How to measure the resistance of the amplifier. The main technical characteristics of amplifiers

(About reducing intermodulation distortions and pride in loudspeakers)

The difference in the sound of loudspeakers when working with various UMPs, first of all, notice, comparing lamp and transistor amplifiers: The spectrum of their harmonic distortion is often significantly different. Sometimes noticeable differences are among the amplifiers of the same group. For example, in one of the evaluation audioshrums, the data of the lamp UMP with a capacity of 12 and 50 W was inclined in favor of less powerful. Or was the evaluation of a biased?

As it seems to us, the author of the article evidenced by one of the mystical causes of transitional and intermodulation distortion in loudspeakers creating a noticeable difference in sound when working with various UMPs. It also offers accessible methods for a significant reduction in the distortions of loudspeakers, which are simply implemented with the use of a modern element base.

Currently, it is considered generally recognized that one of the requirements for power amplifier is to ensure the invariance of its output voltage when changing the load resistance. In other words, the output resistance of the umzch must be small compared to the load, constituting at no more than 1/10.1 / 1000 from the resistance module (impedance) of the load | Z N |. This point of view is reflected in numerous standards and recommendations, as well as in the literature. Even this parameter is specifically entered as the damping coefficient - k d (or dumping factor) equal to the ratio of the nominal load resistance to the output resistance of the R output amplifier. Thus, at the nominal load resistance of 4 ohms, and the output resistance of the amplifier 0.05 K d will be 80. The current standards on the HIFI hardware require that the value of the damping coefficient high-quality amplifiers It would be no less than 20 (and recommended - at least 100). For most transistor amplifiers available, K d exceeds 200.
The arguments in favor of the small R exhibition (and, accordingly, high K d) are well known: this ensure the interchangeability of amplifiers and acoustic systems, obtaining an effective and predictable damping of the main (low-frequency) resonance of the loudspeaker, as well as the convenience of measuring and comparing the characteristics of amplifiers. However, despite the legitimacy and validity of the above considerations, the conclusion about the need for such a relationship, according to the author, fundamentally erroneous!

The thing is that this conclusion is made without taking into account the physics of the operation of electrodynamic heads of loudspeakers (GG). The overwhelming majority of the developers of amplifiers sincerely believes that everything that is required from them is to issue the voltage of the required value on a given load resistance with possibly smaller distortion. The developers of loudspeakers, in turn, seem to have to come from the fact that their products will be powered by amplifiers with a negligible low output resistance. It would seem that everything is simple and clear - what could be the questions here?

Nevertheless, questions, and very serious, there are. The main one is the question of magnitude intermodulation distortionintroduced by the GG when working from the amplifier with a negligible small internal resistance (voltage source or EMF source).

"What is the relationship to this can have an output resistance of the amplifier? Do not fool my head! " - Follow the reader. - and make mistakes. It has, and the most direct, despite the fact that the fact of this dependence is mentioned extremely rare. In any case, not detected modern workin which this influence would be considered on everything The parameters of the through electrical tract - from the voltage at the inlet of the amplifier to sound oscillations. When considering this topic, for some reason, it was limited to the analysis of the behavior of the GG near the main resonance at the lower frequencies, whereas no less interesting occurs on noticeably more high frequencies - A pair of octave above the resonant frequency.

To replenish this gap and this article is intended. It must be said that in order to increase the availability, the presentation is very simplified and schematized, therefore a number of "subtle" issues remained unreasonable. So, in order to understand how the output resistance of UMPs affects the intermodulation distortions in loudspeakers, it is necessary to recall what the physics of the radiation of the sound diffuser.

Below the main resonance frequency when the sinusoidal voltage of the signal is supplied to the winding of the sound coil, the amplitude of the displacement of its diffuser is determined by the elastic counteraction of the suspension (or compressible in the closed aircraft of air) and almost independent of the signal frequency. The work of the GG in this mode is characterized by large distortions and a very low impact of the useful acoustic signal (very low efficiency).

At the frequency of the main resonance, the mass of the diffuser, along with the oscillating weight of air and the elasticity of the suspension, form a vibrational system similar to the load on the spring. The radiation efficiency in this frequency range is close to the maximum for this GG.

Above the frequency of the main resonance of the inertia of the diffuser, along with the oscillating weight of the air, turn out to be large than the power of the suspension of the suspension, therefore the displacement of the diffuser is inversely proportional to the frequency square. However, the acceleration of the diffuser at the same time is theoretically independent of the frequency, which ensures the uniformity of the ACH on sound pressure. Therefore, to ensure the uniformity of the ACH GG at frequencies above the frequency of the main resonance to the diffuser on the side of the sound coil, it is necessary to apply the force of constant amplitude, as follows from the second law of Newton (F \u003d M * A).

