Jump to content

How is the "Golden Mean" calculated in odd-shaped rooms?


JBryan

Recommended Posts

I can't be alone on this one but I haven't seen a lot of information regarding odd-shaped rooms. My basement is an L-shape - my listening area is approx. 28' long and 14' wide on with a 12' x 14' extension to complete the "L". The ceiling varies from about 7'3" to 7'10" with ductwork running everywhere and all is covered with acoustic tile. To make calculations more difficult, there is a 30" wide stairway behind the listening position leaving a bi-sected wall to contend with.

The Khorns are placed on the long wall 12' apart and 12' from the listening position. To facilitate positioning, I have built false corners. The FC enabled me to move the speakers around quite a bit and I positioned them by ear to get the best imaging and soundstage.

I keep trying to use the Golden Mean (1.618 ratio or better) as mentioned quite a bit on this forum and in Dope from Hope but as you can imagine, things get complicated when I start compensating for the variables involved. Is there anyone out there who has come up with a definitive way to calculate the dimensions in odd-shaped rooms? I'm in the process of putting up drywall and this is the opportunity to make any necessary adjustments to room size and treatments. Thanks for your help -Bryan

Link to comment
Share on other sites

I have a thought on a work around - you can use one of the free speaker box programs to find out what the golden ratio would look like for your room. WINISD can take two dimensions (either length, width or depth) and give you the third dimension. You may have to translate the calculations to the greater scale (inches to feet, centimeters to meters). In this way you could set any two dimensions and play from there.

Link to comment
Share on other sites

Thanks Henry,

I've looked at various programs but I keep confusing myself with the dimensions. My room is 7'8" x 28' x 30' x 12' x 14' x 16' x 14'. I usually put in 7.75' x 28' x 30' and then subtract 7.5' x 12' x 14' but that doesn't take into consideration all the other variables such as the stairs. I was hoping someone around here has a program or the knowledge to guide me through the equation and determine the best dimensions for my set up.

Thanks again -Bryan

Link to comment
Share on other sites

----------------

On 4/20/2004 10:23:50 AM JBryan wrote:

I can't be alone on this one but I haven't seen a lot of information regarding odd-shaped rooms. My basement is an L-shape - my listening area is approx. 28' long and 14' wide on with a 12' x 14' extension to complete the "L". The ceiling varies from about 7'3" to 7'10" with ductwork running everywhere and all is covered with acoustic tile. To make calculations more difficult, there is a 30" wide stairway behind the listening position leaving a bi-sected wall to contend with.

The Khorns are placed on the long wall 12' apart and 12' from the listening position. To facilitate positioning, I have built false corners. The FC enabled me to move the speakers around quite a bit and I positioned them by ear to get the best imaging and soundstage.

I keep trying to use the Golden Mean (1.618 ratio or better) as mentioned quite a bit on this forum and in Dope from Hope but as you can imagine, things get complicated when I start compensating for the variables involved. Is there anyone out there who has come up with a definitive way to calculate the dimensions in odd-shaped rooms? I'm in the process of putting up drywall and this is the opportunity to make any necessary adjustments to room size and treatments. Thanks for your help -Bryan

----------------

The rules do not apply. Once you depart from a rectangular configuration calculating the room modes becomes very complex. There are some software packages that take into consideration things like a simple sloping ceiling. Other software can help you evaluate your existing conditions. Most of this is usually confined to professional acoustician's applications. You might want to check out the following:

http://www.etfacoustic.com/contactus.html free demo copy ETF software

http://www.asc-studio-acoustics.com/articles/db1191.htm for overview basics

http://www.partsexpress.com/index.cfm for Prosonus CD 'Studio Reference Disc' which includes a MTF test developed by ASC (above link) for performing a 'listening test' with headphone vs speaker comparisons, as well as many other tests.

If you have access to a DAT machine & mastering software such as Sound Forge you can record the tests to the tape from a mic, download it to your computer & use the real-time analyzer in Sound Forge to see whats going on.

Link to comment
Share on other sites

I would suggest you can check out CARA (Computer Aided Room Acoustics) which was reviewed by Secrets of Home Theatre and HiFi . It is an acoustic software which models and optimizes speaker placement in odd shaped rooms... as it provides quite a bit of flexibility in defining the room. I'm not sure if/how it would treat a staircase though.

Good Luck...

Rob

PS: Artto... have you had a chance to try ETF5? It's a fantastic alternative to a hardware RTA giving you detailed results which you can manipulate it's output visualisation. I've been wanting to post a review / preliminary results from my living room but haven't gotten around to it. I've had a couple of snags but I plan in settling them in the near future. I seem to have quite a bit of RFI noise in my home, and I've been using it on my desktop with an unbalanced 30' interconnect as I haven't been able to get it to work on my portable. Unfortunately at 30' picks up some RFI/EMI and I don't fully trust the software's "optimized noise reduction" test tone.

Link to comment
Share on other sites

Hey thanks Rob! CARA is the one I was thinking of with the simple ceiling slope capability. Seems like they have recently upgraded the software substantially to include more irregularities in the room.

Actually, I'm running ETF, MTF & various sine/white noise/pink noise pulse/sweeps as I write this. I recently made some acoustical changes (hopefully improvements) and wanted to see how the room measures compared to what I hear (I hear major improvement in the low end). On a preliminary run through (I'm out of the room during the actual tests/data recording) the meters on the DAT deck were surprisingly 'flat' at various sinewave frequencies from 20Hz up. Seems to be a dip major around 62Hz (which I think I can correct), but other than that I was really impressed with how flat the system was measuring. I won't know everything for sure until I download the tape to hard disk & run it through Sound Forge. But the preliminary results just watching the meter level sure put a big grin on my face. And my hearing isn't as bad as I thought it was after decades of playing in rock & blues bands (LOL). 12.5KHz was still audible! As much as I hate wearing them, I guess those ear plugs did their job!

