Three Decisions (Part 2)

Around the same time I was setting up a studio of my own, I was also consulting for musician, client and friend Art Halperin, offering suggestions and design ideas as he rebuilt his own studio from the ground up.  Since I prefer to record on location in performance spaces, my own studio was to be used primarily for post-production work and mastering.  Art needed a space in which the musicians could gather to create the original recordings as well.  Both spaces provided opportunities to try out new ideas and to make some new discoveries.

While my own studio was built within an existing space, Art’s new studio, Top of the World, was custom built from scratch.  Everything from the dimensions of the space to the materials from which it was constructed started as an idea, discussed and rolled around until it was decided upon.

I was thrilled to finally have my own work space, with my own gear.  The system was sounding fantastic but there was still room for improvement.  Art had installed similar gear at Top of the World and there too, good as everything sounded, we felt it could be taken up a notch.  What both spaces needed was a full acoustic treatment.  Indeed any space used for serious listening needs this, whether pre-constructed or a newly built room.

Looking at all the available options and trying them, one type of design stood out above all the others, in performance and conveniently, in ease of use as well.  This was the cylindrical design based on a modification of Harry Olsen’s pioneering work in the 1950’s with his “functional sound absorbers”.  Rather than installing separate devices in the room to address each of the three types of room issues (resonant modes in the bass, early reflections in the treble and diffusion), a single type of device addressed all three issues simultaneously.  (For more on resonant modes and room treatments, see Setting up your monitoring environment.)  Since both new rooms needed the treatments, Art and I decided to build our own, so we ordered the raw materials and spent three days building enough cylinders to fully outfit both studios.

The cylinders were stacked in pairs around each room, making columns that evoked mental images of the Parthenon.  The larger diameter, 16 inch (~41 cm) columns were placed in the corners and at the half-way point along each wall, with the corners addressing the fundamental resonant modes for each dimension of the room and the half-way points addressing the first harmonic of each fundamental.  The smaller diameter, 9 inch (~23 cm) columns were placed at the quarter points along each wall to address the second harmonics.  In this way, the resonant modes of both rooms were quelled.

Each cylinder was built with one side that is soft and absorbent in the treble and the other side reflective in the treble.  When placed in the room, the cylinders were oriented with the absorbent side facing the nearest loudspeaker and the reflective semi-cylinder facing away from the speaker.  By doing so, the soft side serves to capture early reflections in the treble, helping to preserve the tonality and imaging from the speakers without interference from the room, while the reflective side provides the diffusion of late reflections—those sounds that have already been around the room—to help preserve the sense of life in the room.

All too often, we see photographs of studios that more closely resemble padded cells, with so-called “acoustic foam” placed everywhere.  Rather than absorb only the early reflections, these tend to soak up the life in the room, making the sound unnatural and simply sitting in the room uncomfortable.  We also see photographs of studios with diffusion placed very close to and facing the loudspeakers.  The result of this is that those early reflections in the treble that should be absorbed are instead splayed in all directions and are therefore guaranteed to reach the listener’s ears, hardening tonality, obscuring low level detail and causing a loss of focus in the images and soundstage presented by the monitors.

Upon installation of the cylinders, the wholesale transformation in the sound of both studios was immediate and obvious.  Where before the treatment, one could hear changes in the bass response as one moved around the room, as soon as the traps were in place, the bass not only remained consistent everywhere in the room but low level detail was much easier to hear, as were the spaces “between the notes”.  The room was no longer “ringing” and filling in the quieter moments between sounds in a recording (or live sounds in the room).  Even conversation became much easier to hear.  The space captured in recordings, whether real or synthesized in the studio, became much more evident.  In effect, the room had gotten out of the way.

I can remember one more thing about the day we finished building the traps for our rooms.  It took a few trips in my Honda to get all of the cylinders for my studio from Art’s patio where we built them to their new home.  I never heard the system in the car sound as good as it did on each of those rides.

There was still one more discovery to be made with regard to the systems in both studios.  It all started at a meeting of a local audio society where one member passed around a 1/2 inch (~13 mm) steel ball and a small “bowl” in which the ball was supposed to sit.  He claimed that using a trio of these devices underneath a component, to lift it from its own feet, made for very positive changes to the sound.  My initial response was doubtful.  How could putting a component on different “feet” change its sound?  Then again, I’d once asked “How can a turntable affect the sound of a record?” and on another occasion “How can a cable possibly make a difference in the sound?”  Both times I ended up learning something and coming to appreciate what I’d learned.  So I decided to embark on another set of experiments.

Over the course of the next month or two, I got hold of a very wide variety of items sold as replacement “feet” for audio and video components and proceeded to audition all of them.  I had sets of cones, spikes, miniature trampolines with elastic suspensions and many other designs.  Most of the devices were claimed by their manufacturers to provide isolation from vibrations that would degrade component performance.  Some claimed to block vibrations from entering the gear.  Others claimed to “drain” vibrations generated by the gear itself.

My expectation was that none of these would have any effect whatsoever.  Once again, it was my very good fortune to learn something new.  Actually, I learned a number of things, first among them was that anything placed under (or atop) an audio (or video) component will change its performance.  I emphasize “change” because the effects I heard from some types of devices were not necessarily positive.  I also learned that a number of the devices sold as isolators were not isolators at all and in fact acted as couplers, performing the exact opposite of an isolator.