The strength acting on the diffuser from the sound coil is proportional to the current in it. When connecting to the GG to the voltage source U current i in the sound coil at each frequency, it is determined from the OBA I (F) \u003d U / Z (F), where Z g (F) is the complex resistance of the sound coil. It is determined predominantly three values: active resistance of the sound coil R g (measured by an ometer), the inductance of L G. The current also affects the counter-EMF, which occurs when the sound coil is moved in the magnetic field and proportional to the movement of the movement.

At frequencies noticeably above the main resonance of the counter-emf, you can neglected because the diffuser with a sound coil simply do not have time to accelerate in half the signal frequency period. Therefore, the dependence Z g (f) above the frequency of the main resonance is determined mainly by the values \u200b\u200bof R g and L g

So, nor the resistance of R r nor the inductance of L g is not particularly constant. The resistance of the audio coil is highly dependent on the temperature (Copper TCS is about + 0.35% / o C), and the temperature of the sound coil of small-sized mid-frequency GGs during normal operation varies by a value of 30 ... 50 o C and and more and quite quickly - for tens of milliseconds and less. Accordingly, the resistance of the sound coil, and consequently, the current through it, and the sound pressure with a constant applied voltage varies by 10 ... 15%, creating intermodulation distortions of the corresponding value (in low-frequency yg, the thermal inertia of which is large, heating the sound coil causes effect thermal compression signal).

Changes in inductance are even more complex. Amplitude and phase The current through the sound coil at frequencies is noticeably above the resonant significantly is determined by the value of inductance. And it depends very much on the position of the sound coil in the gap: with the normal amplitude of the displacement for frequencies, only a few large, rather than the frequency of the main resonance, the inductance varies to 15 ... 40% of various GG. Accordingly, at rated power supplied to the loudspeaker, intermodulation distortions can reach 10 ... 25%.

The above is illustrated by the photo of the oscillograms of sound pressure, taken on one of the best domestic mid-frequency GG - 5GDSh-5-4. The structural diagram of the measuring installation is shown in the figure.

A pair of generators and two amplifiers are used as a source of a two-tone signal, between the outputs of which the test GH is connected, installed on an acoustic screen with an area of \u200b\u200babout 1 m 2. Two separate amplifiers with a large reserve in power (400 W) are used to avoid the formation of intermodulation distortion during the passage of a two-tone signal through an amplifying path. The sound pressure developed by the head was perceived by the tape electrodynamic microphone, nonlinear distortion which is the value of less -66db at the level of sound pressure of 130 dB. The sound pressure of such a loudspeaker in this experiment was approximately 96 dB, so that microphone distortions under these conditions could be neglected.

As can be seen on the oscillograms on the screen of the upper oscilloscope (the top - without filtering, the bottom - after filtering the PVCH), the signal modulation with a frequency of 4 kHz under the influence of another with a frequency of 300 Hz (with power on the head 2.5 W) exceeds 20%. This corresponds to the magnitude of the intermodulation distortion of about 15%. It seems that there is no need to remind that the threshold of the substitution of products of intermodulation distortion is much lower than one percent, reaching in a number of cases of hundredths of percent. It is clear that the distortion of UMPs, unless they have a "soft" character, and do not exceed several hundredths percent, are simply indistinguishable against the background of distortion in the loudspeaker caused by its work from the voltage source. Intermodulation distortion products destroy the transparency and the detail of the sound - it turns out "porridge" in which separate tools And only occasionally heard voices. This type of sound is certainly familiar to readers ( good test On distortion there can be a phonogram of children's choir).

Connoisseurs may argue that to reduce the impedance impedance impedance, there are many ways to fill the clearance of the cooling magnetic liquid, and the installation of copper caps on the magnetic system cores, and the careful selection of the core profile and the coil winding density density, as well as much more. However, all these methods, firstly, do not solve the problem in principle, and secondly, they are subject to complication and increase in the cost of GG production, as a result of which they do not find complete use even in studio loudspeakers. That is why most of the mid-frequency and low-frequency GGs do not have neither copper caps, no magnetic fluid (in such GG when working at full capacity, the liquid is often ejected from the gap).

Consequently, the supply of GG from a high-alone signal source (in the limit - from the current source) is a useful and appropriate way to reduce their intermodulation distortion, especially when building multi-band active acoustic systems. The damping of the main resonance has to be performed purely acoustic path, since its own acoustic quality of the mid-frequency GH, as a rule, significantly exceeds the unit, reaching 4 ... 8.

