EVERYTHING, WE'VE PERVISED

Clearly, without an amplifier there is no life. We begin to choose and, naturally, the first thing we look at is the maximum (we already know what it is) power, what are we fighting for? How to choose it will be a separate and unexpectedly short conversation. But first, let's determine where this power comes from. What makes a separate amplifier such a qualitatively different device compared to the inherited amplification channels built into the head unit? From the previous issue of “V.V.” It became clear: it’s all about nutrition. The amplifier creates an alternating voltage at the output with a swing, from top to bottom, no greater than the supply voltage of the output stages. For the radio amplifier, this is the voltage on board, 12 V when the car is turned off, about 14 V when running. Home component external amplifier - power supply. It receives a constant voltage from the on-board network, turns it into an alternating voltage of a fairly high frequency (tens of kilohertz), the alternating voltage can already be increased using a transformer, which is what the amplifier’s power supply does, and then, already increased, it is rectified again and supplied to the amplifier itself. To how many volts the voltage was inflated during this activity, at such a height the ceiling of the output voltage swing will pass. What follows is simple arithmetic. Let’s assume that out of 12 V on-board the power supply has created 50. In reality it will be two voltages of different polarities, 25 V each, it’s more convenient. This means that the output voltage swing (in each direction) will be no more than 25 V minus some pennies lost in the transistors. Maximum output power it turns out to be 25 squared divided by the load resistance. This is according to Ohm's law, it is inexorable. It comes out to a little more than 150 watts. Only this is the peak value, on the RMS scale - exactly half as much, about 75 W. The numbers are quite real, there are plenty of such amplifiers. Can you get more out of this amp? The first stage of “afterburner” in many models will occur by itself, as soon as you start the engine. For many amplifiers, the output voltage of the power supply is not stabilized and is in proportion to the input. And when, with the engine running and the generator running, the voltage on board becomes not 12, but, on a well-regulated machine, 14.4 V, the voltage at the output of the power source will increase from 50 to 60 V, and the “ceiling” for the output voltage of the amplifier will also rise , and the maximum power will increase to 108 W. Wow, that's a raise, right? Just don’t rejoice too much just yet. Will this make the amplifier play louder? Why is this, exactly? The overall gain, from the signal source to the output, remained the same, it does not depend on the power supply (and if it suddenly did, the component responsible for this would urgently request permanent registration in the trash can), which means it will be as it played. Another thing is that if previously distortion appeared at some volume, when at the peaks of the signal the output voltage tried to jump over the bar set by the power source, now this moment will move to the area of ​​​​higher volume. How far will it move? Let's figure it out. One and a half decibels. One click of the volume, or even not a single one, it depends on the step of the control.

What have we gained compared to " past life"When there was no amplifier at all? There seems to be a lot of watts. And in decibels of the maximum undistorted volume, again, it seems, not very much: 5.4 dB. But this is only “seemingly”, as we will see later, happiness is not just in clicks of the volume control. It is still necessary to organize some kind of harmony between the powers. Look, for example, at what power the acoustics have, and select an amplifier based on that, right?

I did this on purpose, for the purpose of provocation. We talked about how you can ruin acoustics with insufficient power in the last issue, now let's try to do it with too much power. It will be much more difficult, I warn you.

Let's return once again to the phrase that I have said (and written) many times on various occasions, the last time in the last issue. Here it is: “And when we talk about the power of an amplifier, we are talking about what IT GIVES. And when we talk about the power of the speaker, it’s about what HE TAKES.” The maximum power of an amplifier is the one it cannot provide more than, because it begins to distort the signal, and that is not what we bought it for. The maximum power of the acoustics, therefore, is the one it cannot take more than, because WHAT? Does the signal also begin to distort? And it begins to do this at once and little by little, not at all like an amplifier; acoustics do not have a hard limit. In ancient times, there was a Soviet standard by which the so-called rated power of speakers was standardized. Special conditions, frequency bands, and so on were stipulated there; in general, the power was considered such that nonlinear distortion did not exceed 10%. The best bass speaker of that time was called 6GD2, the first number is just the rated power. There were 4 more gas engines, 3 gas engines, and so on, and then they adopted the definition of rated power, which no longer depended on distortion, but on survivability, and all these gas engines at once grew fat to 10, 20, 75, and the like. These GOSTs ordered us all to live a long time, and now power is determined differently, and it is very important to understand this in order to treat this indicator with the attitude it deserves.

I’ll ask you to type this in red, if I forget, you can use a pencil yourself, okay?

The power indicated on the acoustics is not the one at which it should work, but the one that destroys it.

Of course, there must be a relationship between the capabilities of acoustics and the resources of the source of this probable destruction, but this is a relationship, not an identity. Imagine: you bought a car with a maximum speed of 200 km/h. And you got tires with a speed rating of T (190 km/h). What, you can't drive? At 191 km/h all four wheels are in tatters? Or, on the contrary, the tires have a Z speed index (240 or more), and you are knocked off your feet when choosing the right car for such tires. Unreal.

Nevertheless, one often hears (and even reads) how acoustics are selected for an amplifier (and vice versa), looking first at the power, and then at everything else.

