Friday, June 1, 2012

Time Alignment

Many times in my travels as either a mix engineer, or as a design engineer, I come across problems in systems that are due to time alignment issues in the system. Even when I teach people about mixing live, I always have to keep reminding people that sound lives in the time domain, and needs to be considered, especially when putting a rig together. I'm reminded of a system I designed and deployed for a church a couple years ago. It was a typical long narrow cross shaped church. They had decent speakers on the columns going down the room on each side. The speakers were appropriate for the application, but the problem was, there was no DSP delaying the speakers to be in time with each other. The system sounded like a mess. voices were unnatural, and you could even discern hearing the voice twice in some places in the room.  It was very hard to concentrate on the source with all the sonic chaos going on in the room....

In the realm of physics,  sound is kind of slow. It travels at 1,126 ft/sec, which is dependent on temperature, and humidity. Electricity approaches the speed of light, which is 186,282 miles per second. There's a big difference in speed here.  So if you have one speaker in front of another, the sound in the closer will get to your ears first, and the rear speaker will arrive later. Because of the different arrival times, frequencies will be out of phase in varying degrees. The net result is that it does not sound correct. What is coming out of the speakers is not what your ears are hearing. 

Enter a digital delay. Delaying speakers electronically will give you the ability to have the sound from the front and rear speakers to arrive to your ears at the same time. The great advantage is, you can distribute speakers in a space to give you even SPL in all sections of the room. In a long narrow church like I mentioned above, it saves me from blasting the front of the room to reach the back. Once the speakers are up, measure with a tape measure the distance, and either do some math to figure out how many milliseconds of delay you need (google search the formula), or most newer DSP's have field that lets you fill in the distance, and it does the math for you.

After your delay time is set, listen to it, and make sure your source sounds natural. Now, the crucial part to using delay fills: The gain of the speaker is very important. Typically you want the audience to focus on where the original source is coming from, which is probably at the front of the room.  We need to use volume to trick the brain into thinking that the sound is coming from the front of the room, not from the speaker that is aimed at them. I usually start with the front and rear speakers at the same volume. I gradually bring the fill speaker down in volume, until I think I can't hear it anymore. Once I reach that point, I mute that speaker, and see if i can tell the difference between if it's off or on. I might bump the gain a little bit, it is a bit of a "feel" thing. They important thing is that the audio is intelligible, but you don't recognize that the audio is coming from the fill speaker. I repeat the process for each fill down the line, making sure that coverage is seamless, but the focus is up front.

Definitely take your time when you're doing this, and make sure your ears are the final judge. Have fun, and good luck.

No comments:

Post a Comment

You're the Scotty to our Kirk