It is curious that it is this mode of "current" nutrition of the GG, it takes place in tubing umzch with a penter or thyroid exit at a shallow (less than 10 dB) of the OOS, especially if there is a local OC for the current in the form of resistance in the cathode chain.

In the process of establishing such an amplifier, its distortion without general OOS is usually founded within 2.5% and confidently noticeable by hearing when the control path is turned on (the comparison method with the "direct wire"). However, after connecting the amplifier to the loudspeaker, it is found that as the feedback depth increases, the sound is first improved, and then the loss of its detail and transparency occurs. This is especially clearly noticeable in a multi-band amplifier, the output cascades of which work directly to the corresponding heads of loudspeakers without any filters.

The reason for this, at first glance, the paradoxical phenomenon is that with increasing the depth of the OOS on the voltage, the output resistance of the amplifier decreases sharply. Negative effects of nutrition from umzch with a small output resistance are discussed above. In a triode amplifier, the output resistance is usually much smaller than in a penter or thicket, and the linearity before the introduction of the OOS is higher, therefore the introduction of the voltage of the voltage is improving the work of a separate amplifier, but at the same time the loudspeaker head deteriorates. As a result, as a result of the introduction of the OOS on the output voltage into the triode amplifier, the sound, indeed, can become worse, despite the improvement of the characteristics of the actual amplifier! This empirically established fact serves as an inexhaustible food for speculation on the topic of harm from the use of feedbacks in sound power amplifiers, as well as reasoning about the special, lamp transparency and naturalness of sound. However, from the above facts it obviously follows that the case is not in the presence (or absence) of the OOS itself, but in the resulting output impedance of the amplifier. That's where "Dog is buried"!

It is worth saying a few words about the use of the negative output resistance of UMPs. Yes, positive feedback (POS) on the current helps in charge of the GG at the frequency of the main resonance and reduce the power dissipated on the sound coil. However, for the simplicity and effectiveness of damping, it is necessary to pay an increase in the impact of the inductance of the GG on its characteristics, even compared to the mode of operation from the voltage source. This is caused by the fact that the time constant l g / rg is replaced by a large equal to L g /. Accordingly, the frequency is reduced, starting with which in the amount of impedances of the system "GG + Umzch" begins to dominate inductive resistance. The effect of thermal changes in the active resistance of the sound coil increases similarly: the sum of the varying resistance of the sound coil and the constant negative output resistance of the amplifier in percentage is stronger.

Of course, if R is. The mind in absolute value does not exceed 1/3 ... 1/5 from the active resistance of the winding of the sound coil, the loss of the introduction of the POS is small. Therefore, weak at the current for a small additional damping or for accurate warning adjustment in the low-frequency band can be used. In addition, the current and current source mode in UMPs are not compatible with each other, as a result of which the current power supply of the GG in the low-frequency band, unfortunately, is not always applicable.

With intermodulation distortions, we apparently dealt with. Now it remains to consider the second question - the magnitude and duration of the urtems arising in the Diffuser of the GG when reproducing the signals of a pulse character. This question is much more complicated and "thinner."

To eliminate these ghosts, there are theoretically two possibilities. The first is to move all the resonance frequencies outside the operating frequency range, to the area of \u200b\u200bthe far ultrasound (50 ... 100 kHz). This method is used in the development of low-power high-frequency GG and some measuring microphones. With regard to the GG is a method of "hard" diffuser.

So, the third option is also possible - the use of GG with a relatively "rigid" diffuser and the introduction of its acoustic damping. In this case, it is possible to some extent to combine the dignity of both approaches. Thus, the studio control loudspeakers are most often built (large monitors). Naturally, when nutrition, damped GG from a voltage source due to a sharp drop in the complete quality of the main resonance is significantly distorted by ACH. The current source in this case also turns out to be preferable, since it contributes to the alignment of ACH at the same time with the exception of the thermal compression effect.

Summarizing the foregoing, the following practical conclusions can be drawn:

1. The mode of operation of the loudspeaker head from the current source (as opposed to the voltage source) provides a significant reduction in the intermodulation distortion introduced by the head itself.

2. The most appropriate version of the loudspeaker design with low intermodulation distortions is the active multi-band, with a separating filter (crossover) and separate amplifiers for each strip. However, this conclusion is valid regardless of the power regime.

4. In order to obtain a high output resistance of the amplifier and the preservation of a small magnitude of its distortion, an OOS should be applied by voltage, but by current.