So let's do this one last time so as not to return to the question. The numbers on the speaker, accompanied by the words Power, without indicating what is meant by this, do not mean anything, it is part of a modern but deep-rooted tradition. If the speaker manufacturer is at least relatively correct in the figures he gives, then he can indicate the long-term power, and this is the maximum non-destructive (or minimally destructive, don’t forget about this) power supplied to the speaker for half an hour according to the scheme: a minute works - two resting. At the same time, a noise signal is supplied, passed through a filter that cuts off everything below 40 Hz and everything above 4 kHz, so this has almost nothing to do with the tweeter. Now, if the acoustics survived these most difficult half an hour in its life, the used power value is recorded. If it died, it is taken from previous experience with less power. Short-term power is one that will not destroy the speaker (or will destroy it, but at the last moment) after 60 cycles of “yelling for a second - resting for a minute.” All the described procedures imply bringing the acoustic sample under test as close as possible to the edge of the grave, so use them as a guide standard indicator for someone who paid for the acoustics out of his own pocket, it’s somehow not very reasonable. The only type of indicator that even slightly resembles the possible real use of its legal property is rated noise power according to the IEC 268-5 standard, when the acoustics must remain alive after 8 hours of continuous operation on the already mentioned noise signal. It is almost never indicated.

The landmarks here should be different; you shouldn’t look for them on boxes with acoustics.

LINKS, WHERE ARE YOU?

Our in-house experts in acoustics tests have repeatedly recommended (when manufacturers were completely losing their shame and it was unthinkable to remain silent) to measure up to indicators that at least roughly indicate the area possible values. For 6-inch component acoustics, the limits of reasonable risk lie somewhere at 40 and 90 W (this is wide, you already need to look at the design features inside), for 5-inch ones - naturally lower, 30 - 70 W. These are the rated noise power values ​​we consider. You may disagree, but refuting experiments - at your own expense, please.

The numbers, in principle, resemble the common values ​​of the maximum output power of widely used amplifiers, so the simplest, bordering on primitivism, answer to the question of matching the power of the amplifier with the power of the acoustics is already ready: a typical amplifier is suitable for working with typical acoustics. Anyone - with anyone. In principle, if you don’t want to worry, you can use it. But the answer is too simple to in any way claim to be exhaustive, that’s clear.

A slightly more common answer can be found in the experience of the superbisons of the acoustics world. The undoubted superbuffalo is the JBL company, which has equally succeeded in acoustics for home, stage, car and intended for sounding rooms and open spaces. The company's technical circular contains the following recommendation: in the case where the volume level is under control (it explains: this means a home or studio, but not a word about a car), the maximum amplifier power (RMS) can be twice the rated noise power. In cases where control is not perfect (this is about sound systems), parity must be maintained.

Then you need to look at the realities of life. In life, as I have reason to believe, both the amplifier and the acoustics will be used to reproduce music, and not test signals that are only very roughly similar to music. A music signal is not a sine wave or even noise, it is a signal with a large difference between the average value and the peak value. Short-term signal peaks, with rare exceptions, do not threaten the health of the acoustics, which mainly have to resist thermal load, and the heat generated at the voice coil is a function of the average level of the input signal. I had to see in the documentation of the most serious acoustics manufacturers how, next to the very real (and with all regulatory data) long-term power figures, the values ​​of withstand power at short-term (say,

10 ms) peaks. The numbers sometimes reached hundreds of watts, and this is no longer marketing, this is a fact, even a very powerful one, but a very short burst of signal will not destroy the speaker. But the amplifier has a fundamentally different view of level peaks. If the signal level exceeds the maximum power level even for a millisecond, it will be mercilessly decapitated, that is, it will go further along the wires to the acoustics in a distorted form, compared to the original source. This cannot be allowed. And here it makes sense to look at your musical tastes.

TASTES DO NOT MEASURE

Why so? You can also try. I ran a number of musical fragments through the computer and chose ones that were quite indicative in terms of the ratio of average (dangerous for acoustics) and peak (which should be feasible for the amplifier) ​​power. The signal level was measured in decibels relative to the maximum recorded on the disk, but for clarity, I recalculated everything as a percentage of the maximum power. The first picture is 60 seconds of “Procession of the Dwarves” (6th track “Let’s Test!”). If the system is configured so that the largest peaks of the signal do not exceed the amplifier's output power, then in general the speakers will receive about one and a half percent of this power during this minute. Even in those 12 seconds, when the orchestra is completely relaxed, the thermal load will be no more than half the power.

A minute of activity of the Yamato drummers (remember when they came to Moscow?). The signal level is chosen to easily miss the peak of activity at 21 seconds. As a result, the average power of the entire fragment is less than a percent of the maximum, and its most intense part is one tenth of the maximum.

Third example: “In the Pocket” (Kai Eckhardt, “NAIM Sampler”, track images/smiley_icons/icon_cool.gif. The average power is 13% of the maximum, and turning up the volume in a sincere attempt to ruin the acoustics will mean cutting off the numerous peaks caused by skillful drummer's job.

