Journey To The Same Destination

Journey To The Same Destination

Visit the offices of Advanced Automation Corporation (AAC), a systems engineering company based in upstate New York, and you’ll see all of the things you might expect from a technology and engineering workspace: whiteboards covered with meter-long equations, charts comparing the capabilities of various types of satellites, schematics of aircraft propulsion systems, radar testing equipment, and of course a group of engineers huddled around the coffee machine.

Look a little closer, however, and something unexpected becomes evident: nearly every single employee is a musician. Guitars can be found in most offices, and it’s not uncommon for the hallways to be rumbling with high-SPL rock music. After putting in a day’s work on their projects, the employees head across town to a local bar called the Snubbing Post, where they perform as the band Tyler, named for the couple who heads up AAC: CEO David Tyler and his wife Patty, who is the company’s president.

Although the Post represents one of the only places in the area for local bands to perform, the stage and sound system had become quite dated over the years, and the band decided to pool their engineering minds to do something about it. As the band’s audio engineer, I was invited to assist with the project.

The Big Picture

A bit of venue history dictated the deadline: the Snubbing Post has hosted a weekly Open Mic Night every Thursday night for more than 19 years – over 1,000 consecutive weeks – so once the work began, it had to be finished quickly.

The sound system upgrade was just one facet of the overhaul. While various AAC employees and contractors got busy replacing and expanding the stage structure and drum riser, installing new carpeting, painting the walls and updating the stage electrical wiring, I was to tackle the unique challenge of designing a sound system that would be as easy as possible to operate for performers lacking an audio engineering background. Given that the project was funded entirely by donations from the band members (one of whom even purchased and donated an entire drum set!), I had to get as much mileage as possible out of a small fixed budget.

I’ve too often witnessed efforts to try to save money by installing a less powerful system, which is then relentlessly run into its limiters every night and fails far more quickly than a properly spec’d system, so in the long run, it costs less to get it right the first time. After some discussion, we settled on QSC K.2 Series loudspeakers as the cornerstone, even though the purchase would require most of the available audio budget.

 

The rack-mounted Yamaha MG20XU mixer and dbx 231s graphic EQs on stage, with one of the QSC K12.2 main loudspeakers at the edge of the stage.

 

 

 

Two K12.2 boxes would replace the existing left/right mains, which were bulky, ancient passive models driven by a powered mixer (via guitar cables!). The balance of the system would be a single KW181 subwoofer and a trio of K8.2 loudspeakers serving as floor monitors, all fronted by a Yamaha MG20XU 20-channel analog mixer.

The mixer would be vertically mounted in a rack on stage along with four channels of dbx 231s graphic equalizers for feedback control of the main and monitor mixes. (This invites the concern that someone will come in and crank the lows and the highs into the evil smiley face. My solution is to leave the EQs bypassed by default. People who know what they’re doing will know to disable the bypass switch, while those who don’t know any better won’t.)

The rig would have a small footprint and allow onstage musicians to easily adjust their own levels, while larger bands who bring their own engineers could still quickly patch in and mix their show from their own consoles at front of house. The onboard DSP of the K.2 Series meant that any necessary tuning could likely be accomplished in the loudspeakers themselves, eliminating the expense of a system processor.

We also ordered QSC’s mounting brackets to suspend the loudspeakers in a more cosmetically pleasing (and probably safer) manner than the venue’s existing chains-and-plywood solution, along with lock-out covers to guard the loudspeaker controls against tinkering after the final settings were programmed.

 

I knew my time to work in the venue would be extremely limited so I did as much of the verification process as possible at the shop. I tested the cables, console, and loudspeakers individually before assembling and testing the entire system. My first step in loudspeaker verification was to make sure that multiple units of the same model measured the same, which helps detect blown drivers, polarity reversals, or shipping damage.

Figure 1 shows two sets of magnitude traces, with each group of loudspeakers measuring close together. Room effect is clearly visible, but that’s unimportant: emphasis is on detecting any difference between the traces, not the absolute responses.

Figure 1

Next up was getting the loudspeakers as close to my target tuning curve as I could, so that once they were installed, I would only have to compensate for the change in room, and hopefully reduce my tuning time. Due to the reflective shop space, I used a few spatially distributed measurements to get a clearer picture of the overall response.

Figure 2

There was a midrange response bump that I reduced by a few dB. I didn’t want to touch anything below about 200 Hz since the response in that range is so heavily dependent on the local environment, but I did set the onboard high-pass to 100 Hz, which is QSC’s recommendation when pairing these boxes with the KW181. After confirming that the phase response did in fact match the sub in the 63 to 125 Hz range (Figure 2), attention turned to the monitors.

I’ve learned from experience to take steps to reduce the onstage pileup from multiple wedges, especially on small stages. After evaluating several of the DSP presets, I ended up using the onboard EQ to manually create a low-frequency contour that dealt with the wedges’ LF coupling off the floor and provided sufficient LF extension to support a vocal fundamental, but no more.

