A combination of "State Of The Art" and  "Vintage" Recording Technology and Creativity 

By Steve La Cerra
As a producer and engineer, that Michael Sosna has spent quite a bit of time in recording studios. In addition to working with a variety of artists, session musicians, and vocalists, he's worked on more than 1200 on-air projects, including award-winning national and international television spots. Michael's company, Manhasset Bay Associates, Inc., a member of the SOS Enterprises Group, has evolved over eighteen years, acquiring vintage equipment as well as serving as a beta test site for soon-to-be-released studio gear from high-end audio manufacturers. With all that experience under his belt, Mike recently embarked upon designing his latest studio, SOS PRODUCTIONS. Mike had a very specific and unique concept of how the studio would be configured and operated. My mission was to connect all this stuff and make it work.

 Ground Zero
Before examining any of the equipment requirements or technical details of the studio, it's important to understand Mike's vision. The studio is based around three Yamaha 02R's that would be linked together for seamless operation, so they could be thought of (and worked on) as if they were one large-format digital console. In order to obtain the elegant look of a coherent console, Mike contacted Todd Beeten of Sound Construction & Supply, Inc. (Nashville, TN) and ordered a custom oak console that integrates the three 02R's, while also providing a monitor bridge. In addition, Todd designed three matching six-foot racks and a beautiful MIDI workstation.
Equally important to Mike's vision was the ability of the room to be used for transfers between various digital formats such as 16-bit, 20-bit, and 24-bit word lengths. Every step of the way toward completion of this project, Mike and I would keep these goals in mind, basing our decisions upon that vision. Lucky for me, Mike was to handle all of the equipment acquisition, ordering, phone calls, and general legwork, leaving me free to concentrate on the wiring aspects. 

Drop Off The Laundry
Until recently, Mike was using an Otari Soundworkshop Series 34C console. It's integral patchbay had been augmented with external bays, enhancing the patch capability to include microphone lines.  Here was an opportunity to get down and dirty with a new control room, a studio, and three isolation booths. 
Without getting into a laundry list of gear (which is easy because Mike loves his toys!), the studio centers around the three Yamaha 02R's, with Version 2 software. Recorders include three 24- and 32-track systems, including four ADAT XT20's, a 24- and 32-track Pro Tools|24 system with three Digidesign 888/24 (24-bit) interfaces, an Otari MX-80 2-inch, 24-track analog machine (augmented with a set of 32-track heads), three Panasonic SV-3800 DAT machines, two cassette decks, a CD recorder, and several pairs of monitors including Genelec 1031A's, Big Red's, Yamaha NS10M's, and Auratone 5c's. Outboard gear abounds with Avalon and AMEK 9098 mic preamps, Pultec equalizers, Neve, Shep, and George Massenburg Labs parametric equalization, (12) dbx 160 VU series compressors, (12) Lexicon units including a 224XL w/Larc, Klark-Teknik DN780, (2) Roland R880's, TL Audio C1 and EQ1, a Mark Of The Unicorn Digital Time Piece (which serves as the timecode and clock master), a TC Electronic Finalizer+, and a MIDI-controlled feeder for Mike's beautiful Siberian Husky, Sophi. All of this gear, plus the "120-input" Yamaha 02R, would be tied together with a 900-point TT analog/digital patchbay. A new, 400-amp electrical service would be installed by Mitch Glider of MSG Mechanical to cleanly power all the gear. Needless to say, it'd be quite the wiring challenge.

