Posts Tagged ‘frequency’

A new record

Thursday, January 3rd, 2013

Putting a new spin on the long-playing vinyl record, an editor at Instructables has devised a new method for producing LP records using a rapid prototyping “3D printer”.  Working directly from a digital audio file, Amanda Ghassaei uses the waveform profile to create a 3D computer model of the familiar LP groove, which is then built up in physical form by a UV-cured resin printer.

Despite the cutting-edge 16-micron resolution of the printer, the end result is rather crude, with a frequency response and audio quality as yet far beneath a typical analog vinyl record.  The all-digital noise introduced by the discrete print (in time, aliasing, and in amplitude, quantization) is also harsh compared with the traditional “warm” analog distortion sought after by vinyl enthusiasts and audiophiles.  Even so, one could foresee a niche market for one-off, just-in-time pressing of records to keep alive long out-of-print material (or new material that might be in limited demand).  Even though this can be accomplished with .mp3 files or CD-R discs, sometimes there’s simply nothing like setting needle to vinyl!

[via Wired]

LinkedInShare

Drop-less droplets

Monday, September 24th, 2012

In a setup that’s equal parts science and Harry Potter, scientists at Argonne National Laboratory acoustically levitate liquids in midair to further critical pharmaceutical reasearch. Using a technology originally developed by NASA to simulate microgravity conditions, the pharmaceutical droplets are suspended in midair using standing waves of inaudible ultrasound generated by small speakers above and below.

By suspending a drug this way—free from any container or other physical contact—scientists can study its various forms and the ways it might be absorbed by the body. Not to mention putting on a pretty cool show in the process!

LinkedInShare

FFFFT (the fast fast fast Fourier transform)

Wednesday, January 25th, 2012

No, it’s not the sound of the air being let out of your tires!

Researchers at the Massachusetts Institute of Technology have just published a ground-breaking computational method for analyzing digital signals, including sounds and images.

The Fourier Transform is a way to break a complicated signal down into its most basic components, and it’s how computers manipulate things like acoustic and visual information—everything from your jpeg and mp3 files up to complicated acoustic measurement and analysis gear that consultants like ourselves use daily.

Fourier Transform

The last major improvement in the efficiency of the Fourier Transform came in the 1960s, with the advent of the “Fast” Fourier Transform (often denoted FFT).  That was a long time ago, but the FFT is still the method of choice for on-the-fly number crunching in everything from cellphones to video games to high-end audio and graphics workstations.

The new algorithm that MIT has devised (a “nearly optimal sparse Fourier transform”) is substantially faster than the FFT for a large range of realistic and useful cases—up to 10 times faster.  It isn’t hard to imagine that such a major leap in efficiency will lead to smaller, cheaper, and more powerful electronics, since the work they do under the hood just got a whole lot easier!

[via MIT News. Graphic: Christine Daniloff]

LinkedInShare

Catch a wave

Monday, May 9th, 2011

You may know that the sounds you hear travel through the air as waves, but the invisibility of air makes this concept a tricky one to visualize.  For those who like physics demonstrations (and who doesn’t), we recently came across this video of a series of pendulums—and the pendulum is perhaps the most accessible form of wave motion we witness in everyday life.

A pendulum’s length determines its frequency, just as sound waves in air have a frequency that corresponds to pitch.  The demonstration superimposes different frequencies to illustrate traveling waves, standing waves, beats, and “random” noise, which are all phenomena that come from mixing different frequencies together.

LinkedInShare

Visualizing modes

Friday, December 18th, 2009

The Graves on SOHO VoIP blog tipped us off to this cool video showing the vibration of a square plate at different frequencies. By covering the plate with salt, we can see the areas where the plate vibrates a lot (the salt rolls away) and the areas where it doesn’t vibrate at all (the salt collects). These spots and lines of little or no movement are called nodes. As the driving frequency (and sound) gets higher and higher, the patterns (called mode shapes) get more and more complex (and cool looking!)

This behavior is a great example of why we don’t use big speakers (woofers) to generate high-frequency (treble) sound. Instead of moving in unison like a piston, the speaker cone resonates internally at high frequencies, and different parts of the cone are moving in different ways, like a group of uncoordinated smaller speakers. This causes a poor frequency response, with lots of peaks and dips — often referred to as “breakup”. Smaller drivers (tweeters) are more at home with high-frequency sound, since their resonant “breakup” occurs at frequencies near or beyond the limit of hearing.

This is also a great way to visualize the analogous effect of room modes. Instead of a vibrating plate, a room is full of vibrating air, and at certain frequencies there will be points or areas in a room’s volume — nodes — with very little sound. This is most pronounced in hard, reverberant rooms, and at low frequencies. Still, even though the effect is more subtle in normally absorptive rooms, it can wreak havoc with the reproduction and recording of low-frequency sound, so identifying and managing room modes is a common task in the design of recording studios and listening rooms.

LinkedInShare