Golden ratios and standing waves

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Based on the consensus that: enclosures dimensions should optimally be derived using golden ratios to reduce standing wave effects, I have a question...

My enclosure volume is approx. 98l which I'm building into 244x418x962mm enclosure.

Using W=v/f, I caluclate standing waves at:
1393 Hz
813 Hz
353 Hz

If a golden ratio is used to derive box size: 280x467x747mm, then standing waves exist at:
1214 Hz
728 Hz
455 Hz

The offending frequencies are very close and I see no reason to closely adhere to the golden ratios? ( or have I missed something :cannotbe: )

I was going to asymmetrically fill three walls with fiberglass to compensate.

Any comments?
 

Attachments

  • standing-waves.gif
    standing-waves.gif
    7.9 KB · Views: 829
X^2=X+1. 1/X=X-1 Mmmm....

My opinion is that the golden ratio is of better use for the external dimensions of the box so that it looks "right".

but some of the effects ascribed to it by Cardas and others just strike me as voodoo for impressionable audioholics.
Fibonacci numbers are really great but I don't think it applies universally. I'm surprised that the revered *Magic Power Cords* :rolleyes: are not in fact 1.6180339.... metres long. ;)
 
Just had a look at that spreadsheet. Looks like what I caluclated are actually second order harmonics, the fundamental freq's being half my calculations. How does each harmonic affect the sound - to what degree in amplitude?

At the moment, I think that all rectangular enclosures suffer more or less equally (irrespective of whether the dimensions are golden ratiofied)- and that the important thing is to make sure each dimension is (assuming all parallel walls) different - to avoid adding to another standing wave. Agree?

Edit:
:bulb: By looking at the harmonics (occuring at each doubling of the fundamental) I've just realised that you need to avoid enclosure dimensions that are multiples of others, as the harmonics will re-inforce each other. I think I may have just justified the reasoning for golden ratios :scratch:
 
Excuse my ignorance, I'm still in the early stages of learning this stuff, but I've been wondering for a while; Doesn't using non parallel sides in an enclosure pretty much deal completely with this problem? I'm sure it's just not that simple, or else you would see no rectangular boxes and yet they abound.

Chris
 
sfdoddsy said:
Sorry to be a cynic, but what consensus? I find the Golden Ratio/Fibonacci etc interesting as an example of coincidental math, but some of the effects ascribed to it by Cardas and others just strike me as voodoo for impressionable audioholics.


Steve, you've apparently had my house bugged.


Doesn't using non parallel sides in an enclosure pretty much deal completely with this problem?

No. It just means that the eigentones will not follow the same pattern as a rectangular box. A sphere doesn't have parallel sides, but it resonates quite nicely. Boxless is still the best (and can be pulled off easily, since you CAN box drivers which are reproducing wavelengths large compared to box dimensions without fear of standing waves). Second best is a box with adequate damping and designed with minimizing reflections back through the cone(s) in mind.
 
mcp, here's are the standing wave modes for my current design. I see issues with (for example) the 3rd harmonic (height) and the fundamental (width) re-inforcing each other. Are these the problems you were seeing?

It seems clear to me now that using the golden ratios in my design would have resulted in much less overlap of standing waves harmonics. I'm convined of this reasoning now, though I have no idea how much affect this has ;)

using non parallel sides in an enclosure pretty much deal completely with this problem?
My understanding of non-parallell walls is that you're creating/dispersing the standing waves over a wider frequency range, and in doing so the combined mode is of a lesser amplitude.
 

Attachments

  • standing-waves2.gif
    standing-waves2.gif
    6.3 KB · Views: 737
Vikash said:
My understanding of non-parallell walls is that you're creating/dispersing the standing waves over a wider frequency range, and in doing so the combined mode is of a lesser amplitude.

Not really... it's just that the standing wave is no longer a pure axial wave and is much more difficult (without complex software) to predict.
 
Vikash said:
At the moment, I think that all rectangular enclosures suffer more or less equally (irrespective of whether the dimensions are golden ratiofied)- and that the important thing is to make sure each dimension is (assuming all parallel walls) different - to avoid adding to another standing wave. Agree?

