Difference between revisions of "Why Is There No Impulse Right Speaker With High Efficiency?"
(Die Seite wurde neu angelegt: „Ein nominal hoher Wirkungsgrad bei Kleinsignalanregung besitzt keine Aussage. Oftmals wird ein besonders hoher Wirkungsgrad bei einer besonders günstigen Frequen…“) |
(Artikel als "Löschkandidat" gekennzeichnet: Keine Beschreibung eines spezifischen Gerätes, Eher Werbung) |
||
| (23 intermediate revisions by 12 users not shown) | |||
| Line 1: | Line 1: | ||
| − | + | {{delete candidate}} | |
| + | {| class="wikitable" border="1" | ||
| + | |- | ||
| + | | | ||
| + | === What does the efficiency rate say? === | ||
| + | A nominal high efficiency at small signal excitation has no statement. Often a particularly high efficiency is indicated at a particularly favorable frequency. However, the characteristic sound pressure defines the average sound pressure in the midrange. A loudspeaker with a high characteristic sound pressure can therefore have a low ("characteristic") sound pressure in the bass range or in the high frequency range. The specification is therefore not valid for the entire bandwidth, especially not for the low frequency range, is misleading and does not allow a comparison of loudspeakers without taking other parameters into account. Drivers with high efficiency often have this, for example, by using a short voice coil and light paper cones. The maximum achievable sound pressure suffers from this. In addition, the linearity of the magnetic field (B-field) plays a decisive role, as does the linearity of the moving parts. This leads to non-linear excursion with strong distortions and, due to the diaphragm, to an early uneven drop of the dynamic phase.<br /> | ||
| − | + | Time-directed loudspeakers are louder than other comparable loudspeakers because of the high impulse amplitudes. Music is a dynamic event. The transients (transient processes) are of particular importance for the perception of loudness. The nerve activity of the auditory sense is strongly increased in relation to the transients and the auditory sense has its highest sensitivity. Loudspeaker measurements should take this into account, but are usually carried out in the steady state. | |
| + | This justifies why two different loudspeaker models with, for example, the same characteristic sound pressure, excited with music, can certainly reproduce differently loudly. '''A time-corrected summation sounds louder than a non-time-corrected one with otherwise the same efficiency in steady state due to the higher impulse amplitudes.''' | ||
| − | + | The step response of both models provides an indication of this. | |
| − | + | If the design is not correct in time, energy is wasted in the impulse peaks, the transients. The model with the correct step response provides more energy per time, sums correctly, produces significantly higher amplitudes in the transient, and is therefore louder and more efficient. | |
| − | + | Models with incorrect step responses stretch the energy content over time and in the process no longer create the possible maximum amplitudes in the transient. The "impulse dynamic efficiency" decreases with it. But the impulses / transients are the loudest events in music! | |
| − | + | The indication of the sound pressure level does not give any reliable information about the efficiency of a loudspeaker. Only in connection with dynamic measurements one comes closer to the truth. <br /> | |
| − | + | | | |
| + | [[File:2341319742 g.jpg]]<br /> | ||
| + | ''[[Myro Amur C]]''' | ||
| + | |} | ||
| + | {| class="wikitable" border="1" | ||
| + | |- | ||
| + | | | ||
| + | === The tweeter === | ||
| − | + | In the high frequency range (cut-off frequency 20 kHz) one will hardly get above 90 dB/Wm without a parasitic membrane resonance. However, a resonance is not a (sensible) usable sound pressure, it is the worst possible area (slow decay, phase rotation, high distortion). For example, if the diaphragm resonance is at 20 kHz and is not very damped, the tweeter will hum strongly at the top and might achieve 95 dB/Wm at the resonance peak. But which audiophile listener wants to listen to such a design? Should this be the end of the line at 15 kHz? Furthermore, the frequency range covered by resonance is not available for impulse reproduction.<br /> | |
| − | + | Sometimes the use of horn loudspeakers - also in the midrange - can be a remedy. However, this leads to strong phase rotations at the transmission ends. | |
| + | | | ||
| + | === Midrange and upper bass === | ||
| + | |||
| + | If you design a speaker to be a bandpass with less bandwidth, you sort of increase the Q and get a midrange hump. That is, the sound pressure level in the mids goes up, and at the transmission ends it goes down. | ||
| + | So you can get well over 90 dB/1Wm in the midrange. In principle, this applies to every single driver. By using a multi-way design with narrow-band individual drivers, it is possible to increase the sound pressure level in the transmission ranges covered by the drivers. However, the problem of the limiting ranges remains. | ||
| + | |} | ||
| + | {| class="wikitable" border="1" | ||
| + | |- | ||
| + | | | ||
| + | [[File:P 100106.jpg]]<br /> | ||
| + | ''[[Myro La Musica 2005]]'' | ||
| + | | | ||
| + | === The low frequency === | ||
| + | Often, high efficiency concepts forgo bandwidth. This results in early, steep phase shifts. | ||
| + | If they do not forgo bandwidth and are to achieve more than the low end of a small bookshelf speaker, then they necessarily need a large-dimension bass, because the radiated sound power is related to the radiation impedance and this depends on the diaphragm area. (The radiation impedance is also to be seen in the step response among many other properties). Acoustic power also depends on diaphragm velocity, so it depends on frequency and on the quality of resonance of the oscillating system. However, the efficiency does not correlate exclusively with the diaphragm size, because a prerequisite for a high efficiency is a large diaphragm area ''and'' low moving mass if the chassis parameters are sensibly designed. A strong drive essentially increases the efficiency in the midrange. | ||