Of course, the author understands that the proposed method of reducing distortion is not a panacea. In addition, in the case of the use of a finished multi-band loudspeaker, the implementation of current nutrition of its individual GG without alteration is impossible. An attempt to connect a multi-band loudspeaker as a whole to an amplifier with an elevated impedance will lead not so much to a reduction in distortion, how much to the sharp distortion of the frequency response and, accordingly, a failure of a tonal balance. However reducing intermodulation distortions of the GG almost an order, and such an affordable method, clearly deserves decent attention.

S.Ageev, Moscow

2014-02-10t19: 57.

Audiophile "S Software

PROLOGUE: The output impedance for headphones is one of the most common reasons why the same headphones can sound differently depending on where they are included. This important parameter is rarely indicated by manufacturers, but at the same time it may cause significant differences in sound and significantly affect headphones compatibility.

Briefly: All you really need to know is that most headphones work best if the output impedance of the device is less than 1/8 of the headphone impedance. So, for example, for the 32-ohm grados, the output impedance must be a maximum of 32/8 \u003d 4 Ohm. Etymotic HF5 is a 16-ohm, because the maximum output impedance should be 16/8 \u003d 2 ohms. If you want life confident that the source will work with any headphones, make sure that its output impedance is less than 2 ohms.

Why is the output impedance so important? At a minimum of three reasons:

The larger the output impedance, the greater the voltage drop with smaller load impedances. This drop can be large enough to prevent the "scold" low-level headphones to the desired level of volume. As an example, BEHRINGER UCA202 can be brought with an output impedance of 50 ohms. It greatly loses in quality when using 16 - 32-ohm headphones.
Headphone impedance depends on the frequency. If the output impedance is much larger than zero, it means that the stress falling on the headphones will also change with the frequency. The more output impedance, the greater the non-uniformity of the frequency response. Different headphones will interact in different ways (and usually unpredictable) with different sources. Sometimes these differences can be significant and quite tangible on the rumor.
As the output impedance increases, the damping coefficient decreases. The level of bass, which was calculated for headphones in design, with insufficient damping, it can significantly decrease. Low frequencies will be more burst and not so clear (smeared). The transition characteristic worsens, while the depth of bass (more recession on low frequencies). Some people like those who like "warm lamp sound", such an undeplexed bass may even come true. But in the absolute majority of cases, it gives less honest sound than when using a low-voltage source.

Rule of one eighth: To minimize each of the above effects, it is only necessary to provide an output impedance at least 8 times smaller than the impedance of headphones. Even easier: divide the impedance of headphones to 8 and get the maximum impedance of the amplifier, which allows you to avoid auditions.

Is there any standard for the output impedance? The only such standard I know is IEC 61938 (1996). It establishes the requirement for the output impedance of 120 ohms. There are several reasons why these requirements are outdated, and at all are not a good idea. In the STEREOPHILE article about the standard value of 120 ohms, literally as follows:

"Who would not write it, he clearly lives in the world of dreams"

Must agree. Perhaps the value of 120 ohms was still acceptable (and then, hardly) before the appearance of the iPod and before the portable devices were generally gained wide popularity, but no more. Today, most of the headphones are designed completely different.

Pseudo standards: Outputs for the headphones of most professional installations have a resistance of 20 - 50 ohms. I don't know any one that would correspond to 120 ohms, as in the IEC standard. For consumer-class equipment, the output impedance value is usually lies within 0 - 20 ohms. With the exception of some lamp and other esoteric developments, most of the audiophile high-end equipment has an impedance below 2 ohms.

Ipod influence: Since in 1996, a 120-ohm standard was published, from low-quality cassette players, through portable CD players, we finally switched to the cooking iPod hobby "Ami. Apple helped make high quality Portable, and now we have in circulation at least half a billion digital players, not counting phones. Almost all portable musical / media players work from single rechargeable lithium-ion batteries. These batteries produce voltage just over 3 volts, which usually gives about 1 volts (RMS) at the output under the headphones (sometimes less). If you put a 120 ohm resistance to the outlet and use the usual portable headphones (the resistance of which lies in the limit 16 - 32 Ohms), the speed of playback is most likely insufficient. In addition, most of the battery energy will be dissipated in the form of heat on a 120-ohome resistor. Only a small part of the power will come to headphones. This is a serious problem for portable deviceswhere it is very important to extend the battery life. It would be more effective to serve all the power to headphones.