 Live Wire
My assistants for this project were Shane Valentine and Hon Pun Chan. Ordinarily, we'd get together a list of gear, determine which I/Os need to be accessed, and how many patchbays are needed to accommodate those points, and then write up specs for wiring harnesses (snakes). Typically, these specs would be faxed to a wiring manufacturer for custom termination, and a few weeks later we'd get boxes of snakes with various connectors at one end and the other end tinned and stripped - ready to be soldered to the bays. But this gig was different because Mike's studio had been previously wired, though with significantly different gear. The result was that there were hundreds of feet of multipair wire that could be used on the renno. Good news: Mike didn't have to buy all new wire. Bad news: Guess who gets to prep the existing multipair? Shane, Hon, and yours truly (ugh!). 
Paramount to determining what wire could be retained was the location of the gear. Mike spent quite a bit of time and care physically laying out the flow of the room and the arrangement of the consoles relative to the racks and synth station. By the time he was ready for the wiring design, he already had a very good idea of where he wanted the various gear to be located. This made my job easier, and should be one of the first steps in any control room design, because if you fudge the physical layout of gear, it's almost guaranteed that you'll have at least one snake that will not reach wherever it is that you most need it to reach. 
Once we knew gear locale, Shane, Hon, and I measured routing distances from the racks and console to the patchbay (with an extra foot or two for good measure). Then we figured out how many audio lines we'd need to run between the bays and the gear. Note that we measured routing distance, not physical distance. As an example, the rack closest to the desk was only about three feet away from the patchbay. But the wire routing distance between them was 16 feet (down the back of the rack, under the floor into the cable raceway, and up the desk into the patchbay). Our measurement of "distance to the patchbay" also included enough slack to pull a bay out of its rack slot and onto the producer's desk for easier soldering-in-place or future modifications. 
We then made a list of the harnesses we'd need and assigned unique control numbers to them for identification purposes. The existing snakes were laid out, measured to see what could be kept, and labeled with a control number. A spreadsheet was set up for each control number, listing the harness' description (e.g., "Otari MX-80 outputs"), the number of audio channels in the snake, audio source and destination per channel, ID number, and connector type. Control and ID numbers were assigned to every snake and correlated to a patchbay location.
Mike wanted the first row of patch jacks to be the microphone lines coming in from the studio to the control room, so we started with "Control 1a" for the first group of microphone lines (1 through 9), which came from the drum booth. Control 1b was used for microphone lines 10 through 30, which are fed from the other iso booths, and the live room via a separate harness. Numbering snakes this way helps keep brain damage low when it comes time to solder the lines to the patchbay: any Control 1X bundle goes to the first row of the bay. 
Shane and Hon used Panduit ID stickers to number each individual audio line with an ID number to match the patch point. For example, microphone line number 15 was ID'd with "115" because it would be soldered to row 1, point 15. After the lines on a snake had been ID'd, Shane and Hon used 3/16-inch, clear heat shrink over the label to make the ID tag permanent. Eventually this would be done to more than 30 snakes - a total in excess of 800 audio lines (ouch).

 02R x 3...
Before we go any further, a bit of an explanation of our 02R configuration is necessary. In order to link the three 02R's, cascade cards must be loaded into rear-panel slots on the console. Almost all of the analog inputs and outputs from all three consoles would be used in our system: mic inputs, line inputs, stereo inputs, inserts, and aux sends. The major difference between the three consoles is that the 02R designated as the "master" would be the only console that had its master L/R I/Os (both digital and analog) connected. As you sit behind the desk, the consoles are numbered 1 through 3, from left to right. Console #3 is our master, so this is the one that has the master I/Os connected. 