| + | A large diaphragm and low moving mass results in a large equivalent volume (VAS) and thus large enclosure volumes. However, large diaphragms with low mass tend to be unstable and exhibit transmission characteristics that make it difficult or impossible to achieve the other goals set, such as designing the [[filters]] to be in the correct impulse response. In addition, low bass at well above 90 dB/Wm requires a resonant system (bass reflex or otherwise). If you make the diaphragm light (thin and e.g. made of cardboard), it will not resonate effectively with its whole size, but will break up into partial oscillations. It is practically impossible to make a light membrane stable. The result is completely uncontrolled diaphragm oscillations, whose interferences generate sound waves in phase opposition and interact with other drivers, even to the point of cancelling them out. This also contradicts high efficiency. It is not the objective to produce an arbitrary sound mixture, consisting of resonances, in order to generate as much level as possible. This also contradicts the demand for precise reproduction. <br /> | ||
| + | A smaller but stable bass can then reproduce a bass louder under certain circumstances. To try to achieve low bass and efficiency together is even more difficult. That's why speakers with "high efficiency" are tuned high in the bass (bass reflex / high pass filter). But a relatively high tuned bass reflex has an unfavourable group delay and a correspondingly strong phase rotation in the high pass. This means: the signals are shifted more on the time plane. It's helpful to know that the sound portion below 50 Hz in music is at vanishingly small percentages. Therefore, if a loudspeaker converts e.g. 95% of what happens in a first-class way, then a lot has already been achieved. | ||
| + | |} | ||
| + | |||
| + | Advantage: Large diaphragm areas are more favourable in terms of radiation impedance. This results in a better bass attack (better transient response). The first half wave of a large diaphragm is usually much louder in relation to the following half waves and corresponds more to the excitation frequency than with small diaphragms. Consequently, the graph of the step response is flatter from the starting edge! The speed is due to the more favourable radiation impedance, but not to the high efficiency. A high efficiency in the bass range with high quality at the same time can only be achieved with a great effort for physical reasons. To achieve this you need a large effective diaphragm area and a correspondingly large low-resonance volume with stable walls (we're talking washing machine size here) or a gigantic bass horn. But not every big bass has a large effective diaphragm area! | ||
| + | |||
| + | So with "high efficiency loudspeakers" one is forced to accept innumerable typical faults. High power handling is also made possible by filters with steep flanks. These, however, destroy the high impulse amplitudes and distort the signal structures. | ||
| + | All in all, this results in the typical PA sound that is associated with "live sound" by people who are used to it. A PA sound without these errors practically does not exist - because then it would no longer be a PA sound! | ||
| + | |||
| + | Simplified, one can draw the following conclusion (assuming a linear frequency response): | ||
| + | *A wide transmission bandwidth, i.e. high efficiency in the low bass and upper high frequency range, requires an adjusted, relatively low level in the midrange. | ||
| + | A small transmission bandwidth allows a higher characteristic sound pressure in the midrange. | ||
Latest revision as of 13:17, 31 October 2020
ContentsWhat does the efficiency rate say?[edit]A nominal high efficiency at small signal excitation has no statement. Often a particularly high efficiency is indicated at a particularly favorable frequency. However, the characteristic sound pressure defines the average sound pressure in the midrange. A loudspeaker with a high characteristic sound pressure can therefore have a low ("characteristic") sound pressure in the bass range or in the high frequency range. The specification is therefore not valid for the entire bandwidth, especially not for the low frequency range, is misleading and does not allow a comparison of loudspeakers without taking other parameters into account. Drivers with high efficiency often have this, for example, by using a short voice coil and light paper cones. The maximum achievable sound pressure suffers from this. In addition, the linearity of the magnetic field (B-field) plays a decisive role, as does the linearity of the moving parts. This leads to non-linear excursion with strong distortions and, due to the diaphragm, to an early uneven drop of the dynamic phase. Time-directed loudspeakers are louder than other comparable loudspeakers because of the high impulse amplitudes. Music is a dynamic event. The transients (transient processes) are of particular importance for the perception of loudness. The nerve activity of the auditory sense is strongly increased in relation to the transients and the auditory sense has its highest sensitivity. Loudspeaker measurements should take this into account, but are usually carried out in the steady state. This justifies why two different loudspeaker models with, for example, the same characteristic sound pressure, excited with music, can certainly reproduce differently loudly. A time-corrected summation sounds louder than a non-time-corrected one with otherwise