Mix Interview: Dr. Peter D'AntonioTHE SCIENCE OF MUSIC 6/01/2012 5:00 AM Eastern
Dr. Peter D’Antonio’s devotion to acoustics is driven by his love for music. As a bass player and singer (he still gigs today), he brings an understanding of how music should sound into his designs. As a scientist (with a Ph.D. from the Polytechnic Institute of Brooklyn), he brings an expanding search for knowledge to help create tools that make great sounding spaces. It was the combination of musician and scientist that, in 1974, led him to develop a widely used design for modern recording studios utilizing a temporal reflection free zone and reflection phase grating diffusors.
In 1983, D’Antonio established RPG Diffusor Systems, Inc., leading the way in the sound diffusion industry. Since then, the name RPG has become synonymous with acoustical research and innovation. He is the holder of numerous trademarks and patents for a wide range of number-theoretic, fractal and optimized diffusing and absorbing surfaces.
Dr. D’Antonio has led the industry to develop methods for measurement and documentation of acoustical treatments, believing that a design performance must be proven, and not simply theorized. He served as Chairman of the AES Subcommittee on Acoustics Working Group SC-04-02, which published AES-4id-2001 for diffusion coefficient standardization; is a member of the ISO/TC 43/SC 2/WG25 Working Group, which published ISO standard 17497-1:2004 for scattering coefficient standardization and ISO 17497-2 for diffusion coefficient standardization; and has served as adjunct professor of acoustics at the Cleveland Institute of Music, since 1991. He is a Fellow of the Acoustical Society of America and the Audio Engineering Society and a professional affiliate of the American Institute of Architects.
Acting on his belief to share knowledge, he has become a mentor for students, which he is now focusing on. He has lectured extensively, published numerous scientific articles in peer review technical journals and acoustical and architectural magazines and is the co-author of the reference book “Acoustic Absorbers and Diffusers: Theory, Design and Application.”
I understand that your initial inspiration for becoming involved in professional audio came, indirectly, from Manfred Schroeder…
It did. I had built a recording studio in my home in the early 1980s—I’m a bass player and singer. But the studio was just not really sounding the way I had hoped that it would. Being a physicist—at the time working for the Naval Research Laboratory—I did a literature search on recording studios, and there wasn’t a single article on the physics of it. I did come across some articles by SynAudCon about a live-end/dead-end situation. So I thought I would give that a try. It mentioned the use of an absorptive front and a diffusive rear, so then I started doing some research on diffusion and came across the article in 1980 in Physics Today by Manfred Schroeder.
As I started researching what these reflection-phase gratings [RPG] actually were, it turns out they were just two-dimensional periodic repeats of a series of divided wells. So understanding and being able to model what a reflection-phase grating was was second nature to me back then. And that’s why we were able to do so much modeling, and eventually designed a series of these surfaces. Then, just following my tendency as a scientist, I made a presentation at the Audio Engineering Society meeting. And it was actually in the session chaired by Manfred Schroeder, which was a bit intimidating. (Laughs.)
At that first AES meeting in New York I met Bob Todrank at an after-session party, and he was building the Oak Ridge Boys' studios. So we got to talking, and he was interested in what I was doing. So he took a chance, and we put the first surfaces into the Oak Ridge Boys studio in Hendersonville. It was a big success. The design that we created back then sort of became the de facto standard for studios, even to the present day where you absorb the early reflections and create a diffuse environment in the rear. Our markets back then were pretty much recording studios, period.
Then it kind of evolved to broadcast studios, then to home theaters, and eventually wound up in worship spaces and schools. At the moment, schools are our biggest market, because a school is basically a microcosm of all of our markets. We have products now for classrooms, auditoria, physical plants, swimming pools, auditoriums, atria, AV rooms—pretty much every room.
The surface treatment you’re talking about is the Quadratic Diffusor, correct?
Correct. And like any other mechanical system it had three issues that were not seriously problematic but that could be improved upon. One was that it had a limited frequency response, just like a single loudspeaker has a limited bandwidth. So to overcome that we developed this diffusing fractal (Diffractal), where we have a diffusor within a diffusor, sort of like a coaxial loudspeaker.