Miking the Lines?
Microphone inputs to the 02R's would be in row two of the patch bay, with row one half-normalled down to row two. Here's where things became tricky because the number of mic inputs in an 02R depends on how you look at the situation. Inputs 1 through 8 on each console have XLR mic inputs with phantom power, and a separate TRS jack for the line input. Inputs 9 through 16 accept mic or line level on a single TRS jack, but don't provide phantom power. We could have brought up channels 9 through 16 (the TRS jacks) as "mic" inputs and normalled microphone lines 9 through 16 to "mic" inputs 9 to 16. But then every time we wanted to run a line-level signal into a channel between 9 and 16, we'd have to patch that signal into a "mic" input. Confused? That's exactly the point. Imagine a guest engineer trying to figure out that he needs to patch a CD player into a jack labeled "mic in!" A second problem would be that we couldn't normal a line-level signal to a mic input, so every time you'd need a line in, you'd have to patch. 
Our solution was rather unorthodox. We viewed the situation as if there were a total of 24 microphone inputs: eight with phantom power for each 02R (how convenient - this makes half-row on a Switchcraft TTP96 patchbay). The first 24 microphone lines are normalled to these 24 mic inputs. Eventually, when we labeled the patchbay, we'd use a color code for each 02R. The last six microphone lines are normalled to outboard mic preamps.
OK, so what about line inputs? Each 02R has 16 (analog) line inputs - channels 1 to 16 (on channels 1 through 8, the line input is known as the "B" input). Sixteen line inputs x three 02R's = 48 line inputs - another row of 48 patch points. We arranged the line inputs sort of like the mic ins. Console number one (the one on the engineer's left) is home to line ins 1 through 16. The middle console is home to line ins 17 through 32; console 3 hosts line ins 33 through 48. 
A total of 24 inserts were also included in the patchbay - eight for each 02R. Though they'd be labeled 1 through 24, in reality these inserts were on channels 1 through 8 for each of the three consoles (these are the only inserts on an 02R). Since our mic inputs were on 1 through 8, wiring the inserts like this gave us the ability to patch analog processors on the mics during the recording process. A similar concept was applied for the 02R stereo inputs (four pairs per console). The first set comes to the bay designated as stereo inputs 17A through 24A, the second set as 17B through 24B, and the set from the last console as 17C through 24C (this makes another half-row). These would also be color-coded on the patchbay labels to match the inserts and line inputs.
The remainder of the patchbay was laid out mostly for Mike's working preferences and design goals. One of the more interesting aspects was that we normalled the analog outputs from the Digidesign 888/24's to the inputs of the 24-track machine, facilitating 20- and 24-bit A/D and D/A transfer capability to and from all formats. Synth outputs from the MIDI workstation are normalled to 02R analog line inputs 1 through 24, while the outputs of the MX-80 are normalled to 02R line inputs 25 through 48. Because of the fact that more than 15 reverb and echo units were installed, none of them were normalled to the 02R's - they all must be patched.
In addition to these analog normals, a lot of hard-wired digital connections were made, many by the installation of 12 expansion cards in the 02R's (five ADAT cards, three AES/EBU cards, and four cascade cards). As mentioned previously, each 02R has a cascade card installed so that bussing and aux sends between the three consoles may be digitally linked (one of the consoles has two such cards). As a result, turning up send one on channel two dumps the signal via cascade from console #1 to the master console, which then routes the signal to an outboard processor. (This is a simplification. For more detail you can consult the 02R user's manual.) A total of five ADAT cards were installed, each with optical, digital audio I/O to and from the XT20's (this was in addition to the analog I/Os). Three AES/EBU interface cards accommodate the three Digidesign 888/24 interfaces. 
Finishing off the digital routing is a standard Switchcraft TT96 patchbay used for patching and normalling of digital stereo pairs. Interestingly, the analog patchbay has no normal from the 02R master output to SV-3800 #1. That's because this normal takes place in the digital domain via the digital patchbay: AES/EBU master out of the 02R is normalled to the AES/EBU input of SV-3800 #1. The S/PDIF master out from the 02R is normalled to SV-3800 #2 S/PDIF in, allowing Mike to simultaneously mix to two DATs without patching. Two-track returns were handled similarly with SV-3800 #1's AES out normalled through the digital bay to master 02R's digital 2-track return #1, SV-3800 #2's S/PDIF out normalled to digital 2-track return #2, and the S/PDIF out of the CD-R normalled to digital 2-track return #3.
Since this arrangement leaves the analog L/R output from the master 02R unused, it was normalled to the cassette deck inputs. Although one cassette deck would initially be installed (a TASCAM 122 Mk III), Mike anticipated adding a second, so the wiring is prepped for deck #2. Inside the bay, the master L/R out patch points were paralleled to a second pair of jacks, enabling us to normal the one set of master outs to the two cassette decks - again to reduce patching. Output from the 122 Mk III was normalled to the +4 analog return of the 02R. CD out (from a Sony consumer CD player) was normalled to the "10 analog return of the 02R. It's interesting to note that Yamaha has built in gain compensation for the "10 input, so you don't have to lunge for the volume knob when switching between the various inputs in the monitor section. 
Some of the extra goodies in the patchbay include four, four-point mults, two phase reverse pairs, breakout pairs to a rack panel in the control room for guest gear, video ties to a video monitor and video decks, eight ties to the studio shop, and a total of 20 ties to the separate MIDI suite area. In all, 18 02R analog aux outputs appear at the bay: six from each 02R. Although it's the six sends from 02R #3 that act as the master, we felt it'd be a good idea to have the option of connecting sends from any of the consoles - just in case. (According to Mike, "The sends on the slave consoles have actually been quite the blessing and very, very useful during tracking, overdubbing, and mixing.")

