Friday, October 10, 2014

Rome Wasn't Built In A Day, And Neither Was My Studio

Back in 2011, I built myself a humble studio, and after having painstakingly labored to get the acoustics to my satisfaction (see here), I recently did something crazy: I changed everything.

This was not a case of me having a sickness.  Nor do I particularly enjoy torturing myself, as “tuning” a room can be very tedious and frustrating work.  No, this was about the never-ending quest for superior acoustics.  You see, I wanted to get new speakers and a 2nd subwoofer, but the problem was that I would not be able to integrate them into the existing setup.  So I had a choice to make: either maintain the status quo, or uproot my existing setup.  Well, to hell with the status quo!

Let’s proceed like a movie that shows you the ending first, and then jumps back to the beginning.  This is the new studio:

This is the old studio:

In comparing the before and after you will notice, A) the orientation within the room has changed, B) a window has magically appeared, and C) there’s less acoustic treatment in the room (about half as much).  The change in orientation was mainly a practical one - with larger speakers, and more of them, changing the orientation better accommodated all the pieces of the puzzle.  As for the window, it was always there but previously it was boarded up with a custom-made panel in order to create a continuous and uniform wall surface (and it was completely obscured by acoustic treatment).  In the old setup the room measured better with the panel in the window, but in the new orientation it measured better without it. I get to be exposed to actual daylight during the day!  

For those unfamiliar with this topic, when I refer to measuring a room, I’m talking about measuring the frequency response of the room - i.e., how sound propagates throughout the space and collects at the listening position.  This is achieved by setting up a microphone to record the dispersement of sound emitted from the speakers - then that recording gets analyzed by acoustic software and presented as a set of data, which can be used to make comparisons.

You might be wondering, how could I possibly achieve equal acoustic results using half the amount of acoustic treatment?  The truth is, I don’t have equal results…I have better results.  This miraculous feat was achieved through a combination of factors.

To start with, much of the acoustic battle can be waged with strategic placement of the speakers and listener.  As you change the position of the speakers, you change the way the room is excited and the way in which sound reflects off the walls, which affects the resulting sonics at the listening position.  And as you change the listening position, the amalgamation of those reflections that reach the listener are altered.

In the old room, the speakers were on console shelves.  This meant that the main speakers and the listener could only be moved together as a single unit, thus limiting the potential acoustic progress you can make using positioning alone.  With the new speakers on dedicated stands, now the speakers and listener could be independently moved, allowing more progress to be made.  In addition, adding a 2nd subwoofer into the equation provides further flexibility, as there are more placement possibilities with two subs compared to one (as well as other acoustic benefits), thus allowing even more progress to be made.

So the first thing that needed to be done was to remove all of the existing room treatment, which resulted in a room that looked like this:

And a garage that looked like this:

Swap out the old speakers for the new speakers, bring in the additional sub, and let the games begin.  In the end, with more possibilities for speaker / listener placement in the new studio, more acoustic issues were able to be addressed with calculated positioning, which meant there was less heavy lifting that needed to be done using acoustic treatment.  But make no mistake, there was still lifting to be done.  Enter Modex Plates.

Modex plates are pressure-based traps that get flush mounted to a wall surface.  I had never used such treatment before, but based on its reported effectiveness, I decided to take a chance and buy a few.  These plates are gargantuan in size (roughly 3’ by 5’) and quite cumbersome to maneuver singlehandedly - two of the plates I purchased weighed 65 pounds each, and the other two plates I purchased weighed 80 pounds each (FYI, I did actually weigh them).  This posed a significant challenge when it came to measuring their acoustic effect in the room, since I didn’t want to permanently mount them until I knew where they would yield the best results.  However, if not mounted or physically held against a wall by a human being, they would topple onto the floor.  But having people hold the Modex in place during a measurement would skew the frequency response, making it difficult to isolate the effect of the treatment from the effect of the bodies in the room.  I needed the room to be vacant, but I also didn’t want to have to put a ton of holes into my walls while experimenting with where these plates should go.  So I put my thinking cap on.  And then I got out the chop saw.

The plan was to lay a couple of boards on the floor in order to raise the plate above the floor trim, thus allowing it to be pressed evenly flush against the wall.  Then to prevent the plate from tipping over, I’d use non-slip pads and pieces of 2 x 4 with the ends cut at a 45 degree angle to hold the Modex in place, and they would in turn be held in place by dumbbells.  The plates were said to be most effective where boundary surfaces intersect (e.g. room corners), so I focused my efforts in the rear corners, as features of the room made that the only viable option for symmetrical positioning of these behemoths.

Success!  I tested each plate type accordingly.  Then I realized I also needed to measure all 4 plates together, in order to assess their collective influence.

Great.  But I also needed to measure how the plates would perform when placed closer to the ceiling.

And of course, all 4 plates near the ceiling at one time.

Brilliant!  I was now ready to commit to placements for 2 of the plates.  As for the others, the jury was still out.  What if I tried putting them on the actual ceiling?  How could I accomplish that without actually permanently mounting them?  Hmmmm.  I’ll place speaker stands on top of plant stands that have an adjustable height via a rotating top.  The speaker stands have a wider base than the plant stands, so I’ll use a sheet of plywood as a medium between the two.  Then with the help of an assistant, we’ll rotate the top of the plant stand until the Modex is firmly wedged into the ceiling.