Then, because these are periodic devices the energy is focused in very specific directions. We wanted to eliminate that to actually provide uniform scattering. We solved that by using modulation. We have patented a modulated array now, like a Bessel function loudspeaker, where the array has the same performance as the individual speaker.
And finally, because the original number-theory sequences consisted of integers, a simple prime 7 sequence being 0142241, the wells were related by integer multiples. At some frequency, all of those waves are in phase—so at a given frequency you have a good diffusor, and then you have a reflector, and then a good diffusor and then a reflector. We overcame that by developing an optimization code based on boundary-element methods, where the well depths were not integerly related. So the ultimate diffusor at this point is what we call a Modulated Optimized Diffusor (Modffusor). The optimization gets rid of the integer well multiples, and the modulation gets rid of the periodicity.
And today you have moved beyond diffusion.
We quickly realized that to sustain a company in perpetuity, we couldn’t live on bread alone (laughs), on diffusors alone. So we started looking at creating other types of surfaces: absorbing surfaces, combined surfaces that absorb and diffuse. We realized that we needed to get into the general architectural marketplace, and that’s kind of where we are today.
But at the same time, we’ve grown our research department significantly. We have testing for absorption coefficients. In fact we have the only reverberation room in the world that has incorporated the new ISO 354 recommendations. We have the only Impedance Tube in the United States that goes from 63 Hz to 4,000 Hz, a special design. We can measure scattering coefficients, and we can measure diffusion coefficients. So this capability has formed a pretty strong relationship with the acoustical consulting community.
We brought in a product called TopAkustik, which is an absorptive wood system. And we’ve tremendously grown that product line. We introduced the first micro-perforated wood system with 30,000 holes per square foot. We call that the Topperfo-Micro, and it has just exploded in the marketplace. Architects for the first time don’t have to see the acoustical aspect of a product. And then we introduced microperf and microslit transparent absorbers. These Clearsorbers are very popular because you can see through them, you don’t need any Fiberglas in the cavity. We just introduced a completely recycled glass absorber called Quietstone Light. So the product line has really dramatically expanded.
You are well known in the community for your testing methods on your products. Can you explain the process?
There are basically four tests. The first two are random incidence rev room and normal incidence impedance tube absorption coefficient tests, with the most reliable one being the impedance tube. We actually measure the pressure at several different positions, and that is a standard. You see, the upper limit of an impedance tube is essentially given by the speed of sound divided by two times the width. Now there were two things that we found problematic with the circular commercial tubes. To get a full bandwidth you usually have to use several different diameters, and even that is a problem because it’s hard to cut a circular sample. So we made a square tube at 6.3 inches square, and we sum the output of four microphones placed at a quarter of the width and height.
The reason for that is when they’re at a quarter, the first order and the third-order modes have a different sine at each microphone and cancel and the second order mode is at a null. What that does is, in a 6.3-inch tube, which would normally have an upper limit of 1,000 Hz, we can now take it up to 4,000 Hz. So I don’t know of any other tube that goes from 63 Hz to 4,000 Hz. That’s what we did to expand the bandwidth for the normal-incidence absorption.
For the random-incidence absorption, we’ve been doing quite a lot of monitoring over the years of the data from different laboratories. And sad to say, the random-incidence absorption coefficient is a very poorly understood number. We’ve seen several round robins, the most recent one in Europe where you can have +/- 0.2 for the absorption coefficient. I’ve been collaborating with the international, the ISO 354 Committee, and I’ve been giving a lot of papers about this, some of which can be downloaded from the Technology section of our new Website.
Also, we introduced the idea of calibration. We calibrate everything we do in this industry, but we don’t calibrate the room that measures our absorption coefficient. So we’ve started by calibrating, and we do this with the impedance tube as well. We calibrate the room zero. So if we’re measuring a 4-inch piece of Fiberglas, we put in a 4-inch solid MDF panel, and we treat that as the room zero.
Then, when it came to the diffusion coefficient, which we started developing ages ago in the early ’80s when I started to figure out how to measure these diffusors, I borrowed a word from crystallography and developed the goniometer. We have a ground plane goniometer with 37 microphones. And I’ve published several papers on this that are also on our Website.
