Today we tell about selecting of loudspeakers for home speaker systems.
Often people say that the last criterion of assessment of quality is human ear. We will consider whether so it.
If say exactly, quality of sound very accurately described by specification (numbers and diagrams).
Why then says: last criterion of assessment of quality is human ear?
Manufacturer don't measure and don't give all diagrams. Often it is technically impossible (too much measurements).
For most people (non - audio technics - specialist) this diagrams is hard to understand.
Besides for right interpretation we must integrate all measured data.
It is not way. What to us to do?
Why more large loudspeaker (diameter of head) is more preferably?
More diameter - more amply push the air in room.
Not any big head will well sound. Don't tempted on big and cheap acoustics.
Big concert acoustics not always is a sound quality standard. Concert acoustics for great sound pressure (on open air also).
Demands for harmonic distortions is more soft than for home acoustics.
If you have place on floor, using of the floor speakers is preferably.
"Have place" it is direct way (without barriers) for sound wave.
If don't have place on floor, use bookshelf speakers.
Somebody say:"I can enforce any speakers sound equally"
It is not absolutely correct statement. What do they mean?
Speaker and room is integrated system. You can't "listen speakers". You can listen system.
Into "system" we must include amplifier and (sometime) cables. But we rarely have all components simultaneously.
Quality of speaker depend by quality of loudspeaker's headers.
Its specification include apart from Amplitude-Frequency and Phase characteristics also Total Harmonic Distortion.
View now approximate reasons of forming of prices of speakers
It's not hard rule of course. For different countries the ranges of prices is different also. But we will see some trend.
For enforcing any speakers sound equally we can correct Amplitude-Frequency and Phase characteristics by equalizer.
But we cant exactly correct this characteristics.
We remember also: these characteristics is various for different points of listening room.
But we can't affect to Total Harmonic Distortions and speaker's push air capacity.
Than loudspeaker systems for $500 differ from speakers for $3000?
Fidelity of playing for low and very low levels of loudness.
We remember about design and expensive materials also.
Than loudspeaker systems for $100 differ from speakers for $500-$2000?
Fidelity of playing for all levels of loudness. More transparency.
For prices more $500-$1500 the increment of quality is less than into range < $500-$1000.
We often say about transparency of sound
What is in technical terms? Or it only subjective characteristic for our ears?
Certainly we can to measure the transparency.
Transparency is complex of characteristics, which define the capability of non-distorted throughput for device.
If to say simply: Transparency = minimum harmonic distortions + flat frequency characteristic (less +/- 1 dB in declared band).
Usually we apply term "transparency" about hi-frequencies (5-10 kHz). We qualify it as "air" or "airy sound". What for low frequencies?
The same. But here speak not about "air", and about clarity of low frequencies. Bass notes don't merged. No bass hum.
What is harmonic distortion?
We send sine signal to input of device. At output we have this sine + other sines with other frequencies (named as harmonics). Other sines is harmonic distortions.
Than reflections and re-reflections are harmful?
Imagine: we motionlessly stand on one place. The sound (air) wave press alternately to the ear-drum and in the opposite direction.
Reflections has time delay relative to speaker's direct wave.
Therefore we have next cases:
1. Speaker's direct and reflected waves push the ear-drum in same directions simultaneously. We have more loudness, than without reflections.
2. Speaker's direct and reflected waves push the ear-drum in the opposite directions simultaneously. We have less loudness, than without reflections.
These cases differ for various points of listening room and for different frequencies. But these cases is permanent for various times and for different levels of loudness.
What do for eliminating these problems?
We can eliminate the reflections. But we may get "dead sound".
For case of adding speakers direct and reflected waves, we can decrease loudness for these frequencies.
I recommend to use the parametric equalizer pronouncedly
It can sufficiently precisely compensate the superfluous loudness.
For case of inter-suppressing speakers direct and reflected waves, we can't do anything. Only suppress reflection or change listening point.
How suppress reflection? Use sound-absorbing and sound dispersive materials for wall's surface finishing, installing sound traps. You can preliminary calculate room's acoustics (without assistance or with specialist). Otherwise - only experimental tuning.
Whether it is possible to use an equalizer for increase of loudness? No.
We increase loudness of speaker's direct waves. And reflected waves loudness increased too.
Difference of direct and reflecting waves remain constant.
Now about "dead sound"
Reflections emulate concert room/hall. They give to us rear sounds (like multichannels systems).
Into sound-insulated and anechoic room we feel discomfort. We listen own bloodstream even. Such room does not suit for listening.
Other reasons give us frequency distortions also.
Mechanical resonances of room (at separate frequencies) give increasing of loudness. Compensation - with parametric equalizer or sound traps.
Mechanical dampings of room (at separate frequencies) give decreasing of loudness. Compensation - with parametric equalizer also.
Why recommended using parametric EQ but not multiband EQ?
Parametric equalizer allow better localization. It have minimal touching of "good" frequency bands.
For correction of frequency distortions need find frequencies of "resonances" and "dampings".
Finding by way measurement.
Simplest (most expensive) method - measurement with panoramic SPL-meter. We see real-time spectrum (loudness by frequency) for sound in listen point.
You can use usual SPL-meter:
manual measurements at separate frequencies.
You can use own ears. For in need sweep signal or sequence of frequencies.
Sweep signal isn't very good for detecting by ears of frequency value.
Sequence of frequencies - resonances may be between these frequencies and difficulties with memorization of loudness for ear.
"Audiophile Inventory" now make researching for simple method of speaker calibration without additional equipment. It will not the precise method. But it will allow to calibrate your system on one's own, and train hearing.
Also you can use hardware geneкator of sound frequencies or software generator at notebook.
It is possible to recommend following records for test bass resonances:
Cezaria Evora - album "Nha sentimento" - song "Vento de Sueste"
Goran Bregovitch - album "New Collection" - songs "100 Lat Modej Parze", "Gas Gas"
VA - album "Deep Bass. Low Frequency Sound Adventures" - song "Final Frontier"
What you should listen?
Bass notes must listened as separate notes with sufficient (non quiet, non excessive) loudness, not muttering, not buzz.
For control high frequencies you need use jazz and acoustic guitar records. Listen sounds of cymbals and string's scratches. These sounds should be clear, without "dirt". Recollect the live concert.
Yuri Korzunov, developer, Audiophile Inventory