Link to comment
Share on other sites

Artto,

After seeing the post on ETF I downloaded it and would like to play w/it some but i'm really not sure where to start. What process do you go through to determine how flat the speaker response is? A play by play would be nice if you could spare the time...not too much, just a "Room Acoustics measurement and Analysis with nothing more than a sound card and a free program" for dummies guide.

Most importantly...what type of mic do you use that has a good range and flat response?

Thanks very much

Link to comment
Share on other sites

First of all let me say that Ive let basic acoustic principals be my guide in getting things set up. Second, there is (IMHO) no one measurement test that tells it all. And third, after youve got things done reasonably well by ear, then I run the tests to see if they confirm what I hear. And to find any remaining anomalies.

The other thing to remember is that you are NOT measuring how flat the speaker response is. You are measuring the response of the room/speaker(s)/listening position interaction with each other. This involves not only frequency response, but things like reverberation time relative to frequency, room articulation, etc.

My procedures are probably going to be a little different than you might be doing, primarily because Im not going to drag my Dell workstation down to the music room. And since my room is already built, Im not concerned with modeling at this point. Simply measuring room/system performance. And I already have some other software, so Im not using the ETF Acoustics software. But maybe this can give you an idea of what to do & look for.

I set up an Oktava MC012 (MK012 outside US) condenser mic with the omni directional capsule http://www.oktava.net/ at the primary listening position. This mic has very flat frequency response with the omni capsule down to 30Hz. I run this mic to a Mackie VLZ1202 mixer that feeds the Sony DAT (digital audio tape) deck. I download the DAT tape via the regular RCA analog outputs to the computer through its normal audio-in on the sound card. I then record the DAT tape to hard disk.

Today I was using the Prosonus Studio Reference CD which probably has many (if not more) of the tests contained in the ETF software.

First start by calibrating the output level of the speakers with a 1KHz tone. I use a Radio Shack SPL meter. I put it right up against the grill cloth of the midrange horn. Make sure the output is balanced (ie: if you have a derived center channel speaker its normally supposed to be set -3dB lower than the side speakers, & make sure the side speakers are the same output).

Then get a reference sound level set (be careful, start at a low level, some tests may tax your system) on the tape deck or your computer if youre going in direct (by-passing the DAT part).

The first tests I ran were sine wave tones. I dont particularly like using these because a very small movement in location or direction (mic or ears) can have very major changes on the results or what you hear. This is why I leave the room when I run these tests. At some frequencies simply raising your arm or leaning one direction or the other can affect the results. The test frequencies I used are 15.5Hz, 31, 50, 62.5, 100, 125 & so on up to 16KHz.

I also used several sine wave sweeps, sine wave bursts, time/energy/frequency sweeps, white noise & pink noise, both continuous & bursts.

Sound Forge has a built in real-time analyzer. So I can accurately see the results, not only of frequency response, but see the decay time at certain frequencies, how various tones interact with other frequencies & so on.

I know this may seem complicated. But there are really several aspects to all of this. And thats why I keep repeating BROADBAND-ABSORPTION&DIFFUSION-BROADBAND. Its not just about frequency response, or any ONE of the above.

Link to comment
Share on other sites

Thanks for the response, I played w/it with a crappy mic that came with my audigy 2 sound card and did a few frequency sweeps. Sound Forge showed me that it jumps all over the place, which i actually noticed during one of the 15-280 sweeps I did, after 40hz or so its almost like a roller coaster with peaks and vallys. I'm not sure what to do about it since this is my theater room and my bedroom. Oh well, i'm leaving this apartment in 2 weeks and moving into my stepdads old house, maybe then I'll be able to tweak/tune the room.

Link to comment
Share on other sites

Yeah, I wonder how many rectangular rooms actually exist in fact, much less those with a Golden Ratio. There is usually at least a doorway which breaks up the solid surfaces.

I've got a one bedroom condo. The main room has two fairly good corners with a big window in what I consider the front. The rear has a pass through to the kitchen and a hall. Even closing the doors off the hall, there is some question about the effect of the wall which defines the kitchen.

The good news is that there is nothing I can do about any of this. So I don't worry.

Gil

Link to comment
Share on other sites

  • 2 weeks later...

I am building a new home and will have an area in my large new basement that I am planning to build a dedicated theater completely closed off with a single (maybe double french doors at the rear entrance) door.

The room dimensions unfinished are as follows:

Length: 27 feet 5 inches

Width: 13 feet 9 inches

Height: finished height will be approximately 8 feet for the entire room

Once finished I can basically subtract 8 inches from each dimension to account for each framed/sheetrocked wall, leaving me with 26' 9" and 13' 1".

How does this measure up using the golden rule? Is this a good room size/dimesions relative from an acoustic perspective? What is the rule and what would you do giving these dimensions to work with.

I appreciate any feedback, advice or help you guys offer.

Thanks. Brett

Link to comment
Share on other sites

There are several methods for determining ideal room proportions. The Golden Mean is just one of them. The proportions are 1:1.26:1.59 (height:width:length). It affects primarily the lower frequencies as the room size gets smaller.

Your room is rather narrow compared to its length & available ceiling height. This means you will have some substantial bumpiness (both peaks & dips) in the bass response. In some locations in the room certain frequencies may seem excessive while in other locations these same frequencies may seem to almost disappear.

The Golden Mean (or other methods) are an objective. It doesnt necessarily mean that the room cant be made to sound good

Link to comment
Share on other sites

Guest
This topic is now closed to further replies.
×
×
  • Create New...