Question for Eric: ratios, modes, and parallel walls... | John Sayers’ Recording Studio Design Forum archive

Eric, I hope you see this and feel inclined to respond, since this is an issue that has been bugging me for a while now. It's all about room modes / standing waves / room ratios. Basically, my question is this: How far off parallel (perfect rectangle) do the walls of a room have to be, before standing waves no longer "stand", such that the modes and the ratios can no loner be calculated using all those simple "room mode calculators" and "room ratio calculators" that one can find all over the internet? I mean, if a rectangular room that is 7m wide and 10m long (for example) has a wall that is out of parallel by just 1mm, then I would imagine that this won't make any difference at all, and one could still safely use the mode calculators and room ratio rules with no hesitation. But if that same room has walls that are 3 m off parallel (ie, 7m wide at one end, 4m at the other, but still 10m long), then I would imagine that the actual modes would be way different from those predicted by calculators, and you could no longer even talk about "ratios" at all. ( As far as I know, "ratios" only apply to parallel rooms. ) So my question is: How far off parallel do walls have to be before it makes a big enough difference that waves don0t stand and simple rectangular ratios don't apply? Can you still use a mode calculator to judge (roughly) the design for a room with walls splayed 6° and the ceiling at 12°? Or 10°? or 20°? Or even 1°? That's what is bugging me. Or maybe another way of asking that would be: at what splay angle for side walls / ceiling is it no longer valid to consider the room as being a rectangle, and therefore no longer subject to ratio / mode calculations for rectangular rooms? And the bigger question is: why? My "feeling" is that this must be related to wavelength somehow, and that perhaps it becomes an issue when the size of the difference in parallelism becomes large relative to the wavelength you are worried about, so even a slight difference affects high frequencies, but you need a fairly large difference before it becomes an issue for low frequencies. Does that make sense, or am I way off base there? I've seen many questions along these lines on this and other forums, but I've never seen a satisfactory answer from a real acoustic expert. This whole issue has gotten me intrigued, but I can't find any discussions on it! - Stuart -

Soundman2020 wrote:
Eric, I hope you see this and feel inclined to respond, since this is an issue that has been bugging me for a while now. It's all about room modes / standing waves / room ratios. Basically, my question is this: How far off parallel (perfect rectangle) do the walls of a room have to be, before standing waves no longer "stand", such that the modes and the ratios can no loner be calculated using all those simple "room mode calculators" and "room ratio calculators" that one can find all over the internet? I mean, if a rectangular room that is 7m wide and 10m long (for example) has a wall that is out of parallel by just 1mm, then I would imagine that this won't make any difference at all, and one could still safely use the mode calculators and room ratio rules with no hesitation. But if that same room has walls that are 3 m off parallel (ie, 7m wide at one end, 4m at the other, but still 10m long), then I would imagine that the actual modes would be way different from those predicted by calculators, and you could no longer even talk about "ratios" at all. ( As far as I know, "ratios" only apply to parallel rooms. ) So my question is: How far off parallel do walls have to be before it makes a big enough difference that waves don0t stand and simple rectangular ratios don't apply? Can you still use a mode calculator to judge (roughly) the design for a room with walls splayed 6° and the ceiling at 12°? Or 10°? or 20°? Or even 1°? That's what is bugging me. Or maybe another way of asking that would be: at what splay angle for side walls / ceiling is it no longer valid to consider the room as being a rectangle, and therefore no longer subject to ratio / mode calculations for rectangular rooms? And the bigger question is: why? My "feeling" is that this must be related to wavelength somehow, and that perhaps it becomes an issue when the size of the difference in parallelism becomes large relative to the wavelength you are worried about, so even a slight difference affects high frequencies, but you need a fairly large difference before it becomes an issue for low frequencies. Does that make sense, or am I way off base there? I've seen many questions along these lines on this and other forums, but I've never seen a satisfactory answer from a real acoustic expert. This whole issue has gotten me intrigued, but I can't find any discussions on it! - Stuart -

Stuart, first off - you seem to misunderstand the concept of splaying walls - by this I mean the reasons for doing so (which do not include defeating room modes). Take any room - make it rectangular is shape - and it is reasonably easy to calculate the rooms modes - as you begin to change the room's shape geometrically you make it more difficult to predict the modal activity..... you don't make it go away - you just sort of spread things out to some degree or another....... The reason for splaying walls and ceilings is related to high mid and high frequencies - the ranges where the frequencies become more "directional" in nature (for want of a better term). Splaying surfaces will help to counteract things like flutter echo, some early reflections, comb filtering, but it will not stop the buildup of low frequency zones in the form of peaks and valley, although it might just lessen their intensity if planned and executed properly. Again - extremely complicated. For this purpose a standard room calculator (designed for rectangular spaces) is pretty much useless....... The rule of thumb for the splaying of walls is a minimum total of 12 degrees between 2 surfaces....... so in a control room (where you want symmetry) 6 and 6 would get you there....... (that is meaning 6 degrees splay on your left and an equal 6 degrees on your right. Now - as far as your question regading at what point does this take place? Technically - any variation of distance will affect the room immediately - although the vaiance might be so little that the human ear might not be able to discern the difference - but it would be there nonetheless. It would have to be - but there are so many other things that affect it as well - so that even in a perfectly square room you might find some anomolies that you can't explain. Things like temperature and humidity - these can change the measurements in a room considerably...... with all other things being equal. Which is why part of the standards published (Picture ISO - ASTM) all have requirements for the measuring of Temp and humidity as a part of the work required in reverb rooms......... it isn't enough to just measure the reverb times........ the exact same signal, for the exact same duration, for the exact same length in time can produce different results just by allowing the temperature range to change - or the humidity....... Even in a perfect room (from a dimensional point of view) you will have different results than the results indicated by the exact model - this due to the fact that the models calculate based on borders which are infinite in mass - and thus no sound escapes them........ but we cannot build those same rooms......... In reality - it really doesn't make sense to place a whole lot of faith in these calculators other than getting your head around the basic degree of anomolies you expect in the end.......... Rod