Having discussed Power and Nonlinearity Distortion in the previous two parts of this series, we will now turn our attention to the remaining amplifier characteristics of interest: Gain, Frequency Response, Noise, Input Impedance, Damping Factor, Idle Power Consumption and Voice. Admittedly, Voice is not a specifiable characteristic, but it is a very important concept and is being mentioned for completeness.
Gain
Gain is the degree of amplification, the ratio of the output voltage to the input voltage. As such, it is the most fundamental aspect of the amplifier. It is, however, rarely of interest because modern amplifiers have nearly universal gain.
Gain can be expressed as a voltage ratio, or it can be expressed in “dB”. We have the following definition and conversion formula:
Gain = Vout / Vin, and
Gain (dB) = 20 Log (Gain)
Gain is almost invariably 28.3, and was chosen by the industry so that 1 V at the input would produce 100 W of power into 8 Ohms. That makes the standard Gain (dB) to be 29dB. Of course, small variations around these numbers are going to be encountered, but they are inconsequential.
Frequency Response
This is the range of frequencies over which the amplifier will respond without prejudice, and nearly always includes the entire frequency range of the human ear, 20Hz–20KHz. The specification usually gives the amount of drop off at the lower and upper limits, most typically 3dB, though 1dB is sometimes stated.
Most modern amplifiers exceed this range considerably. Though not intuitively, there is value to exceeding the range of the human ear. The reason for this is complex, and includes considerations such as phase distortion and nonlinearity distortion.
Phase distortion is related to time delay, and time delay errors can upset the calculations of the brain in its effort to determine exact location and nature of sound. Unfortunately, phase distortion sets up about ten times earlier than drop off. That is, if the upper limit of frequency response is 20KHz, the associated phase distortion begins around 2KHz, which is right in the central part of the human hearing range. For this reason, we would prefer the upper limit of frequency response to be about 200KHz. This limit is rarely achieved, but something around 100KHz is fairly common. Phase distortion is not very consequential to the listener at the lower limit of frequency response, and so we can simply ask that the amplifier handle the lowest frequencies of music, making 10Hz a good number to think about.
Nonlinearity distortion, discussed in Part 2 of this article series, almost universally increases when the signal approaches the upper frequency at which phase distortion begins to appear. This gives a second reason to prefer that the upper frequency limit be around 200KHz, if attainable.
Noise
Noise can be thought of as all of the power produced by the amplifier that is not correlated to the music signal itself. Think of an air conditioner firing up while you are trying to listen to a soft pizzicato passage of a symphony. It is not related to the music at all, but it is distracting, if only in a subconscious way.
There are four ways noise is typically specified: Output Noise Floor (or simply Output Noise, specified in units of nV/Hz), that is, nanovolts per root Hertz), Input Referred Noise Floor (more commonly just Input Noise, specified in units of nV/Hz, that is, nanovolts per root Hertz), Output Noise Voltage (specified in units of uV, that is, microvolts), and Signal to Noise Ratio (specified in dB, referenced to full output power). These four rating are related to each other and very nearly communicate the same information.
Input Referred Noise Floor = Output Noise Floor / Voltage Gain.
Output Noise Voltage is usually specified from 20Hz-20KHz, it is all the noise over the entire audible spectrum, represented as a single voltage. In most cases we can state to good approximation that:
Output Noise Voltage = 141 * Output Noise Floor.
Signal to Noise Ratio = 20 Log (Vout max / Output Noise Voltage), where
Vout max = (Load * Maximum Power into Load), where
Load is the reference load resistance into which the maximum power is specified.
For example, consider an amplifier with the following spec’s:
Output Noise Floor 50nV/Hz
Gain 28
Maximum Power 500W into 4 Ohms
Then,
Output Noise Floor = 28 * 50nV/Hz = 1.4uV/Hz
Output Noise Voltage = 141 * (1.4uV) = 197uV
Vout max = (4 * 500) = 44.7V
Signal to Noise Ratio = 20 Log (44.7V / 197uV) = 107dB
The foregoing conversion formulas come in handy for comparison shopping, since any one of the specifying formats may be used.
What is important to realize is that modern amplifier noise is rarely audible at the listening position, even when highly sensitive horn speakers are used. One may reasonably ask, “Why then should I care?” Due to subtleties in the way that human hearing works, it is possible for small amounts of noise to obscure very low-level information in the music signal, even though the noise by itself cannot be recognized or even heard. The degree to which this can happen depends on where in the amplifier the noise is generated and the degree to which the noise mixes with the audio signal. This gets into the topic of non-linear noise, but that is beyond the scope of this article. Notwithstanding, less noise is better, as a rule.
