|
Posted by atilsley169 on
08-14-2012
|
|
fiogf49gjkf0d
Audio Sonar Acoustic Levelling (ASAL)
As an engineer, I’ve been intrigued by the various room treatment systems out there…most of which are sound (excuse the pun). My own 4-way front-loaded playback system is very dynamic…and suitable room treatment is required to bring out the best in the system.
Of late, I’ve been collaborating with an acoustics engineer to merge two applications to obtain the best possible acoustic outcome for my listening environment. This is radical…and as far as I can see, a world first.
The standard application for assessing room acoustics is to run audio sweeps using a mix of software packages….some commercial, some free-ware. These applications will show peaks, troughs, nodes etc. All good stuff. The outcome of such assessment would be to tailor the audio output using either passive or active input. However, the primary function of this approach is to ‘tame’ the output…which may or may not be desirable ultimately.
This is where it gets interesting.
Running a parallel pseudo-military spec sonar signal will provide an intimate acoustic image of the listening area, down to the nearest out-of-place book, cushion or chair. Sonar sweeps use microscopic detailed analysis to map the very imprint of the room. The outcome of this process, is that minor object relocation (or MOR) can dramatically level out the audio signal, thus reducing or minimising the need to alter the output at the source.
I have been dabbling with this now for 6 months…and I have found the acoustic improvements staggering, compared with the standard room acoustic treatments. There are possible patent and IP issues to consider here, so there’s only so much I can say in relation to the various algorithms and devices used…but I’m wondering if other enthusiasts have considered alternate room treatment solutions…?
Andrew
|
|
|
|
Posted by Romy the Cat on
08-14-2012
|
|
fiogf49gjkf0d I am sorry; I do not understand what you are proposing. Sonar might read for you the reflective quality of room but how does it affect the output of your playback? I mean: how you use the data the you got from sonar reading to navigate your decisions?
|
|
|
|
Posted by Paul S on
08-14-2012
|
|
fiogf49gjkf0d Andrew, the "holographic imaging" reminds me of certain FFT "scans". You use the term, "SONAR". Surely, test tones are actually in the "working" audio band? What frequencies are involved? Can you say a bit more about generation and "reception" without committing a crime?
I was impressed when a friend recently demonstrated for me an "app" for his i-Phone that takes an audio "snapshot" of a test tone, even a sharp whistle, using the mic built in to the phone. He uses the app at work to quickly diagnose ambiant acoustics, to speed set-up of giant sound reinforcement installations, but of course it lacks the "acute self-awareness" of the system you debut here. How cool would it be to have a 3-D "portrait" of not only the speaker output, but also the room (and contents) in which the speakers are "embedded"! And how cool to turn an arm chair 5 degrees south in lieu of Tube Traps!
Best regards,
Paul S
|
|
|
|
Posted by atilsley169 on
08-14-2012
|
|
fiogf49gjkf0d
Thanks Romy, Paul
Paul, the i-phone app sounds interesting…sort of capturing the concept. With the audio map established, only minor room tweaks are required to assist overall room improvement (leading to Optimum Object Placement, or ‘OOP’). For example, we can calibrate the sonar testing range from a relatively deep 50Hz up to 800Hz. The transducer must be set at the initial frequency for each test. When testing low frequencies, simply moving an arm chair will result in a different audio plot…while higher frequencies can be better levelled using angled books, cushions, etc. In fact, I have used small round felt pads inserted into the mid-range horn (550Hz to 4,000Hz) to level out a problematic 720-750 frequency null. I then re-run the audio sweep to re-assess levels.
|
|
|
|
Posted by atilsley169 on
08-16-2012
|
|
fiogf49gjkf0d So, now I am overlaying the sonar map on top of the audio output sweeps, to create a hybrid audio map. This then means we create an algorithm to input to a suitable audio reader. I then place this reader between the source (eg cd, dac) and the amp. Leveling then is assured.
|
|
|
|
Posted by Paul S on
08-16-2012
|
|
fiogf49gjkf0d Andrew, this seems like a classic case where the "inversion" processor would be - essentially- DEQ/DSP. I realize that most people do not worry about DSP, but in my own experience to date it simply kills the sound, no matter the other gains from a flatter response, even in cases where the source is CD. It would be wonderful to finally discover a way to DSP that I can accept in terms of end-use Sound, as I would then use DEQ in a heartbeat. How do you deal with this, or do you simply hear/feel no reason to concern yourself with the matter?
Best regards,
Paul
|
|
|
|
Posted by atilsley169 on
08-16-2012
|
|
fiogf49gjkf0d DSP is only partially involved. The real key is the audio 'reporting' from the sonar plots. You can pick and choose your fixes...from limited/aggressive DSP, to limited/aggressive object room placement (ORP). My ocus is on the latter as new way of radically improving in-room audio quality.
|
|