Echo in Sanctuary

[audiomatic]

I'm attending a church that has a very obvious echo in the sanctuary. You can clap your hands in the middle of the room and hear several very fast repeats. I'm shopping around for ideas as to how we can determine the cause. One sound tech thinks the echos are coming from the side walls, and one says they're coming from the back wall; they've been doing this for some time, and I've only been attending a short while and I'm deaf in one ear, so I have no idea except that I'm fairly certain that the ceiling is O.K.

Short of recording the echo and playing it back slowly enough that we can estimate the milliseconds between repeats, is there a way that we can make an educated guess as to which walls may benefit from treatment? It's a fairly new building and hasn't yet been paid for, so there's no budget for correcting the problem. I'm just hopeful that something can be done to better define the problem so that a solution can be devised and costed.

The sound system is a separate matter. For now, I'd like to try to get the room under control.

Thanks in advance for any ideas.

[cmchamp]

sound will propagate between any parallel surfaces.

[Josh Scott]

It is nice to see someone fixing the room before spending money on a PA.

[Brad Weber]

I've seen people temporarily hang batt insulation, draperies and even sleeping bags to assess if they can isolate the offending surface.

You mentioned it being a fairly new building, was there an Architect or Acoustical Consultant involved in the design? If there was then you might want to contact them and get their help.

If there wasn't, then it might be worth considering hiring someone to assess the existing acoustical conditions and make recommendations. I understand that this sometimes seems a financial impossibility but the alternative may end up being spending money on ineffective treatments.

[audiomatic]

Thanks guys,

Brad, I'll do some rough measurements to see what it may take to temporarily hang sleeping bags, etc., and look into who designed the building. I'm most interested in your suggestion to have someone assess the existing acoustical conditions and make recommendations. If I had a better understanding of what was involved in this, I would like to do it myself. I don't want to take work away from anyone, but my day job isn't very interesting and this kind of thing fascinates me. I know I could slap a couple of 2X4s together, record the room, slow it down on playback, and read milliseconds between repeats, thereby estimating the distances between offending reflective surfaces. Is there something else I could try, such as a model or a particular testing device? What would an expert do to "assess the existing acoustical conditions?" If I can afford it, I would jump at the opportunity to do it myself. If not I'll shop around for nearby experts to do it for me.

Thanks!

[cmchamp]

I can tell you what I've been taught to do.
1: set up no less than 2 omni directional microphones in the sanctuary, (for recording) the first in the front 1/3rd, the second in the rear 1/3rd, but no on any symmetrical axis.
2: record multiple 24" latex balloon pops. In order to achieve a broadband impulse (noise) the ballon must be popped immediately after being inflated.

From here I deviate from the instruction because I don't have the software my teacher wrote for himself to calculate RT60 times across 8 octaves.

I use my DAW to create multiple tracks of each balloon pop with octave band filters. I then apply a noise-gate filter to cut below -40dB (as my DAW doesn't have a -60 option). I find the sample # at the beginning of the impulse and then at the end of the gate, do the math, then calculate from the resulting RT40 number, the equivalent RT60 number.

C.

[buggyboy]

just a thought...why not create a sound absorbing box of sorts in which you clap (or other wise make noise) so that the sound coming out of the box is traveling in more of less one direction?

of course, the waves will spread out, but it might give you enough to hear the difference when you point the box at different walls.

you might get away with using a speaker to generate the directed noise.

and of course, be suspicious of parallel walls...especially if they are very hard surfaces.

another thought, sound reflects just like billiard balls (ya ya, still). if you have a helper stand and clap...while you move around the room, you might be able to follow the echos to a surface.

and finally, if two walls are playing ping pong with sound waves...they should eventually make feedback. the frequency of the feedback may prove useful in determining what walls are involved.


(standard disclaimer, i know nothing, nothing!)

Good luck

[cmchamp]

http://paws.kettering.edu/~drussell/demos.html

for a bit more understanding of how sound waves work

[buggyboy]

I was tired when I wrote that last post.
two things,

one. My main impulse here was to provide options, not necessarily the best, or even good options, but more ideas. There is always the "right" way to do things, but most church sound techs have no budget, no help, only a passion to help the ministry however they can. (sliver and gold have i none, but I do a have a little free time, and the willingness to try to fix acoustics with no budget)

two. since we are dealing with using "clapping" as a diagnostic tool, we can throw out all the "right" solutions immediately, and focus on solutions that perhaps have a prayer of being slightly useful. hence the idea for moving around the room while a helper claps. the perceived reflections should change as he moves around the room. as he gets closer to a problem wall some reflection ought to be time shifted earlier and perceived as direct sound. when that happens you might be able to sort out which reflections are coming from which wall.

same basic idea with the box idea. it ain't even close to "right" but maybe it would change the sound enough to isolate some bad walls.

so no, these ideas aren't "right" but I'd rather give a fellow cash strapped sound tech some marginal options than no options at all (or demad he use the "right" solutions, because that just isn't going to happen.)

Best of luck to the OP.