the same efficiency in steady state due to the higher impulse amplitudes. The step response of both models provides an indication of this. If the design is not correct in time, energy is wasted in the impulse peaks, the transients. The model with the correct step response provides more energy per time, sums correctly, produces significantly higher amplitudes in the transient, and is therefore louder and more efficient. Models with incorrect step responses stretch the energy content over time and in the process no longer create the possible maximum amplitudes in the transient. The "impulse dynamic efficiency" decreases with it. But the impulses / transients are the loudest events in music! The indication of the sound pressure level does not give any reliable information about the efficiency of a loudspeaker. Only in connection with dynamic measurements one comes closer to the truth. |
The tweeter[edit]In the high frequency range (cut-off frequency 20 kHz) one will hardly get above 90 dB/Wm without a parasitic membrane resonance. However, a resonance is not a (sensible) usable sound pressure, it is the worst possible area (slow decay, phase rotation, high distortion). For example, if the diaphragm resonance is at 20 kHz and is not very damped, the tweeter will hum strongly at the top and might achieve 95 dB/Wm at the resonance peak. But which audiophile listener wants to listen to such a design? Should this be the end of the line at 15 kHz? Furthermore, the frequency range covered by resonance is not available for impulse reproduction. |
Midrange and upper bass[edit]If you design a speaker to be a bandpass with less bandwidth, you sort of increase the Q and get a midrange hump. That is, the sound pressure level in the mids goes up, and at the transmission ends it goes down. So you can get well over 90 dB/1Wm in the midrange. In principle, this applies to every single driver. By using a multi-way design with narrow-band individual drivers, it is possible to increase the sound pressure level in the transmission ranges covered by the drivers. However, the problem of the limiting ranges remains. |
The low frequency[edit]Often, high efficiency concepts forgo bandwidth. This results in early, steep phase shifts.
If they do not forgo bandwidth and are to achieve more than the low end of a small bookshelf speaker, then they necessarily need a large-dimension bass, because the radiated sound power is related to the radiation impedance and this depends on the diaphragm area. (The radiation impedance is also to be seen in the step response among many other properties). Acoustic power also depends on diaphragm velocity, so it depends on frequency and on the quality of resonance of the oscillating system. However, the efficiency does not correlate exclusively with the diaphragm size, because a prerequisite for a high efficiency is a large diaphragm area and low moving mass if the chassis parameters are sensibly designed. A strong drive essentially increases the efficiency in the midrange.
A large diaphragm and low moving mass results in a large equivalent volume (VAS) and thus large enclosure volumes. However, large diaphragms with low mass tend to be unstable and exhibit transmission characteristics that make it difficult or impossible to achieve the other goals set, such as designing the filters to be in the correct impulse response. In addition, low bass at well above 90 dB/Wm requires a resonant system (bass reflex or otherwise). If you make the diaphragm light (thin and e.g. made of cardboard), it will not resonate effectively with its whole size, but will break up into partial oscillations. It is practically impossible to make a light membrane stable. The result is completely uncontrolled diaphragm oscillations, whose interferences generate sound waves in phase opposition and interact with other drivers, even to the point of cancelling them out. This also contradicts high efficiency. It is not the objective to produce an arbitrary sound mixture, consisting of resonances, in order to generate as much level as possible. This also contradicts the demand for precise reproduction. |
Advantage: Large diaphragm areas are more favourable in terms of radiation impedance. This results in a better bass attack (better transient response). The first half wave of a large diaphragm is usually much louder in relation to the following half waves and corresponds more to the excitation frequency than with small diaphragms. Consequently, the graph of the step response is flatter from the starting edge! The speed is due to the more favourable radiation impedance, but not to the high efficiency. A high efficiency in the bass range with high quality at the same time can only be achieved with a great effort for physical reasons. To achieve this you need a large effective diaphragm area and a correspondingly large low-resonance volume with stable walls (we're talking washing machine size here) or a gigantic bass horn. But not every big bass has a large effective diaphragm area!
So with "high efficiency loudspeakers" one is forced to accept innumerable typical faults. High power handling is also made possible by filters with steep flanks. These, however, destroy the high impulse amplitudes and distort the signal structures. All in all, this results in the typical PA sound that is associated with "live sound" by people who are used to it. A PA sound without these errors practically does not exist - because then it would no longer be a PA sound!
Simplified, one can draw the following conclusion (assuming a linear frequency response):
- A wide transmission bandwidth, i.e. high efficiency in the low bass and upper high frequency range, requires an adjusted, relatively low level in the midrange.
A small transmission bandwidth allows a higher characteristic sound pressure in the midrange.
<zurück: Myroklopädie>
<zurück: Myro>