Now Hear This!
Quite a few monitoring options were installed to accommodate Mike's choice and location of monitors - a tricky proposition since he's using a mix of active and passive monitors. The control room output of the master 02R goes to the bay and is normalled to the input of a line-level switch box. One output of this switch box is hardwired to the input of the (active) Genelec 1031A's. A second output of this box is hardwired to the input of a Bryston 4B power amp. Output from the Bryston amp comes back to a second (speaker-level) switchbox, which then routes the power amp to either the control room mains (Big Reds), nearfield #1 (Yamaha NS10M), or nearfield #2 (Auratone) monitors. A third output from the line-level control room switchbox is sent to a Yamaha P2200 power amp that is used to run another pair of NS10M's located at the synth workstation. 
Similarly, the studio output from the master 02R is normalled to a second line-level switchbox, which outputs to two additional amplifiers:a Furman HA-6 headphone amp for cue purposes and a second Yamaha P2200 for the studio monitors (still under debate). All of the speakers and the headphone lines employ Monster Cable speaker cable. Intended to interface with Furman HR-6 remote headphone boxes, the Monster Cable speaker lines for the headphone feed terminate at male XLR receptacles in the studio and iso booths.
To be continued.... Next month we'll take a look at the clock synchronization and MIDI routing of SOS Productions.

Michael Sosna of The SOS Enterprises Group is an Independent Producer/ Audio Engineer who has worked with Patti Austen, Randy Bachman (BTO), Hiram Bullock, Michael Bolton, Ron Carter, Johnny "Clyde" Copland, Miles Davis, Anton Fig, Lou Gramm (Foreigner), Benny Gramm, Thelma Houston, Whitney Houston, Al Jareau, Lenny Kravitz, KISS/Paul Stanley, Will Lee, Delbert McClinton, Lenny Pickett, Lou Rawls, Nile Rodgers, Philippe Saisse, David Sanborn, Paul Schaffer, Phoebe Snow, Grady Tate, Southside Johnny, Stanley Turrentine, Martha Wash and Bebe & Cece Winans. He has also produced and engineered audio, music, sound design, and voiceovers for companies such as Apple Computer, AT&T, Chevy, Coke, Burger King, Cadillac, Caress, Dupont, Folgers, Ford Trucks, General Motors, Gillette, HBO, M&M's, McDonalds, NBC, Pepsi, Pizza Hut, Royal Carribean, Visa, and for television spots featuring Aretha Franklin, Michael Jackson and Madonna. Mr. Sosna has the unique ability to develop talent and to bring out the best in people in and out of the studio. He was recently appointed Music Producer and Audio Consultant for "Cyberjam".