bhendrix Posted May 16, 2009 Share Posted May 16, 2009 We have an interesting challenge in our church sanctuary with speech intelligibility of the singers. The room seats 1200 and is layed out in a flattened diamond shape with the chancel in one of the corners of the diamond. The room is wider than it is deep. Heavy black drapes are hung on the side walls from about 10' down. The walls above are brick. The back wall is mostly glass. Aisles are carpeted and pews are fabric upholstered. The ceiling is vaulted wood about 45' high at the peak. We are using a KP-682 with 2 Ki-362's on the on the left side and a KP-682 with 2 Ki-362s on the righ side of the stage with a center cluster flown overhead in center. The cluster is about 40' in the air and the two floor stacks are about 40' apart. SM58 mics. Intelligibility is good when only one person is speaking. When the band is playing, the music sounds good, but you can't understand the words being sung. Any thoughts on how to begin improving intelligibility? Quote Link to comment Share on other sites More sharing options...
djk Posted May 16, 2009 Share Posted May 16, 2009 How does it sound with the center cluster doing vocals only, and no vocals in the LR mix? Quote Link to comment Share on other sites More sharing options...
drewby2 Posted May 17, 2009 Share Posted May 17, 2009 I was very taken with the design of the Pompideau Center in Paris a few decades ago. One of the professional journals reported on their computer controlled variable acoustic sound absorbtive/reflective system. This idea has been cooking in my brain for at least 30 years. The computer controlled servo motors just made the auditorium easy to tune. One could manually turn acoustic panels for a less than high tech and less expensive system. As I understand the Live End Dead End acoustic design (LEDE) making everything dead is not a good idea. This design technique deadens one side of every parallel surface. The best acoustics are just live enough and just dead enough to sound good. It is reverberation and sound slap that create the multipath nodes that ruin articulation and promote articulation loss. I am less a good theoritician and more an experimentor. That is why the variable acoustic idea appeals to me. My own church had similar issues, when I sang, I could hear my voice slapping back from the back wall. I always wanted to try this out. applummer@sbcglobal.net Quote Link to comment Share on other sites More sharing options...
colterphoto1 Posted May 20, 2009 Share Posted May 20, 2009 Bill, instead of looking to acoustics as the source of the problem, could it be in the MIX? Are there too many instruments clogging the frequency bands occupied by the vocals? Tweak some mid out of the guitars and keyboards and see how you fare for starters. If a single mic sounds intelligible on your system, it may very well not be an acoustical problem. Quote Link to comment Share on other sites More sharing options...
drewby2 Posted May 21, 2009 Share Posted May 21, 2009 The process that colterphoto1 is talking about is "masking" if two sources have the same sets of harmonics then only the louder one is heard clearly. This could be the reason for your problem. The answer is changing the tambour(sp?) of the source by equilizing them so that everything in the mix has a unique harmonic signature. But if the problem is articulation loss (the consonants in speach) then I think the solution is solving multi path issues. If sound is bouncing off the parallel walls there are time and phase cancellation issues. Quote Link to comment Share on other sites More sharing options...
bigrfish Posted June 2, 2009 Share Posted June 2, 2009 Bill, As I can understand from some reading I did on the subject, you are looking at a "booger" of a situation. Nobody that wanted to have good sound in a room would ever build one like that, yet that is exactly what happens, then they want a super result, where you have intelligibility of speakers, singers, band, singers and band together, and one part a lot of people forget, is that the worshippers have a better experience and get more "plugged in" if they can hear themselves and one another, so you've got the most difficult job there is. Before you hire the acoustic consultant from out of town, there might be something you can try....Just off the top of my head, here are some things that won't cost much money and might help: 1. Try to get the band to play as quietly as they will, in a non-frenzied, controlled, recording-studio-like manner. 2. Don't run any instruments other than drums through the board....unless there is no choice. Want to reinforce sound, not make sound. 3. If you are processing the house with delay or reverb, quit. Do use some compression on the praise team vocals. 4. Lower the overall house level. I have been in some churches where it was more brutal in there than an Allman Brothers concert. That's nt worship, it's spectacle. 5. Consider signal chain integrity and look at the quality and engineering of your mics. I know everybody swears by SM 58's but look at something else like a Beyer Dynamic with a supercardioid pattern...you want something with relative fidelity that will not pick a lot of noise in the area...directional...that will reject both adjacent sound/noise/choir/band and also room acoustics. 6. Like one of the other posters said, selective equalization might be the key....that's when you call in the guys with the spectrum analyzers who do the room treatments, and I am suspecting you are going to have to do it because of the bizarre room characteristics and larger than life needs...but you might get it a little better in the meantime... 7. You probably already know this, but the MUSICIANS need to know it...they just need to play in tune on time and not worry about what it sounds like in the room...no volume creep, tone changes, etc. That is the sound guy's job. It is totally impossible for a player to know either what he sounds like in the mix, or what he NEEDS to sound like in the mix. 8. If certain instrument amps are problematic because of zeal run amok, put them suckers in a little isolation booth just for an amplifier and mic that...also, put the drummer in an isolation booth, or try to deflect some of the drums off of the hard surfaces behind the drummer... 9. Would it hurt or help to drop that cluster? Would the altar guild be up in arms?? God Bless and good luck with it...I feel your pain, and I ain't Bill Clinton... Chuck Quote Link to comment Share on other sites More sharing options...