Then there was a need to have another kind of a parameter called the scattering coefficient for computer modeling programs. The diffusion coefficient basically says how uniform the scattering is, like a loudspeaker polar balloon. But in addition to the absorption coefficient, modeling programs need another parameter, the scattering coefficient, which says that when a ray hits the wall, a certain amount gets absorbed, but a certain amount will get scattered in non-specular directions as well as the specular direction. The scattering coefficient basically is a metric which says how much of the energy that’s scattered is directed away from the specular direction. So you follow your specular rays, but you also have to then monitor what happens to the non-specular rays.
Let’s move on to absorption. You have talked previously about your membrane absorbers, and I know that you make broadband absorbers as well as frequency-specific membranes. Do you have a preference as to what goes where in a room? What style goes where?
Membrane absorbers, as you’re well aware, have a limited bandwidth, so we offer them at third-octave center frequencies. Because we have the big two-foot-by-two-foot impedance tube that measures down to 20 Hz, we can verify that they’re actually absorbing where we say they’re absorbing. Too often, people build their own Helmholtz resonators, and I can’t imagine they absorb where they think they absorb. Those are frequency specific.
I tend to think of those as more Band-Aid-type solutions for specific frequency problems. For broadband low frequency we introduced the Modex Plate, a 50Hz to 250Hz absorber and the Modex Broadband, which absorbs from 40 Hz to 5k, both being only 4 inches thick. Their mechanism is pressure, so that’s why we tend to put these pressure devices in the corners, where the pressure is the highest; the closer you can get them to the trihedral corner, the better. We try to position them in corners from floor to ceiling. And an additional one on the ceiling if you can. Any ceiling-wall intersection is another good place for them. If it’s a porous material, then you obviously want to space it from the wall, or make it thicker. Everything we do, we try to get a broad bandwidth response. I’m not a proponent of anything that’s 1-inch thick (Laughs), whether it be Fiberglas or foam, because you are filtering the response of room reflections.
Do you have any specific advice for engineers or musicians in the home studio market? Basic things to help them improve their rooms?
I would say in these smaller studios, if I were to do anything, I would do it in limited coverage area, but do it as broadband as possible. Any 4-inch product—four inches or more—is the way to go. Our Harmonix product is a good start.
And then for absorption, the 4-inch BAD panel to me is a universal solution for these rooms. Put it in the mid-third of the room. And the curved BAD panel is even a better option. So if I were doing a small project studio, I’d put curved BAD panels on all the walls in the mid-third, covering as much area as you can afford. Then I would use our Modex Corner membrane absorbers and stack them floor to ceiling. To provide broadbandwidth control, use a combination tuned to different third octave center frequencies. If budget allows, use our Modex Plate or Broadband absorbers.
And then I like to always use a cloud, because if you have a cloud over the middle of the room, you can load it up on top with absorption to control the decay time if you have to. And you can fill it with some diffusing surfaces. It’s a good place to hide the wires and introduce lighting.
And the clouds you like are diffusive, so they’re hard clouds that you can pack soft stuff up above.
Yeah, they’re diffusive clouds. You can just hang an island of T bar, and fill it. Then another effective solution that can actually be built by anyone is a soffit between wall and ceiling, where you create an absorbing using 2- or 4-inch Fiberglas vertically and horizontally forming a 90-degree soffit with an air cavity. That’s a good way to control low end as well.
That’s all good advice. Is there anything I haven’t asked that you’d like to get in to this article?
Yeah, there’s just one more thing, as I’m stepping out of day-to-day operations. A few years ago, I formed a new company called the Chesapeake Acoustic Research Institute, and it has essentially three goals: education, experimentation and exploration. Educationally I have begun mentoring doctoral and masters students, as well as providing seminars at all of the schools that offer acoustical programs. Much of the research that I mentioned to you earlier is done under the auspices of CARI. RPG will go on as a manufacturing company. And that’s going pretty well, actually.
And I’m happy to say that my band is still playing and having fun. We have six people now—two guitarists, sax, keyboard, piano, and I do the singing and play bass. We’re having fun. We like to do parties. We like to do complex songs. We do Steely Dan, Toto, some of the harder ones. Got kind of tired playing the blues. (Laughs.)