Headphone design: So for the same weekend impedance company-producers develop their headphones? As of 2009, more than 220 million iPods were sold. IPod and similar portable players in the headphone market are similar to 800-pound gorillas. Therefore, it is not surprising that most developers began to create headphones in such a way that they were well compatible with iPod. This It means that they are designed to work with the weekend impedance of less than 10 ohms. And almost all High end-sized headphones are designed for sources that comply with the rule 1/8, or having an impedance close to zero. I have never met audiophile headphones intended for home usedeveloped in accordance with the ancient 120-OMO standard.

The best headphones for the best sources: If you beat the most observed high-end amplifiers for headphones and DAC "Ami, you will find that almost all of them have a very low output impedance. Examples are Grace Designs products, Benchmark Media, Headamp, Headroom, Violectric, etc. The majority of high-end headphones are best manicing themselves in combination with the same equipment with equipment. Some of the most well-proven headphones initially have low impedance, including various models from Denon, Akg, Etymotic, Ultimate Ears, Westone, Hifiman and Audeze. All of them, as far as I know, have been designed for use in combination with a source with a low (ideally zero) impedance. Also, the Sennheiser representative told me that they develop their audiophile and portable headphones for sources with zero impedance.

ACH question: If the output impedance is more than 1/8 of the headphone impedance, the non-uniformity of the frequency response will be observed. For some headphones, especially reinforcement (balanced anchor) or multi-drive, these differences can be colossal. Here, as 43 ohms, the impedance is influenced by the ACH Ultimate Ears Superfi 5 - quite tangible unevenness of 12 dB:

Output impedance 10 ohms: Someone can take a look at an example above and think that such significant differences are manifested only with resistance in 43 ohms. But many sources have an impedance of about 10 ohms. Here are the same headphones with a 10-ohm source - still distinctly audible unevenness of 6 dB. Such a curve leads to the weakening of bass, a pronounced accent on average frequencies, muted high and fuzzy phase characteristic due to a sharp failure by 10 kHz, which may affect the stereo panorama.

Full-sized Sennheiser: Here are the full-sized Sennheiser HD590 with increased impedance, with the same 10-OMO source. Now the unevenness above 20 Hz is only a little more than 1 dB. Although 1 dB is not so much, unevenness is in the field of "buzzing" lower, where any accent is extremely undesirable:

How damping works: Any dynamic head, whether it's headphones or speakers, moves back and forth as music play. Thus, they create sound oscillations, representing a moving mass. The laws of physics say that a moving object is inclined to remain in motion (i.e. possesses inertia). Damping helps to avoid unwanted movements. If not going to the details too, the undepleted speaker continues to move when he has to stop. If the speaker is transmissible (it happens rarely), its ability to move the resulting signal is limited - imagine that the speaker is trying to work immersed in the maple syrup. There are two ways to damp the dynamics - mechanical and electric.

Jumping cars: Mechanical damping is like car shock absorbers. They bring resistance, because if you swing the car, it will not scold up-down long. But the depreciation also adds stiffness, because it does not allow the suspension to change its position in full compliance with the relief of the road. Therefore, it is necessary to search for a compromise: soft shock absorbers make a trip to the softer, but lead to swaying, tightly make a trip less comfortable, but prevent rocking. Mechanical damping is always a compromise.

Electrical perfect: there is the best way control the unwanted displacement of the diffuser, it is called electric damping. The coil and magnet in the dynamics interact with amplifier To control the displacement of the diffuser. This type of damping has fewer side effects and allows developers to create headphones with a smaller level of distortion and better sound. As a car suspension, capable of more accurately adapt to the relief of the road, optimally damping headphones can more accurately play the audio signal. But, and this is a critical moment, electrical damping effectively only when the output impedance of the amplifier is much less impedance of headphones . If you turn on the 16-ohm headphones into the amplifier with an output impedance of 50 ohms, the electric damping will come to no. This means that the speaker will not stop at the moment when he must stop. It looks like a car with worn shock absorbers. Of course, if the 1/8 rule is respected, electric damping will be sufficient.

Acoustic suspension: In the 70s, the situation has changed, since transistor amplifiers became popular. In almost all transistor amplifiers, 1/8 rule is observed. In fact, the majority corresponds to the rule 1/50 - their output impedance is less than 0.16 ohms, which gives a damping coefficient 50. Thus, the speaker manufacturers have the opportunity to develop better speakers using the benefits of low output impedance. First of all, the first closed speakers were developed with acoustic suspension from Acoustic Research, Large Advents, etc. They had a deeper and accurate bass than those similar to the size of the precursors designed for lamp amplifiers. It was a great breakthrough in the field of Hi-Fi - thanks to new amplifiers, it was now possible to significantly rely on electric damping. And it is a pity that such many sources are lagging behind 40 years or more.