bigrfish Posted June 2, 2009 Share Posted June 2, 2009 That's a fine point. some of the old head Muscle Shoals engineers talk a lot about managing bandwidth. Same concept. Quote Link to comment Share on other sites More sharing options...
colterphoto1 Posted June 2, 2009 Share Posted June 2, 2009 Yes, kill any delays or reverbs in the mix, compress gently (just enough to shave off the loudest parts), and tweak eq away from instruments on songs with vocals. How about mic technique? We're not on the verge of feedback because the numerous 'vocalists' don't know to step up to the mic, are we? Since you say that a single vocalist speaking is intelligible, it might be an issue with multiple mics or the mix of mics/band. If multiple mics are left open but unused they might be adding mush to the sound by amplifying room and band noise (which might be out of tempo and phase by the time it reaches the mics) in this case a good soundman killing unused mics (use the MUTE buttons so you can leave the master faders alone), or get gates for the multiple mics. Mic selection might be an issue- SM58's aren't much good for choral mics. 849 condensers or something similar might be better here. What board are you using Bill? Can you subgroup secondary mics and put a compressor on that whole mix? Quote Link to comment Share on other sites More sharing options...
bigrfish Posted June 2, 2009 Share Posted June 2, 2009 I saw a DIY project once where there was a floormat/switch thing that when the vocalist stepped up to the mic, it was turned on, and when he stepped back it was turned off. Don't know how they controlled transients, probably some kind of filtering network...but interesting. There's more to do than you can imagine, and the hardest part is to get people to do things subtly different than they have been "doing it before". good thread, guys... Quote Link to comment Share on other sites More sharing options...
Professor Thump Posted June 2, 2009 Share Posted June 2, 2009 Bill, It is hard to answer this question with more specifics on the setup. I would start by having someone layout the design in a CAD model so you could predict the ALCONS (RT60) from the computer. ACOUSTACAD, CADP, or EASE can do this for you. I believe Roy has a copy. Until you compute the Array angles and absorbtion coefficients of the walls you will be running around in circles. Once you have it modeled you can go and adjust angles on the array and even placement of the array to optimize the setup for intelligibility. You can also adjust the sabines to the wall to see what treatment would give you. This could also be as simple as mic placement. Good source localization and good speaker placement are starting points. Quote Link to comment Share on other sites More sharing options...
bigrfish Posted June 7, 2009 Share Posted June 7, 2009 Bill, Have you tried a BBE Sonic Maximizer in the signal chain? One thing I noticed about mine is that the phase correction or time-of-arrival modification makes it so that I can pretty much hear everything that is in the mix with a lot of airiness and freedom from effort compared to listening with it off or bypassed. I do not think I would ever want to go without it...subjectively on a nominal scale I would say I get about a 20-25% increase in the ability to hear everything. That seems to be its' main function. Chuck Quote Link to comment Share on other sites More sharing options...
bhendrix Posted June 7, 2009 Author Share Posted June 7, 2009 Hi Chuck, The first thing we're going to work on is the rear window reflection in the sanctuary. When we listen over headphones, the vocals are clear and understandable, so this seems to be a room issue. The next step will be to use the KP-600s by themselves without the overhead center cluster. I still have some work to do to get the 600's up and running. Waiting for amps and EQ settings. Quote Link to comment Share on other sites More sharing options...
bigrfish Posted June 7, 2009 Share Posted June 7, 2009 Bill, The KP 600's should be able to give you sufficient air movement without having to push hard enough to make the people perceive a "loud" PA system, and the parishoners/worshippers should be able to hear both themselves and the output of the house system with no problems. I know it is going to be worth it, all this concern and care and work. Godspeed as you get it done. Sounds like you are on the right track. When you get it "tweaked" and running as you like it, I would love to visit and "fellowship" with y'all!! Chuck Quote Link to comment Share on other sites More sharing options...
colterphoto1 Posted June 8, 2009 Share Posted June 8, 2009 Bill, with the big stacks you'll be blowing away the parishoners in front while those in the back get no volume. If the sanctuary has any height at all, a hung system should give more even dispersion throughout the room. I don't think the KP600's are your answer. Plus I think they'd look pretty oppressive in there. Quote Link to comment Share on other sites More sharing options...
