If we consider that the audio signal in a playback system begins at one or more sources and proceeds, eventually, to the speakers, then we find the preamplifier in a special position. It is the first component that is common to all of the sources. Dividing the system into front-end and back-end, the collective ensemble of sources forms the front-end of the system, and the line-up of preamplifier, power amplifier(s), and speaker system forms the back-end. Looking more closely at the back-end, the preamplifier is for source selection, volume control, and perhaps a little bit of signal gain, the power amplifier is for driving the speakers, and the speakers are for converting electrical signals into pressure waves in the listening room. Of the three back-end components, we should have the highest expectations of the preamplifier because its job is the least difficult. Some get this backwards and hold the preamplifier to the lowest expectations, thinking that any preamp will do. This is a grave mistake. Any error the preamp makes will be impossible to eradicate downstream, no matter the hefty prices paid for the rest of it. Because its job is simple and, in fact, easy, we have every right to expect it to be virtually perfect, don’t we? And, if one component has opportunity to be perfect, should we not demand that it be perfect so that at least one box can be checked off, in the face of so many other variables that are hard or impossible to nail down completely?
What makes the life of a preamplifier so easy? It is dealing with a small input signal, about 1 V. It is usually expected to have only a small amount of gain, say about 10 dB, which means the output signal is not expected to be more than about 3 V at full volume. Being that we don’t usually listen with the volume at max, the output voltage is often much lower than the input, perhaps only about 0.3 V. Furthermore, it is driving a power amplifier with an input impedance of about 30-100 k-Ohm, so it only has to produce about 10 uA (10 millionths of an amp) or less of current, unlike the 1-10 A or more that a power amplifier has to deliver. It should be perfect.
For the purpose of further discussion, let us consider the topic of voicing. We consider a voice to be any distinct quality of the sound, any signature that superimposes itself over the sound in a recognizable way. A voice can be intentional: a signature sound that a manufacturer wants their equipment to be known for. An example of this is the “tube sound” that some are very fond of. Alternatively, a voice can be an unintentional and unfortunate quality that is not necessarily desired, something that creeps surreptitiously into the sound of a piece of equipment. We could think of a voice as a guest, welcome to the party or not, that adds its own twist to the tenor of the evening. By extension, voicing may be thought of as the imparting of a particular voice to a piece of equipment. The same word has been used to describe a design process that either imparts a desired voice or that eliminates all traces of a voice, the latter rendering the equipment completely neutral. The concept is useful in the description and understanding of audio components in general, not just preamps. In recognition of variances in taste and system design procedure, it must be conceded that only neutral voiced equipment can deserve the title of perfect, for neutrality alone is singular and objective.
There are many ways in which the preamplifier can fall short, drag its feet, or fail to measure up to expectations of perfection. Some of these are summarized below:
Capacitor Voicing
When capacitors are present in the signal path, the voice of those capacitors are imposed on the signal. A great many very expensive capacitors have been developed to minimize this problem, or for the explicit purpose of tuning the complete circuit for a given desired end-to-end “sound”. There has never been a capacitor that does not have any voice, so when they are implemented, there must be very careful listening and, perhaps, compensation, without which the capacitor voice will be added to everything you listen to through the preamp. It is possible to eliminate capacitors from the signal path of the preamplifier, depending on the architecture, though some manufacturers choose not to do this.
Resistor Voicing
Unlike capacitors, resistors are impossible to completely eliminate from the signal path. Every resistor has its own voice as well. The degree to which the resistors “voice” the signal is dependent on the architecture of the preamp, and on the quality of the resistors used. As with capacitors, very expensive resistors exist, and some of these come extremely close to perfection. It is less common, though not unheard of, to use resistors to intentionally impart a voice to the sound.
Attenuator Voicing
The attenuator is the technical name for the volume control. It is made of an array of switches and resistors. Here again, the manufacturer must make choices that will affect the overall voice of the preamplifier. Should they intend to create a voiceless product, they will be driven to use expensive resistors.
Tube Voicing
This is probably the example of voicing most people are familiar with, whether or not it is their cup of tea. The sound of tubes is a much debated asset, but those on either side of the fence must admit that tubes offer a very convenient way to voice a component, as tube rollers will adamantly assert.
Noise Infection
Some noise enters the preamp through the AC power cord (the AC grid is very noisy), and some is created within the circuitry. Virtually every component in the circuit is a source of noise. Choices of architecture and components determine how much noise is filtered from the AC connection, how much is created within the circuit, and how much is shielded so that the noisy power supply cannot infect the sensitive amplification circuitry. Noise is of particular concern in preamplifiers because the signals being processed are very small in amplitude and are easily degraded by very small amounts of noise.
A bit of digression is in order. While some voice effects are intentional, noise is universally understood to be bad, and the designer/manufacturer should task himself with reducing it as much as possible. Voicing, on the other hand, falls into two distinct philosophical camps. Some believe that a component should have a signature or voice. They might say something like, “I love this company’s preamplifier because it is warm and sweet.” In the other camp are those who believe that the preamp, or anything else for that matter, should be perfectly neutral. Their motto is, “If it is there to amplify, let it amplify, and nothing else.” This camp enjoys a distinct advantage over the other, in that a system comprised of neutral equipment can be upgraded with a single neutral component without changing or imparting a voice. Not so for the first camp, for they must consider that they have a sweet preamp and a dry source, for instance, and they are bound to keep it that way or start the tuning all over again. Many non-neutral components can be strung together to achieve a particular sound, but that sound is likely to change if any one component is replaced or upgraded, and reacquiring the desired sound can be a daunting task thereafter. This is not to trivialize the process of finding neutral equipment, for it is not trivial at all.