Sub wash across the stage would be significant, so extended LF response would be unnecessary.

Figure 3 shows the out-of-the box response (black) versus the final EQ choice (red), where I opted to reduce the wide bump around 1 kHz. Again, there are significant room effects in play, so I considered this a good start with the recognition that I’d likely need to do some tweaking once they were installed.

Figure 3

I suspected that the ramp-up response above 4 kHz would sound too harsh, but again I decided to hear it in context on stage. This tuning was copied to the other wedges, then I lowered the high-pass filter a bit on the box assigned to be used as a drum monitor because drummers often ask for some bass in their monitor.

The Real Point

This seems as good a place as any to address the dual-channel elephant in the room: isn’t it a bit ridiculous to send in a system tech with an analyzer to optimize a sound system at a venue not much larger than a basement? There’s a certain feeling of detachment than can creep in when reading work of some of the industry’s optimization pioneers – “Sure, you get to play with million-dollar systems, and I’m stuck mixing speakers on a stick.”

My point in sharing this project is to demonstrate that regardless of scale, the goals are remarkably the same: minimal spectral variance and minimal level variance. I could be in a 15,000-seat arena or a bar with standing room for 15, and either way, I want everyone to hear the same show at the same level.

The biggest mistake I made in my journey of learning about system optimization is fearing the analyzer. This stuff isn’t black magic, it’s the nuanced application of basic principles. Even the analyzer technology itself is easily attainable – the excellent Room EQ Wizard software is free to download and can be used to support all basic system optimization tasks. Simply put, there’s no barrier to entry.

If you’re curious, there’s no excuse not to be out there doing this stuff. Even with this simple-as-they-come, one-box-per-side point source rig, we were able to achieve a remarkable level of front-to-back tonal consistency, with nothing more than a game plan and the loudspeakers’ onboard DSP.

Installation Day

The venue’s layout is asymmetrical, shaped like a capitol “L” with the stage positioned along the short dimension, which means the house right loudspeaker position is doing most of the heavy lifting for most listeners.

Figure 4 shows that the front-back level variance is significant, about 12 dB from the front of the listening area to the guest mix position, but that’s the inescapable physics of point source geometry, and in this context it’s not necessarily a bad thing.

This is a bar, after all, and the level drop-off past the mix position means that patrons can retreat further away from the stage to converse or move closer to watch the performance. A 12 dB front-to-back difference is also less than one would expect based on the distances: aiming the on-axis points of the loudspeakers at the furthest part of the coverage allows us to play the off-axis level loss against the inverse squared law’s level gains as the listener approaches the loudspeaker.

Figure 4

The red line above the magnitude trace in Figure 4 indicates that the coherence (shown here at guest mix position) isn’t fabulous for such a small space. There are plans to install acoustic treatment on the walls and ceiling around the stage, which should improve system intelligibility as well as help with the stage volume of the acoustic drum set, which is one of the biggest challenges mixing in this space.

However, overlaying the traces (Shift-click in Smaart, thanks to Jamie Anderson at Rational Acoustics for this handy tip!) allows us to overlook the level variance and reveals that the tonal response over the listening area is actually very consistent (Figure 5). We can have high confidence that what we hear at mix position will translate well to the whole audience.

Figure 5

Although my personal tuning preference tends toward a flatter target curve, the goal with this system is for musicians to sound good with as little work as possible (insert musician joke here).

Tilting the system response closer to the slant associated with modern concerts should (hopefully) allow bands to get closer to this sound without having to use as much EQ at the mixer, which is a good thing considering that the target users are not sound engineers.

Figure 6 shows the same three mic positions after rolling back the high-frequency response of the mains and adding the subwoofer. The slight dip around 2 kHz turned out to be quite audible upon listening to my reference material, so I brought the response back up by about 2 dB in that range and everything sounded good.

Figure 6

When I fired up the monitor wedges, the HF response noted earlier was indeed excessive, and it was easy to bring it back into line with a single hi-cut filter. With a comfortable 20 hours to spare, the new stage and system were ready. I’ll be listening as artists take the stage at tomorrow’s open mic night with an ear toward any final tweaks.

In Context

Loudspeaker verification in process at the shop; this type of prep ahead of time paid off come installation time.

Takeaways for me on this project were twofold: on a technical level, spending some on tuning the individual loudspeaker responses before getting into the venue really speeds up the process later on: the final tuning in the room took less than 15 minutes. Getting all the little things right along the way leaves very little “big” work to be done.

On a broader scope, this project reminded me how important it is to filter all specification and optimization decisions through the lens of how the client will use the system.

I would have done a lot of things differently if I were putting together my own rig, but I’m confident that this finished product presents a much better fit to support the venue’s bands and open mic night performers.

What’s more, I had a great time working to support a group of engineers and musicians making a meaningful improvement to something so important to them.

Leave a Reply

Your email address will not be published. Required fields are marked *