DrWho Posted June 8, 2009 Share Posted June 8, 2009 How does it sound with the center cluster doing vocals only, and no vocals in the LR mix? Have you done this yet? If you've got REW handy you could capture an impulse of your room and then I can convolve it with some music on my end to hear what you're hearing....I suppose we could look at the measurements too. An RT60 would probably tell you quite a bit in this scenario. The only reason I want to hear it is because it's a lot easier to identify things rather than trying to sift through the noise of a measurement. Have you verified that when the band is playing that the only thing you hear in the mic channel is the person's voice? And not bleedover from other instruments on stage? What kind of processing do you have on FOH? Any feedback suprressors or other EQ trying to notch out feedback? With an SM58 you should be boosting the highs a bit, and cutting the lower mids (around 250-500Hz-ish). The low cut should definitely be engaged, but depending on the subwoofer implementation you'll probably need to cut the LF shelf a notch too. If you put a person on stage without a mic and he shouts back at you, can you understand what he's saying? Quote Link to comment Share on other sites More sharing options...
colterphoto1 Posted June 8, 2009 Share Posted June 8, 2009 " If you put a person on stage without a mic and he shouts back at you, can you understand what he's saying?" oooo, that's a good one. Take the whole PA out of the mix and listen to see if its the room. Of course it doesn't take speaker placement into account, not unless you hang the 'shouter' up where the cluster is! LOL Quote Link to comment Share on other sites More sharing options...
Professor Thump Posted June 8, 2009 Share Posted June 8, 2009 Hi Chuck, The first thing we're going to work on is the rear window reflection in the sanctuary. When we listen over headphones, the vocals are clear and understandable, so this seems to be a room issue. The next step will be to use the KP-600s by themselves without the overhead center cluster. I still have some work to do to get the 600's up and running. Waiting for amps and EQ settings. Bill, I suspect your problems are related to the natural reverb of the room and not the signal processing. Especially if the master mix through headphones is acceptable. Major suspects are the brick and glass which both have low absorbtion coefficients. Parallel walls are terrible on top of this. Your flutter echo in the ceiling area above the curtains is probably ringing. A little wall treatment will improve the situation greatly. You can think of foam as an open window for sound. Once it hits the foam it does not reflect. For the sound mix you best speaker cluster for a reverb free mix is probably the center cluster hanging. This is because it is more optimized for angles of the room. If you were to trace the rays of acoustics the stage speaker being parallel to the floor would radiate to the back wall and then reflect to the front parallel wall. The hanging speaker might project to the back wall but at a skewed angle which would then bounce to the floor and congregation. People absorb sound well, thus it will be better in a full house. May I make this suggestion? If possible mix the vocals to mono and send them through the center cluster only. I am flying blind because I don't know the SPL mapping of each speaker cluster on the floor but the rear of the room would probably sound better with mono center only. You can then mix the instrument to the L/R stacks. Secondarily loud acoustic instruments could be contained in a plexiglass iso chamber. (drums are the main culpret) This would help to avoid fader creep from sound spilling over from the stage. Quote Link to comment Share on other sites More sharing options...
djk Posted June 8, 2009 Share Posted June 8, 2009 "May I make this suggestion? If possible mix the vocals to mono and send them through the center cluster only." I'm glad to see someone finally repeated what I said in the very first reply on this thread. Quote Link to comment Share on other sites More sharing options...
drewby2 Posted June 9, 2009 Share Posted June 9, 2009 What an interesting idea. Like a home theater have a center channel for the dialog. This would eliminate phaseing and nodes in critical areas and reduce articulation loss. A human voice, unamplified comes from a single point. How can we expect our hearing apparatus to figure out what this "coming from everywhere" sound means? The multipath reverberations would increase the points of origen. This is a recipe for mud. . In my own small church (a few years ago) we put up a central cluster with three 60 dregree dispersion enclosures. After the original design, a sound person seperated all the enclosures of the cluster. This put the sweet spot of each speaker on the ailse. It increased multipath problems and decreased intelligibility. A central cluster is ideal for the intelligibility of speech. The intelligibility of the spoken word should be the first priority of a Christian house of worship. I prefer sound reinforcement to be transparent. I don't like to hear a sound system. I like to hear the voice and music. Quote Link to comment Share on other sites More sharing options...
colterphoto1 Posted June 9, 2009 Share Posted June 9, 2009 Or you could mix the vocals to mono and route them through the center cluster only. [] Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.