how to figure room ratios in a non-rectangular room?

Started by biggator on 21 March 2009. 3 replies.

Originally posted at johnlsayers.com, topic 11822.

Searched through the reference area and a few books and couldn't find the calculations anywhere so I figure I'd try here. I'm tweaking my design - and was curious (due to a question from Soundman2020) how close it was to an 'ideal' room ratio. The room is not rectangular (it's a typical Sayers-style control room design.. as he drew the original plans for me) - and has a ceiling that starts at approximately 6' at the front, rising to about 9' in the center of the room. Sloppy Sketchup drawings can be found here.. viewtopic.php?f=1&t=11741 Pictures here:
Image not preserved: working_temp1 (Small).png
Image not preserved: working_temp2 (Small).png
So how does one go about calculating room ratios in a room with so many angles? Thanks!
So how does one go about calculating room ratios in a room with so many angles?
A couple of recent threads that you might find interesting, on this very subject: viewtopic.php?f=1&t=11719&p=85468&hilit=ratio+non+rectangular#p85468 viewtopic.php?f=1&t=11772&p=85791&hilit=ratio+non+rectangular#p85791 Summary: The way I see it, (and I haven't noticed any acoustics expert argue against my non-expert reasoning yet), you can use a "best-fit" rectangle that approximates your room's real dimensions, and calculate your ratio based on that. Provided that your room is reasonably rectangular (not round, or triangular, or hexagonal, or something like that!), then a "best fit" rectangle will approximate reasonably. It won't be totally accurate, but will give you a reasonable ball-park estimate, valid up to around 200 Hz, roughly, provided that your walls/ceiling are not splayed at huge angles. As long as the "bet fit" rectangle stays within about half a meter of the real dimensions, you'll get a reasonable rough idea of your ratio. - Stuart -
One of the problems I see with best fit, even though I may have used that suggestion before, is that it is inaccurate. The extensive work that went into ratios and modes and all those catch words we throw around are based on six hard boundaries rectangular in shape. Simple math relations based on length of wall, length of wave, ray, trace, et. al. and the interaction within these set boundaries. So when the best fit is applied it is, based on the ideas that went into the original work, and in my opinion, no longer of any value. Everything has changed. Now I certainly do not know what the correct answer is but I would suggest taking the best fit with a grain of salt. I suppose with the 1000 sg. ft. and smaller rooms that is becoming typical, a best fit might apply. To paraphrase Philip Newell, "the room modes become the room acoustics." Even when the ratios are being recommended by celebrated authors like Everest, Newell, etc. they still are aimed at the hard boundaries of rectangular rooms. Knocking an individual boundary out of plane by 6 degrees is often suggested because it was found that up to and including 5 degree, had no effect and higher numbers produced a negative effect. I would wonder how a best fit method would have worked for Hans Jenny's study of wave phenomena and where would Mandelbrots' fractal math be if best fit were applied? If I had a choice I would start with a rectangular layout and work splaying, trapping, front panels, even speaker baffles, into that. Rather then the other way around.
Bigator's room is, in fact, very close to rectangular anyway, except for that slight kink around the soundlock.