What is the output impedance at my device? Some developers make it clear that they seek to reduce the output impedance as much as much as possible (such as Benchmark), while others indicate its products to their actual value (for example, 50 ohms for Behringer UCA202). Most, unfortunately, leave this value to the mystery. Some equipment reviews (for example, in this blog) include the measurement of the output impedance, since it significantly depends on it, as the device will sound with those or other headphs.

Why is such a large number of sources have a high output impedance? The most common reasons are the following:

Protecting headphones - More powerful sources with low output impedance are often able to submit too much power to low-level headphones. In order to protect such headphones from damage, some developers increase the output impedance. Thus, this is a compromise that adapting the amplifier to the load, but at the price of deterioration of the parameters for most headphones. The best decision - The ability to select two levels of amplification. Low level allows you to set less output voltage For low impedance headphones. Also, the additives can be used in the current, so the source will automatically limit the current for low-level headphones, even if a too large level of amplification is installed.
To be different - Some developers specially overestimate the output impedance, arguing that it improves the sound of their device. Sometimes it is used as a way to make the sound of the product different from the sound of competing products. But in this case, each "separate sound" that you get completely depends on the headphones used. For some headphones, it is perceived as an improvement, with the others as much as a significant deterioration. It is most likely that the sound is largely distorted.
It is cheap - A higher output impedance is the easiest solution for cheap sources. This is a cheap way to achieve stability, the simplest protection against short circuit; It also allows you to use less high-quality operational amplifiers, which otherwise could not even be able to dig even 16 or 32-ohm headphones. By successively connecting to the output of some resistance, all these problems are solved by the price in some cent. But for this, the cheapest solution has to pay a significant deterioration in the quality of the sound on many models of headphones.

Exceptions from the rules: There are several headphones allegedly intended for use with high output impedance. Personally, I am wondering, myth is or reality, since I do not know a single specific example. However, it is possible. In this case, the use of these headphones with a low-voltage source can lead to the transmissible dynamics of bass and, as a result, to the response from the planned achkh developer. This may explain the individual cases of "synergies" when certain headphones are combined with a specific source. But this effect is perceived purely subjectively - for someone as expressiveness and the detail of the sound, for someone - as excessive stiffness. The only way To achieve adequate work - use the low-level source and follow the 1/8 rule.

How to checked: If you are interested, whether the sound quality does not suffer because of the output impedance of the source, I can suggest purchased for 19 $ FIO E5 amplifier. It is equipped with an exit with a practically zero impedance and it will be enough for most of the headphones with impedance

TOTAL: If only you are not absolutely confident that your headphones sound better with some kind of better output impedance, it is better to always use sources with impedance no more than 1/8 from the impedance of your headphones. Or even easier: with impedance no more than 2 ohms.

TECHNICAL PART

Impedance and resistance: These two term in some cases are interchangeable, but technically they have significant differences. Electrical resistance is indicated by the letter R. and has same value For all frequencies. Electric impedance - the value is more complicated, and its value is usually changing with the frequency. It is denoted by Bukkova Z.. As part of this article, the unit of measurement of both values \u200b\u200bis ohms.

Voltage and current: To understand what impedance is, and what is in question in this article, it is important to have at least a general idea of \u200b\u200btension and current. The voltage is similar to water pressure, while the current is an analogue of the flow of water (for example, liters per minute). If you let water from your garden hose, without attaching anything to its end, you will get big flow Water (current) and can quickly fill the bucket, but the pressure near the end of the hose will be almost equal to zero. If you use a small nozzle on the hose, the pressure (voltage) will be significantly large, and the flow of water will decrease (it will take more time to fill the same bucket). These two values \u200b\u200bare associated with inverse dependence. The relationship between the voltage, current and resistance (as well as the impedance, under this article) is determined by the law of Ohm. R can be replaced by Z.