Manufacturers use selection of internal components, resistors and capacitors, or even wire, for example, to achieve the “sound” they are shooting for, be it a particular voice or complete neutrality. How is one to know what voice, if any, his preamp has? This is where the investment (or borrowing from a good friend) of a high-quality headphone set and headphone amplifier comes in extremely handy. One example of such a headphone setup is made by Stax, and it comes with a tube amplifier. The quality is of a sufficiently high level to reveal small nuances in the following test. Bear in mind that absolute accuracy is not important here. What is important is a high degree of resolution so that very subtle differences can be heard.
With the preamp installed in your system, power off your amplifier(s) and remove the interconnects from the amplifier input jacks. Connect these interconnects into the input of your headphone amplifier instead. You will need a pair of couplers that allow you to connect the input interconnects directly to your output interconnects, eliminating the preamplifier from the signal path. The couplers we are taking about are short, solid barrels with a connector at each end. Do not use coupling cables, because the introduction of cable length will corrupt the experiment. If you are using XLR cables, you need a pair of female-to-male XLR couplers. If you are using RCA cables, you will need a pair of female-to-female RCA couplers. Pull the interconnects from the preamp and connect them together with the couplers. Use the volume control on the headphone amplifier to achieve a desired volume. Begin by playing a piece of music through the headphones, paying careful attention to every aspect of the sound (sound stage, timber, dynamics, etc.). Next, remove the couplers and connect the cables to the preamp input and output jacks, making sure to match the previous volume by using the preamp volume control (not the headphone amplifier volume control – it should be left as before). Listen to the same piece of music again, carefully noting any differences in sound that you are aware of from the previous listening. Expect to have to repeat this procedure a few times, until you home in on the differences, if there are any. Differences in sweetness, dryness and soundstage size can be due to imparted voicing from the internal components. Differences in soundstage size and “magic” or “bloom”, particularly reductions in these when the preamp is in circuit, can be the result of noise pollution within the preamp. Understanding all of this is perhaps academic. What is important is that, if you belong to the neutral component camp, you want to hear no difference. To the extent that a simple pair of cable couplers should reasonably be expected to be neutral, the preamp should sound identical to the couplers. If you think your preamp should sound sweet, then it should sound sweet compared to the cable couplers, et cetera. Audiophiles usually differ around sweetness and resolution, but tend to be unified with regards to soundstage size and background quietness (blackness). Simply put, audiophiles tend to universally prefer the largest possible soundstage and the blackest possible background noise level. Most people will not be pleased to discover that their preamp shrinks the soundstage or pollutes the background.
Understandably, most people will not go through this process. They will, instead, simply insert a component like a preamp, and ask themselves whether they like the sound of the whole system more or less than with the previous one. For those who are more intentional about this, they will find that the headphone experiment allows them to better understand what they have now and what they may be auditioning. If what they have is of the tube variety, they may take the opportunity to voice their preamp to what they think it should sound like, be that neutral or sweet, or whatever. Admittedly, if one gets the whole system to sound the way he wants it to, this could all seem academic. If, on the other hand, one is after a completely neutral system or one that is nearly neutral, this subject is far from academic. The reason for this is simple: It is far easier to build a neutral system out of neutral components than to achieve neutrality with a lineup of non-neutral components that are hoped to perfectly compensate for each other.
Philosophically speaking, one might consider a focus on the preamplifier unjustified unless a similar understanding of all the other components can also be had. But, consider that, if a neutral system is desired, the preamp is the component most possibly capable of being perfect in that regard. This makes it a good place to lay first and lofty expectations. Should this seem like an invitation to save money on the preamp, be advised, it is not. Rather, it is council to pay the price for perfection where it can be had, and to expect nothing short of it after doing so.
If a voiced system, that is, a system that is not neutral but that has a certain signature voice, is desired, a further plug for neutral components is nonetheless respectfully submitted. It can be far easier to voice a system whose voice is to come from a single piece of gear than to achieve the voice from a complex lineup of individual voices. For instance, one may decide to use a tube DAC that has just the right amount of sweetness, and run neutral equipment all around it. If a change of voice is ever desired, he knows exactly which component to change in order to achieve it. Should any other piece of gear need upgrading, it is a neutral piece and would be replaced with a better component that is also neutral. This way, upgrading doesn’t have to mean revoicing. “How does one go about finding such neutral components?”, one might ask. It is no easy task at the outset, but the further developed the neutral system becomes, the easier it all gets. This is because, once achieved, any further upgrades can be expected to improve the system without revoicing it. This becomes an objective expectation that is easily verified with component substitution.
The information contained herein is presented in good faith and with all hopes of helping the reader to find the best equipment for an audio system. Neither the author, nor Bella Sound LLC, assumes any responsibility or liability due to loss on the part of the reader, owing to or believed to be owing to the content of this article, or resulting from or believed to be resulting from an application of the information contained herein, regardless the level of expertise engaged thereto.
