has to do with even and odd order harmonics, and soft clipping. In guitar rigs the answer is pretty complex. One has to look each stage clipping and successive stages distorting that clipped signal - plus the tone circuit, the power supply, and output transformer - plus the speaker frequency, distortion and compression characteristics. The major overlooked part is that the tone circuit boosts the upper mids quite a bit, then follows most of the distortion, then the speaker rolls off the top end. This in effect, attenuates the higher harmonics of the distortion which we tend to be unpleasantly sensitive to. It would not be such a bad idea to "weight" distortion specs so that each "harmonic" was multiplied by its "number" to arrive at a final meaningful spec. Most solid state circuits produce fantastic distortion specs but when they do clip, it usually sounds disgusting. Remember the feedback that is used to correct for non-linearity. Clipping is non-linear and the feedback fights to correct it but cannot. The circuit stays very linear until the threshold of clipping then abruptly runs out of headroom. The abruptness causes more upper harmonics to be created. The transistor circuit recovers from clipping less gracefully too because the feedback requires time to stabilize. Odd or even harmonics ? Symmetrical push pull There are other time domain distortions in solid state that become even worse near clipping. Manley designs use very little feedback and provide unusually high headroom. We minimise low level and time based distortion but understand that when our circuits are driven to clip that they should still sound "good' and musically useful. We allow for "artistic license".
7) True, tube gear is often useful in making digital sound better.
Much tube gear does seem to improve the sound of digital audio. There are probably as many reasons as there are tube products. Generally it helps to avoid too much of any one technology in a signal chain. The result of a purely digital chain is usually a purely "digital" sound. If every component can colour the tone then it becomes akin to cooking. You probably avoid saturating your food with sugar or salt or hot spices because it can easily become inedible. If cooking analogies seem farfetched, try oil painting with only a few shades of blue. This principle holds true in electronic design, acoustics, and mixing and probably most artistic endeavors.
8) True, musicians like the "sound" of tubes.
Could the reason that musicians like the sound of tubes, be because they are extremely familiar with the real sound of their instruments? Is it because tubes reproduce the subtle nuances and dynamics better? Does the emotive "feel" get lost in op-amps? Does a good tube mic capture the details better? Does the singer hear the words with the care she put into them? With Manley gear the answer is yes. With tube gear designed to distort or tone-tint the answer is "not quite" and these questions might have to be avoided. Manley gear seems to want to create the illusion of "no electronics", "no mic", "no window" where the performance is taking place in front of you and touching you heart and spirit directly. Musicians like that.
Most of the funny misconceptions we hear about tubes started off with journalists "taking a guess" and got amplified by ads designed by companies that jumped on the tube bandwagon. Most of these companies copy old designs or employ outsiders to come up with a product that meets the criteria meant for the ad. The cloners usually have no idea why people want a copy of some box but they can build it (more or less) and it will sell. The bigger companies swear by solid state but saw some of the market reach deep into their pockets for tube gear. They wanted part of this trend. If fashion changes - so do they. It is both frustrating and funny. All we are trying to do with this letter is set the record straight and print our opinions based on many years experience with tubes in audio.
Finally, one of the most important reasons Manley designs sound the way they do is that they come from "purist" thinking. Tubes are practical in simple circuits. As the feature list grows, so does complexity and discrete transistor circuits become a viable choice. To lower the cost and add more knobs, op-amps are a good choice. The next step in complexity and lower cost is digital audio and DSP. Now, think about the phrase "less is more". Each step in complexity lost some valuable musical information. Manley prefers the approach of fewer gadgets and more sound. We use better parts rather than more parts. Passive over active. Quality sound is the #1 feature and the only feature that will stand the test of time.
Manley approaches audio equipment with the opinion that every individual part in the signal chain changes the signal. Some of these are major and many are subtle. There is no "perfect" device - each imparts some of its own personality. We make huge efforts to minimise any negative effects on the music. Some of these are easy to measure, some difficult and some seem to be not measurable yet. Transient detail, 3 dimensional imaging and emotional pull are not buttons on the test gear. That's OK, test benches don't make records.
We believe that our intelligent customers care more about sound quality and day-to-day usefulness than whether the box uses tubes, transistors, ICs or pistons. We use tubes because we also strive for sound quality.
Our products are not inexpensive because we always put quality over quantity. It is unlikely that anybody will produce analog products with amazing sound for low budgets. Quality parts cost more, hand wiring costs more, transformers cost more than ICs and then there is those Manley power supplies.... However, if and when we do find ways to improve our products sound and price, we will.
For the purpose of fluid communication, we often refer to our products in comparison to vintage well known gear. For the record, we do not copy designs - old or new - without written agreement of the original companies and then we always credit the original. However, we do improve where we can. We appreciate those who have gone before us and respect the best sounding designs. We appreciate and respect our customers and dealers and always will. Let us know your opinions - enough of ours. Thanks.
For more in depth discussion on distortion, tubes, and transformers, read Tube Talk 2 the Manley Mantra.