Where did the rule 1/8 come from?: The minimum audible differences of the volume, which are perceived by a person - about 1 dB. The fall in -1 dB on the output impedance corresponds to the coefficient, 10 ^ (- 1/20) \u003d 0.89. Using the voltage divider formula, we obtain that when the output impedance is 1/8 of the load impedance, the coefficient is exactly 0.89, i.e. the voltage drop is -1 dB. Headphone impedance can change within the audio frequency band at 10 or more times. For Superfi 5, the impedance 21 Ohms is indicated, but in fact it changes from 10 to 90 ohms. Thus, the 1/8 rule gives us the value of the maximum output impedance 2.6 ohms. If you take the source voltage to 1 to:

Stress on headphones with impedance 21 ohms (nominal) \u003d 21 / (21 + 2.6) \u003d 0.89 in
Headphones voltage with impedance 10 ohms (minimum) \u003d 10 / (10 + 2.6) \u003d 0.79 in
Stress on headphones with an impedance of 90 ohms (maximum) \u003d 90 / (90 + 2.6) \u003d 0.97 in
Non-uniformity frequency response \u003d 20 * log (0.97 / 0.89) \u003d 0.75 dB (less than 1 dB)

Measuring output impedance: As can be seen concept Above, the output impedance forms a voltage divider. Having measured the output voltage without connecting the load and with a known load, you can calculate the output impedance. It can be easily done using an online calculator. The voltage without load is "INPUT VOLTAGE", R2 is a known load resistance (do not use in this case the headphones), "Output Voltage" - voltage when the load is connected. Press Compute, and get the desired output impedance R1. It can also be done using the 60-hertz sinusoids (it can be generated, for example, in Audacity), a digital multimeter and 15 - 33-ohm resistor. Most digital multimeters have a good accuracy near the frequency of 60 Hz. Play 60 Hz sinusoid and adjust the volume in such a way that the output voltage is equal to about 0.5 V. Then plug the resistor and fix the new voltage value. For example, if you received 0.5 V without load and 0.38 V with a load of 33 ohms, the output impedance is approximately 10 ohms. The formula here is as follows: zist \u003d (RN * (Vxx - VN)) / VN. Vxx - voltage without load (idle).

Neither the headphones do not have completely resistive resistance that do not vary within the range of sound frequencies. The absolute majority of headphones are reactive resistance and have a comprehensive impedance. Due to the capacitive and inductive components of the headphones impedance, its value varies with frequency. For example, the dependence of the impedance (yellow) and phases (white) from frequency for Super Fi 5. Below is ~ 200 Hz Impedance is only 21 ohms. Above 200 Hz, it increases to ~ 90 ohms to 1200 Hz, and then falls to 10 ohms to 10 kHz:

Full-sized headphones: Perhaps someone is not interested in intra-channel headphones like Super FI 5, so here is the impedance and phase for the popular Sennheiser HD590 model. Impedance still varies: from 95 to 200 ohms - almost twice:

Matchast: One of the graphs at the beginning of the article demonstrated the non-uniformity of the frequency response of about 12 dB for Superfi 5 connected to the source with an impedance of 43 ohms. If we accept the nominal value of 21 ohms per reference, and the output voltage of the source will take 1 V, the voltage level on the headphones will be as follows:

Support level: 21 / (43 + 21) \u003d 0.33 V - What corresponds to 0 dB
With a minimum impedance of 9 ohms: 9 / (9 + 43) \u003d 0.17 V \u003d -5.6 dB
With the maximum impedance of 90 ohms: 90 / (90 + 43) \u003d 0.68 V \u003d +6.2 dB
Change range \u003d 6.2 + 5.6 \u003d 11.8 dB

Damping levels: Damping damping, as explained earlier, can be either pure mechanical (QMS), or to develop from electric (qes) and mechanical damping. Total damping is denoted by QTS. How these parameters interact at low frequencies - is explained by modeling of the tille - ground. Damping levels can be divided into three categories:

Critical damping (Qts \u003d 0.7) - many consider it an ideal case, as it provides the deepest LC, without any deviations of ACH or excessive stall (uncontrolled displacements of the diffuser). The bass of this dynamics is usually perceived as "elastic", "clear" and "transparent". Most believe that QTS 0.7 provides an ideal transitional characteristic.
Excessive damping (QTS
Weak damping (QTS\u003e 0.7) - allows you to get some gain of the LC with the peak at the top of the RAM of the range. The speaker is not fully controlled, which leads to an excessive "ringing" (i.e., the diffuser does not quickly stop its movement after the electrical signal attenuation). Weak damping leads to responses of Ahh, less deep bass, bad transitional characteristic and an increase in frequency response in the area of \u200b\u200bthe top border of the LF. Weak damping is a cheap way to raise the level of bass costs of their quality. This technique is actively used in cheap headphones, in order to create "fake bass". The sound of non-empty speakers is often characterized as a "spicy" or "negligent" bass. If your headphones are designed for electric damping, and you will use them with a source that has an impedance of more than 1/8 of the headphone impedance, you will receive precisely such, undepleted LF .