Input Impedance
This is a measure of the load that the preceding component, usually the preamp, has to drive. It is expressed in k-Ohms (more typically just k), and may be anywhere from 10k to 100k (where k denotes kilohms), or more. There is no exceedingly important aspect to this number, unless the preceding component is tube based. Tube preamps, DAC’s, etc., typically couple the output signal through a capacitor, and very often the value of this capacitor is 0.47uF. Together with the input impedance of the amplifier, these two values will determine the lower frequency limit of the preamp-amplifier coupling. For aforementioned reasons, we would like the component coupling to allow us to respond at least down to 10Hz, and this constrains us to amplifiers whose input impedance is at least 33k.
Damping Factor
We would like our bass to be tight and controlled, not loose and floppy. The amplifier and speakers work together to achieve this, and the amplifier’s contribution is referred to as Damping Factor. Think of this as a kind of brakes, to stop the speaker diaphragm movement instead of allowing it to ring on, after the signal has said, “Stop!” The definition of Damping Factor is not very illuminating, so we will suffice it to say that the higher the number, the better. Damping Factor is always accompanied by a reference impedance, for instance, @ 4 Ohms, or @ 8 Ohms. It is easy to convert to a different reference impedance, as follows:
To convert 8 Ohm Damping Factor to 4 Ohm Damping Factor, divide by 2.
To convert 4 Ohm Damping Factor to 8 Ohm Damping Factor, multiply by 2.
Example 1: An amplifier specifies Damping Factor as 200 @ 8 Ohms.
The Damping Factor @ 4 Ohms would be 100.
Example 2: An amplifier specifies Damping Factor as 75 @ 4 Ohms.
The Damping Factor @ 8 Ohms would be 150.
Damping Factor is more usually specified into 8 Ohms because it results in a higher number. The rest of our discussion on Damping Factor is with an assumed impedance of 8 Ohms.
Typical values for solid-state amplifiers are in the 50-200 range, with values beyond 200 being essentially of no further sonic significance. Tube amplifiers, on the other hand, rarely achieve damping factors of 50, and many are closer to 20. Thus, solid-state designs have the upper hand here, and careful choice of speakers must be made for use with tube amplifiers. Where tube technology is preferred and tube-unfriendly speakers are also preferred, one should consider a tube amplifier that has a large number of output tubes, or equivalently, simply require a damping factor of at least 50. Of course, be prepared to pay dearly for this creature, both at time of purchase and whenever a retube is deemed necessary.
Idle Power Consumption
This topic is strictly for those who want to minimize their utility bill. It is not a trivial subject because most amplifiers reach their full sonic performance only after at least one hour of warm-up. To avoid the inconvenience of warm-up, some prefer to leave the amplifier on all the time. Should you be of that mindset, you may want to take this specification very seriously, as it could affect your utility bill significantly.
Idle Power Consumption is given in Watts, whereas the utility company charges for kilowatt-hours (KW-Hr). We have the following conversion formula for kilowatt-hours over a one-month interval:
KW-Hr = 720 * Idle Power Consumption / 1000, where 720 is the number of hours in a month.
A typical stereo amplifier may have an Idle Power Consumption of 150W. This comes to 108 KW-Hr over the one-month billing cycle. It is not a trivial effect on most utility bills, and it gets much more serious with large Class-A amplifiers and monoblocks. My personal opinion is that the sonic performance of the amplifier should weigh much more heavily than Idle Power Consumption, and the convenience of leaving the thing on 24-7 is well worth sacrificing in exchange for getting the sound you want.
Voice
This refers to an unspecifiable but undeniable sound signature imparted by the amplifier. Words that have been used to describe the Voice of an amplifier include: bright, edgy, smooth, polite, dull, boring, constricted, open, and the list goes on and on. It is impossible to know the Voice of an amplifier without trying it in your own system. This is because you cannot listen to an amplifier by itself; all you can listen to is the voice of the entire system. You cannot really know the Voice of the amplifier alone, but you can discern if the amplifier Voice is helpful to your system or harmful. Perhaps you will hear no change in system voice when you audition an amplifier. This is probably a good thing, though an improvement to system voice is obviously a better thing.
Many would be inclined to take exception with this point, claiming that an amplifier should have no voice at all. I very much agree with this position, but careful comparative listening would be convincing that very few amplifiers actually live up to this ideal. Unless one can find such an amplifier, he should at least make sure that the Voice he ends up purchasing is compatible with and ingratiating to the rest of his system.
In Summary
Amplifier specifications are helpful to understand, and can allow some degree of comparison shopping to help narrow down the daunting task of auditioning. They do not obviate the auditioning phase – truly, nothing can. The foregoing discussion of amplifier specifications over the last three articles is not really given to save the reader the trouble of listening, but to give the reader a profound respect for just how important this component is to his or her system. The greater the respect for this component, the greater the care and effort that will attend selecting it, and the greater will be the enjoyment of listening to it for a long time thereafter.
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, assume 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.