Don't listen to audiophile delights? We won't force you. Here is a fragment of the soundtrack of the punk rock band Kurban (Turkish and, by the way, quite interesting). Here already - yes, the guys on stage do not rest, and the average power for a long time is about 40, or even more percent of the maximum. But the guidelines, in principle, remain the same as those proposed by the bright heads from JBL, God bless them. It’s just that rock music falls into the category of “flawed control,” which is logical.

An attentive reader may be puzzled here: “Wait a minute, it turns out that we listen to music on one or two, many - ten watts, connected to the acoustics? Why then does it play loudly? You heard it yourself: it’s loud.” I’ll answer: why shouldn’t she play loudly? After all, you can easily manage decibels (even those who couldn’t before). We take any acoustics from any of our past tests and look at the sensitivity indicator. Well, let's say 87 dB, this is the average typical value. This sound pressure will be created by this acoustics at a distance of 1 m with a power of 1 (one and only) Watt supplied to it. This, by the way, is no longer quiet. In order for this acoustics to create a sound pressure level of 90 dB, standard for control listening in a sound recording, it only needs 2 W. Apply 10 W and get 97 dB. It's quite loud. Moreover, keep in mind that we have at least two such speakers, and they sound not in a quiet room, but in the cabin, where there are much fewer losses, and the reflected sounds come to us. What then, you ask, will the speaker do when those same hundred peak, say, watts come to it? Exactly what it should: briefly, within a fraction of a second, it will scream at 107 dB. Give him these 100 watts continuously, in the form of noise or, worse, a tone, and this scream will be his death cry. And so - everything is under control, don’t worry.

In acoustics, everything is measured differently than in the ordinary world. There are several reasons for this, some explanations can take you to the paradise of science, we will not touch them. Others are amenable to simple interpretations. Or they can simply be taken on faith, whichever is more convenient for you.

Human hearing cannot add and subtract. Just multiply and divide. Evolution (or the Creator, take your pick) arranged it this way, as it seems to me, guided by technical expediency. Hearing operates over a huge range of volumes. The sound pressure (measurable, as is known) corresponding to the pain threshold exceeds the sound pressure of the hearing threshold by ten million times (in words, so as not to count zeros). Hearing adapted to this by becoming (by the will of evolution or the Creator) logarithmic. People came up with logarithms later, but they sit in our heads by nature. The logarithmic nature of hearing is that it evaluates the difference in loudness not by how much greater the sound pressure is, but by how many times greater it has become. So (if we now remove all the intermediate chapters of history) a unit of measurement was invented on which absolutely everything in acoustics and electroacoustics is based - the decibel. Who knows everything about this, do not read further, however, I asked about this when I opened this series of publications.

I give the rest, no matter how many there are, the opportunity to master operations with decibels in five minutes and subsequently do it easily and gracefully. So: a decibel is a unit that, if added, means “multiply”, and if subtracted, means “divide”. For example: sound pressure is 3 dB higher. This means double. Another 3 dB? Twice more. More by 1 dB is approximately 1.25 times. 10 dB more - ten times. And vice versa: subtract 3 dB from the sound pressure, and this will mean that it has been halved.

It is enough to remember a few important values ​​so that from them, like bricks, you can form an idea of ​​what this or that value indicated in decibels means.

Here you are:

Power or sound pressure
differ in
Voltage
varies in
1 dB 1.25 times 1.13 times, just a penny
3 dB 2 times about one and a half times
6 dB 4 times 2 times
10 dB 10 times about 3 times
12 dB 16 times 4 times
20 dB 100 times 10 times

That's all: you meet, for example, somewhere around 18 dB, you estimate that it is 12 + 6, take the “times” for these two terms and multiply. You multiply, that’s the whole trick. In our example, 16 by 4 gives 64. Just pay attention: when comparing sound pressures and powers, you need to take “times” from the left column, and when comparing voltages, say, from the right one, this is a trick due to the fact that an increase in voltage, to for example, at the output of the amplifier, doubling the power leads to a fourfold increase in power (the voltage is squared there), and the decibels are the same, there are 6 of them. However, in the future we will mainly operate with powers and sound pressures, so the right column will remain in reserve for now .

What does decibel mean to the ear? A difference in volume of 1 dB (this is for most head units - one click of an encoder or volume button) is heard only by immediately comparing how it was and how it is now. Conduct an experiment: listen to the sound at a volume of, say, 15 on the display, and then 16, get out of the car for half a minute, and let your friend (or even a friend) cover the display with his palm (or palm), and you determine: it’s 15 or 16 ? If at the same time you pass by the cash register less often than five times out of ten (even on the same fragment), it means that your head unit has a volume step of 2 dB, this also occurs. Although there are, of course, talents.

3 dB is perceived as a noticeable change in volume. Not “big”, but simply noticeable. And here comes some bad news that you might have already guessed. The sound pressure created by the acoustics and the power supplied to the acoustics in order to create it live in the same column of our cheat sheet. Consequently, in order to obtain a noticeable change in volume, the supplied power must be doubled. This is what causes all the problems with power. Mainly because of this...