While the couplers changed the sound, the changes were somewhat random from component to component.  The couplers changed the sound by altering the resonant characteristics of the component’s chassis.  Some of these, claimed by their manufacturers to “drain” vibrations from a component, were also referred to by an ingenious term which I would guess was invented by a savvy marketer.  They were called “mechanical diodes”, the claim being that vibrations would pass through them in one direction but would be blocked in the other direction.  I found it very easy to dismiss these claims by showing how a component placed atop these devices would move in direct response to any motion in the supporting surface or shelf.  In other words, the devices were perfect couplers and any path for vibrations is always a two-way street.

Further, with regard to the claim of “draining”, it occurred to me that when something is drained, by definition, I would expect there to be less of it in the place from which it has supposedly been drained.  When I drain the water from my kitchen sink, the result is less water in the sink.  If I run the water while I drain the sink and the quantity being drained matches the quantity being added, the net result is the amount of water does not change.  If the amount of vibration supposedly being drained does not exceed the amount being generated by the component, the net result is the same amount of vibration in the component.  Effectively, nothing has been drained except the customer’s wallet.

The surprise came when I got to the real isolators.  The well designed roller bearings, like the one I first saw at the audio group meeting and well designed air bearings defied my expectations and left my jaw hanging.  These acted as mechanical low-pass filters, devices with an inherent resonance above which they did not transmit vibrations.  (For more detail, see Vibration control for better performance.)  I was not at all prepared for what these did for every single component with which I tried them, which was to improve every characteristic of audio (and video) I know how to describe.  In addition, the improvements were consistent and repeatable from component to component and further, they were cumulative with system performance improving as each additional component was set “afloat”.  The improvements were such that I wondered why others were not shouting the news from the audio rooftops.

I considered how these devices worked and having learned that isolation begins at approximately 1.4 times the resonant frequency of the device, it became clear that the lower the resonant frequency of the device, the sooner it would begin to provide effective vibration isolation.  Further, there was an inverse relationship between how damped the device’s resonance was and how steep the rolloff above resonance was (i.e., how much isolation the device provided).  In other words, the less damping on the resonance, the steeper the rolloff (the greater the degree of isolation).  With this in mind, I thought I could improve on the commercial roller bearing designs I’d tried, so I made some drawings and took them to a local machinist.  (Actually, I spoke with a number of machinists, some of whom provided silly pricing quotes, including one who wanted an additional $75 “set up change”.  Not seeking to be “set up”, I thanked them and went elsewhere until I found the shop I chose to work with.)  I had some prototypes made and these proved so successful, I went back and had enough made to place under everything in the studio.  Partly because they reminded me of the “ball-and-socket” joint where a human femur meets the pelvis and partly in honor of how an old time jazz musician might refer to a place in which they liked to play, I christened my design “Hip Joints”.

While the improvements in digital devices were more pronounced than in say, solid state amplifiers (which nonetheless, still showed improvement), I was in for another shock when I decided to try them under the loudspeakers.  For years I’d been “taught” that speakers must be mounted rigidly and here I was placing them atop the loose springs that were a set of Hip Joints.  Once again, an opportunity for learning presented itself.  I’d never heard the speakers sound so good.  I referred to the speakers directly on the floor as “bound and gagged” by comparison.

While roller bearings provide isolation from vibrations in the horizontal (and rotational) planes, they do not isolate in the vertical plane.  Air bearings, on the other hand, provide isolation in the vertical plane.   Since I wanted to apply multiple-axis vibration isolation, I sought a way to combine Hip Joints with air bearings.  As a result, I made a new set of design drawings and with the aid of a good friend with the unique ability to turn wood into art, created the “Enjoyyourself” racks.  Unlike most racks, which provide clear paths for vibrations from the ground to get into every component they house, the Enjoyyourshelf racks have a separate, fully adjustable air bearing and a set of Hip Joints for each shelf.  (I call it the world’s first piece of furniture with a fully independent suspension!)  Upon first audition of gear on these racks, I heard the sound completely freed from the confines of the loudspeakers and the soundstage (on those recordings containing such a large space) expand well beyond the boundaries of the studio.  By design, the air bearing inflation is adjustable without having to first remove the gear, which allows for changes in inflation while the music is playing—a most informative situation.

As I said in the article cited above, “I’m still having a bit of trouble accepting that the ocean tide or the wind or a truck changing gears 1/4 mile away has such a profound effect on the performance of my audio and video gear.  What I have no trouble with is the results of isolating my gear from these effects.  The performance gains in every parameter I can think of are clear, consistent and repeatable.  Frequency extension into the treble and downward in the bass is improved.  Stereo imaging gets better focused.  The soundstage takes on greater proportions.  Dynamic swings both large and small are more like real life.  Overall, there is a much greater sense of the system getting out of the way, leaving the listener with a considerably increased sense of contact with the recorded event.”

With the installation of the acoustic treatments, followed by the addition of vibration isolation measures to all the components, the studio had “arrived”.