Types of damping: There are three ways to dampen speakers / resonance control:

Electric damping - Already known to us qes, it is similar to recuperative braking in hybrid electric vehicles. When you click on the brakes, the electric motor slows down the movement of the machine, turning into a generator and transmitting energy back to batteries. The speaker is able to perform the same. But if the output impedance of the amplifier increases, the braking effect is significantly reduced - hence the rule 1/8.
Mechanical damping - known as QMS, it is more likely like car shock absorbers. As you increase the mechanical dynamics damping, it limits the music signal management, which leads to greater nonlinearity. It increases distortion and reduces sound quality.
Damping - The case can provide damping, but it is necessary that it is closed - either with a properly configured phase inverter or with a controlled limit. Many top headphones are certainly open, which eliminates the possibility of using damping due to the hull, as in acoustic speakers.

Clicking level: For headphones who have a sufficiently dense landing, such as full-size embracing with tightly adjacent incosses, the developer can take into account the possibility of some additional damping due to the ear shell. But the shape of the head, ears, hairstyle, landing of headphones, the presence of glasses and other factors make this effect almost unpredictable. For overhead headphones, this feature is absent at all. Below you see two graphics depicting the impedance Sennheiser HD650. Please note: resonant peak on LF in open video It has a level of 530 ohms, but when using an artificial head, the value is reduced to 500 ohms. The reason for this is damping due to the closed space formed by the ear shell and the ambush.

Conclusion: I hope now it is clear that the only way to achieve efficient work Bundles Headphones-Amplifier is compliance with the rule 1/8. At least someone prefers sound at a higher output impedance, it is extremely depends on the headphone model used, the value of the output impedance and personal preferences. Ideally - should be created new standardIn accordance with which the developers would have to produce sources with an output impedance of less than 2 ohms.

Information from sponsor

Kupi.tut.BY: Convenient catalog of laptops, laptops prices. Here you can pick up and buy a laptop at a low price. Ease of payment, delivery, quality assurance.

Original Article in English: Headphone & Amp Impedance

Why is it important to the value of the output impedance of the source (amplifier), as it interacts with the headphones and what affects.

The output resistance can be determined in two ways.

1) Disable load resistance. Click the active input source. Test to the output clamping amplifier AC voltage. Calculate alternating current consumed from the source. Determine the output resistance of the amplifier. The replacement scheme of the amplifier implementing this method is shown in Fig.2.11.

Figure 2.11 - Scheme of replacing the amplifier, for calculating R out

2) Determination of the output resistance by load characteristic.

The output chain of the amplifier can be represented as follows with the following model in which the transistor output circuit is represented by the EDC source (Fig. 2.12).

Figure 2.12 - Scheme of replacing the output chain of the amplifier

The load characteristics of the amplifier is determined by the dependence of the voltage on the load from the load current, will be viewed in Fig. 2.13.

Figure 2.13 - Load Characteristics Amplifier

For the output chain of the amplifier in idle modes ( R N.\u003d ¥) and short circuit ( R N.\u003d 0) Determine the values U NHX and I kz:

From the load characteristic it follows that the output resistance of the amplifier:

Provided that, you can write :.

Consequently, the results of determining the output resistance obtained by the first and second methods are the same.

Since the input and output resistance of the circuit with OE is commensurate, then the consistent inclusion of cascades of amplifiers with OE is possible when they are satisfactory agreement. For example, for a two-stage amplifier with reinforcement coefficients to 1 and to 2 and the equality R exhaustion \u003d R in Vx2, we obtain the overall amplification gain.

Conclusions:

The voltage amplifier circuit (OE) has an approximately equal input and output resistance, which allows you to coordinate the input resistance of the subsequent cascade with the output resistance of the previous one when they are sequentially turned on in multi-stage amplifiers. The scheme with one does not allow such an inclusion, since. For consistent inclusion of cascades with about them, it is necessary to include matching cascades that are built according to the scheme with OK (see section.2.3).

Enhancement coefficients with OE and voltage K U.\u003e\u003e 1 (tens) and differ only by phase ratios j OE\u003d 180 °, j Ob.=0°.

Current gain for the circuit with OE ( K I.\u003e\u003e 1), and for the scheme with about ( K I.<1). Поскольку коэффициент усиления по мощности K P.=K u × k iThe scheme with OE has the highest coefficient.

The voltage amplifier circuit with OE is widely used in electronics, but the scheme with OB, despite the number of these disadvantages, is used in accordance with its advantages. These include the highest temperature stability and smaller nonlinear distortion (see Section 5).

8 RC-amplifier frequency characteristics
Sound frequency