Towards Realizing the TPU

[poynt99]

I'll post it right now as a matter of fact:

Spherics:


My understanding is Steven Mark was developing his spherical 3d sound system. In general we can place the location of sound because of frequency shifting of moving sound; the doppler effect, the phase shifting of sound because a wave from a point source on your left side, has less distance to travel to your left ear than to travel to your right ear, sound attenuation and frequency suppression (apparently the sounds from different directions hit the ear and ear canal in different ways that cause certain bands of frequencies to be subtly suppressed - I don't confess to understand how it does this!).

Obviously ideas on faking 3D sound placement require electronics to subtly alter phasing etc. High quality voice coils on speakers can be wound in a bifilar fashion and can be considered aircore type coils. He noticed an unusual frequency peak in a spectrum anlyser when pulsing the bifilar with two similar waveforms... like many inventions start with the phrase 'that's really strange why did it do that'. An unusual freq. peak ON TOP of what was expected indicates some sort of incoming energy. The rest is how to utilise the effect to extract the energy in such a way that you can use some of the extracted energy to make the event that caused the initial energy bigger. i.e. A positive feedback loop.

Got to go!

More golden information! S is accurate in what he says there about the 3-D methodology.

I was going to save this for the document, but since sg brought it up here in S's message, I'll capitulate a little.

In the early to mid 1990's I was fascinated with and in development of my own 3-D sound system of sorts. In the end time and money was lacking so the project was not completed. In the process however, much was learned about this subject.

True 3-D from two speakers is a two-step process: 1) Eliminate the inter-aural (ear-to-ear) cross-talk, and 2) process the audio in the recording stage to include "spatial" information.

I suspect Steven's goal with his speaker research was to accomplish step 1). It is quite possible that his "Audio-Spatial Effects" patent was an attempt at step 2).

Dramatic results in the audio listening experience can be achieved even by step 1) alone. In order to cancel, eliminate, or minimize inter-aural cross-talk, trans-aural (across the head) correction is applied. When sound from the left speaker reaches the right ear, this is unwanted crosstalk, and in order to cancel it we apply an inverted, filtered (HRTF), and time-delayed version of the left speaker audio through the right speaker, so that the two sounds effectively cancel when they reach the right ear canal.

There are a few ways of accomplishing this and here is one from Polk Audio:
http://www.polkaudio.com/downloads/whitepapers/SDA_WhitePaper.pdf

I suspect what Steven was working on was a dual voice-coil, single driver version. One voice coil is used for normal channel audio, and the other is for the trans-aural corrected audio from the opposite channel.

In my document I outline a circuit and experiment to find the kick and/or comp field as best I currently understand them. What I did not mention is the suggestion to pulse the second coil in the pair not only in a delayed fashion, but also inverted 180º (switch the leads), even though spherics stated not to. It can not hurt to try both ways imho, because Steven likely did try it.

The delay time setting is based on the time required for sound to travel across the human face to reach the opposite ear. This inter-aural delay time equates typically to about 400us based on a nominal size head.

.99

Further 3-D Sound Reading:
http://profs.sci.univr.it/~dafx/Final-Papers/pdf/Jost-Jot2.pdf
http://www.google.com/patents?id=mMwOAAAAEBAJ&dq=6643375
http://www.google.com/patents?id=UOcOAAAAEBAJ&dq=6577736
http://www.media.mit.edu/~billg/docs/TR-342.ps.gz
http://books.google.com/books?id=h90HIV0uwVsC&pg=PA165&lpg=PA165&dq=distance++ears+transaural&source=bl&ots=diw9TIv6gX&sig=SoRhCPxS64jyYGclv9mbAqWko9Y&hl=en&sa=X&oi=book_result&resnum=4&ct=result#PPA165,M1
http://homepage.mac.com/cooperbauck/cbpublications/documents/XTCwhitepaper.pdf
http://www.headwize.com/tech/sibbald_tech.htm

[forest]

Guys,

I'm lost.I can't follow that multi-pages extremely long threads about SM TPU, finally finding that no real good information is there. Why not put a sticky thread  to summarize all findings up today ? That way everybody could just realize if that topic is of some interest for them.

Reading all topics is mostly a waste of time.

My little small advice : drop all electronics, think about tips Steven Mark gave us. I see many fighting  with more and more complex oscillators trying to put 3 frequencies or harmonics into ring without results.

Ah...what a waste of time.... 

[BEP]

The delay time setting is based on the time required for sound to travel across the human face to reach the opposite ear. This inter-aural delay time equates typically to about 400us based on a nominal size head.

Or the circumference of the TPU?

Great info .99
P.S.

My wife just pointed out that the delay for my head will probably be around 800us 

[forest]

Don't waste your time please. If you had read Steven Mark tips carefully you would realize some similarities between TPU and cyclone.Check it.We are deceived about Coriolis force being not real obviously...

[Grumpy]

@Grumpy, Another quick question: what waveform type did you use to power the delayed bifilar in that experiment?

square wave