I was fortunate to work several campaign events this year. At the most recent one there was a heavy emphasis placed on audio. It was the last event of the campaign and they were pulling all the stops to win as many votes as possible.
The event took place in a 10,000 seat arena. Two stages were built, one at each end of the arena. One stage was for a rock band and one was for the Presidential Candidate. Each stage had a complete arena sized PA System flown above it and firing toward the opposite end of the arena.
Rock band PA is a no brainer. OK, there’s actually a lot of thought, measurement, and calculation are involved but you get the idea… Now a presidential candidate… a particularly soft spoken candidate using the traditional dual SM57 podium mic system… that’s a bit more tricky, especially when the client wants both systems at equal volume.
Actually, no it isn’t.
There are as many schools of thought regarding system tuning as there are people trying to tune them. Some are based on solely scientific fact, some are based solely on experience, while others are based merely on time available for tuning.
I combine all three.
In fact, I align and tune a PA system the same way every time. I don’t care if it’s Black Sabbath or The Pope using the speakers. The laws of physics still apply and if the system EQ is all hacked up and pulling 6+dB out of the PA before any input is considered, the system will quickly run out of headroom and sound awful.
A basic guide to equalization can be summed up like this:
1.) System Processor Output EQ is used for correcting anomalies between component (driver) and cabinet. These are usually manufacturer recommended filters.
2.) System processor input EQ is used for system to room correction an intonation. This is where the Systems Engineer makes his or her money.
3.) House EQ (either outboard or digital console output EQ) is used in conjunction with processor input, but I tend to leave it flat unless absolutely necessary. This is where the FOH engineer puts in their particular tastes in tonality.
4.) Sub Group EQ is used for intonation of groups of input. This is a great option when afforded the capability.
5.) Channel EQ is used for correction of tone differences between source and transducer. This is the first line of defense. For example: a human speaking at a podium through a PA system should sound like a human speaking anywhere else without a PA system only louder. LISTEN to the human without the PA. Then LISTEN to the human through the PA. Make EQ adjustments within the channel to ensure that the voice sounds as close to natural as possible.
The first step is to time align the PA. The next blog will cover this in detail, so I’ll just glance over this for now. In the most basic terms, time aligning is the process of measurement and time delay adjustment utilized to ensure that all information from the different sound sources (Mains, Subs, Front Fill, Etc…) arrive at the same fixed point at the same time. This ensures that there won’t be destructive interference between PA zones.
Once the system is time aligned, it can be tuned. Contrary to popular belief, PROPER PA tuning doesn’t mean popping in a favorite track and making it “bump”-ESPECIALLY if the source information being mixed sounds nothing like the source information used to “tune”. Before listening to music, use the pink noise generator on the measurement/analysis equipment or onboard the console and conservatively apply parametric EQ filters (don’t touch the graphic eq, we have use for that later) to correct the frequency discrepancies across the spectrum compared to the pink noise reference. This is digitally verifying that the information coming out of the PA is the same as the information going into it.
Now that the PA is in time and “flat”, making it “musical” should be a snap. Here’s where the musical track comes into play. (…a note about test music… MP3 files SUCK! They are over compressed, distorted and much of the musical information within the wave form was lost when being compressed. Use a WAV file or a CD.) “Musical” is a matter of taste. Bear in mind that if you find yourself boosting or cutting multiple frequencies or wide parametric filters within the bands of an output section (i.e. subs, lows, low mid, high mid, high frequencies) it would be more beneficial to leave those EQ filters flat and to turn that output section up or down accordingly.
After the PA is aligned, tuned, and musical the soundcheck can begin. With our dual SM57 podium the first order of business was to find an analogue for the speaker. Presidential Candidates, like our rock star clientele, do not soundcheck. Therefore we need to find a speaker of similar size, demeanor, volume, and vocal intonation… Or someone who can fake it well. That person’s job is to stand at the podium, open a news app on their cell phone and start reading the news until the FOH says stop.
I try very hard to avoid drastic equalization. A good approach to EQ is: make the noise from the mic sound like the noise from the source. It’s that simple. Really. Once the sound coming from the source sounds the same without PA, in headphones and with PA, take your hands off of the channel EQ. (an exception to this could be a kick drum since I’ve never heard a 22” kick drum sound like a typewriter without a ton of EQ help.)
Now that the source and the mic sound as close to identical that they can, it’s time to make it loud. The system WILL feed back initially. This is good . We want it to feed back. We want to know where our trouble spots are. Hopefully your newsreader isn’t a total jackass and doesn’t mimic the tones ringing. There’s nothing beneficial from hearing a 1k feedback tone followed by some prick generating the same tone with his voice because he thinks the engineer can’t hear it.
If the system is properly tuned, all that is left to do are minimal EQ adjustments at the trouble frequencies. Here is where the graphic EQ comes into play. Many engineers can aurally distinguish feedback frequencies, however the foolproof method is to use a spectrum analyzer. As your speaking analogue talks, push the system. When feedback happens it will be visible on the analyzer. Find the corresponding handle on the graphic and pull out 1/2 db at a time until the feedback disappears. That’s right, I said 1/2 db. It’s a common practice to grab frequencies in 3 db increments, however if the system is properly tuned this is overkill. You will find that minor adjustments will suffice and you will have more overall headroom using this method.
All that’s left at this point is to find catering and wait for the show. Walk away. Don’t sit there and over think the EQ. Don’t second guess yourself. Don’t dig a hole before the show starts.