Boom baby!  I admit that was rather unsafe.  Definitely do not try this at home - one error in judgement, and 80lbs will fall right on top of you - it will not be pretty.  But if you’re as psychotic as me, you’ll realize that the only way to get the answers you need is to proceed accordingly (and cautiously).  Ultimately, a comparison of all the various measurements showed the best results came from this arrangement:

The Modex Plates are sitting on custom-made wood platforms, and the top of the plates are held flush against the wall by 1" L brackets that I installed into the ceiling joists.  This was an easier method of mounting than that suggested by the manufacturer, and it allowed me to position the plates right up against the ceiling / wall boundary (which would not have been possible using the mounting brackets that came with the plates).  The other side of the room has a baseboard radiator, so at my wife’s suggestion we installed a permanent shelf into that wall.

That’s pretty sexy.  Now on to testing my other acoustic treatment.  This absorption and diffusion functions by engaging sound as it is in motion, a design that requires an air gap between the treatment and the surface behind it.  As you change the size of the air gap, you change how the treatment performs.  Similarly, as you change the thickness of a panel, you change its effective range.  So I did an extensive trial of testing, using 6’ columns, 4’ columns, and 2’ columns at every given location, with air gaps between 0” and 5”.  I also tested panel thicknesses between 2" and 6", and compared different types of absorption and diffusion.  And I tested the treatment free-standing directly on the floor, and at various distances off the floor using wood platforms.  If you can think of it, I probably tried it.

With myriad measurements in hand, I eventually settled on the following configuration for the rear side walls:

When it came to the front corners of the room where the subs are located, I concocted a framework of platforms that allowed me to stand treatment on top of the subwoofers.

Then I tested various configurations before making a final decision.  Here are some examples that didn't make the cut:

Ultimately, a diffusion panel 2” from the front wall and 4” from the side wall yielded the best results (as to why those specific distances were best, or why diffusion instead of absorption, I have no idea…all I know is it measured best this way).

In the old room, I had many layers of this treatment (made by Real Traps), which collectively consumed a great deal of the available free space.  In the new room, the independent positioning of the speakers / subs / listener along with the utilization of the Modex Plates (which are less invasive due to their flush mounting), resulted in needing to use a little less than half of the prior acoustic treatment.  Which means, I have more space in the room now.  Which means, it’s time to dance.

For those wondering, the black and white panels depicted in these images do different things.  The black panels absorb low to high frequencies, and the white panels absorb low frequencies only.  I had them colored this way when I bought them so that I would be able to easily distinguish each type.  Any perceived color coding of the room is completely accidental - things just happened to work out in the way that they did.

As for the ceiling, the two panels above the speakers (and those to the left and right on the side wall / window) are absorbing the 1st reflection points between the main speakers and listener.  Another 2 panels on the rear ceiling took care of a ringing that would have interfered with recording a vocal or instrument in the room.  Then I placed some small diffusor squares around, which had a subtle but positive effect on the room's high frequency response.  FYI, the blue lines in the following image are chalk markings - left to right indicates the location of ceiling joists, and front to back frames the center of the ceiling.

The last step was to dial in some corrective parametric EQ for the icing on the cake.  For the audio novices among us, EQ is short for equalization, which is a process that is used to modify the sonic characteristics of an audio recording.  With respect to room acoustics, the sonic characteristics of a speaker can be adjusted using EQ, so as to offset certain features of the room - for example, if the room is creating excessive energy at 70Hz, EQ can be applied to reduce the amount of 70Hz before the sound comes out of the speaker in order to compensate for the room’s effect, thereby creating a balanced sound at the listening position.  One of the beautiful things about Genelec’s digital line of speakers (in this case, a pair of 8260s and a pair of 7270s) is that each speaker individually contains a comprehensive set of EQ filters that can be independently adjusted and fine tuned with precision.  Such corrective EQ can be quite beneficial, particularly when dealing with smaller rooms where it can be impractical or impossible to address certain anomalies any other way.  Indeed, if acoustic treatment is a butcher’s knife, EQ is a scalpel.

You might be wondering though, if we can just EQ the speakers to compensate for the room, then why was this entire acoustic treatment fiasco even necessary?  The answer is that EQ by itself is insufficient for the task: A) it cannot compensate for nulls in the frequency response, which is where sound reflections cancel each other out, therein creating a void or “hole” in the audio, and B) in an untreated room, the frequency response can vary greatly as you move within the room, thus rendering any EQ adjustments only relevant to a narrow listening position.  So you need to address as much as possible using acoustic treatment first, before moving on to consider corrective EQ measures.

With everything said and done, and around 250 measurements later, the new room sounds phenomenal.  A couple of tissues taped to the ceiling to diffuse the light, and this studio is officially a wrap.  Yes, it was all worth the effort.  And now that the studio is finally finished, I can begin making some music again.  Well, at least until the next studio escapade!!!!

UPDATE 8/12/15: There has been one more important addition to the studio - see here.