My understanding of golden ratio's (can there be more than one?) is to maximize the (acoustic) difference between each dimension. By sticking to a .66/1/1.66 ratio, you get decent response even without stuffing and dampening the enclosure. I seem to recall it would keep the amplitude of enclosure resonance within a 3db window....

Dan
 
there is _something_ to the golden ratio

at least when it comes to rooms.

Not sure that it means anything in speaker enclosures.

check out

http://www.stereophile.com/showarchives.cgi?171

and

http://www.enjoythemusic.com/tweaks/tweakspage2.htm - scroll down on this one.

FWIW, I did a major renovation and based the floor plan on a listening room that is very close to a golden cubiod. With no other changes, there was a noticeable improvement and I am quite happy that I did so.

Couldn't do an actual A/B - the previous listening room did not exist in the same form _grin_

Still, the room works very, very well. 9x14x23.5

Since the AES uses a golden cubiod for its standard listening room, I can't help but think there are benefits.

regards

Ken L
 
I'm all for room domensions that cut down on resonances. The problem I have is with magic abilities ascribed to the Golden Ratio by some of its more fanciful adherents. I'm all for a good conspiracy theory, but a little gnosis goes a long way.

Also, if you read to the bottom of the Stereophile article, there is a letter from the AES stating that their standard listening room (if they had one) differs considerably from the golden cuboid dimensions Mr. Cardas ascribes to it.

Cheers

Steve
 
I think that only half the story is being looked at in this discussion. The enclosure will have resonances and standing waves at different frequency values. I might even agree that the golden ratio dimensions will optimally spread the frequencies.

But what is missing in the discussion is the location of the driver and the port, in a bass reflex design, and how the selection of the location determines the excitation of each mode. Placement of the driver in the front baffle will determine which modes are strongly excited and which ones not at all. The only way to excite all of the modes in the box is to place the driver exactly in the corner of the three sides of a sealed box.

Lets say we place the driver exactly in the middle of the front baffle. Then the standing waves that exist between the two sides and between the top and bottom will not all be excited. The first half sine waves will have a minimum excitation (driver at the point of maximum velocity) and the first sine full wave will have a maximum excitation (driver at the point of zero velocity). Every other mode will be suppressed or excited in the same manner. The front to back modes will all be excited (driver at the location of zero velocity).

Now if you add a port into the mix (point of zero pressure) then some modes will be supported and others suppressed.

Standing waves in enclosures are very interesting. If you want to try my MathCad "Ported Box" worksheet, you can experiment with the location of the driver and the port to see how they impact the vertical standing waves in the enclosure.

So my conclusion ..... I don't pay any attention to the "golden ratio" when setting the dimensions of my enclosures. I try and account for some of the standing wave resonances and use them to my advantage.

For room mode excitation the same discussion holds. Place you speaker in the corner to get strong excitation of the fundamental room modes. Move the speaker out into the room to get a more balanced response. Depending on the position of the speaker, some of the room modes will be strongly excited, some minimally excited, and some modes missed completely.
 
duhhh, wellll

darn it steve! Do you mean to say you read all of that stuff? letters, etc?

I did read it after you pointed it out _grin_

OK so it isn't "AES standard listening room".

It still is working well for me. FWIW worth, the method for positioning speakers in the same article is working well, and is working better than the several other methods I have tried.

I don't believe there is any _magic_ here, I think there is only math. It does seem plausible that it will work as stated and short of buying software for the specific purpose appears to be as good an approach as any.

I'll have you know that I'm the seventh son, of a seventh son, of a seventh son _really big grin_.

So you really ought to take heed of of what I say _ even bigger grin_

regards

Ken L
 
frugal-phile™
Joined 2001
Paid Member
Vikash said:
I was going to asymmetrically fill three walls with fiberglass to compensate.

At frequencies where the waves stand the fiberglass will be pretty much invisible.

I think I may have just justified the reasoning for golden ratios

The Golden Ratio gets far too much press -- almost any set of irrational ratios will do. The box dimensions will also have an effect on the actual panel resonances (complicated by bracing). As Martin notes the subject is more complex than it may seem on the face of it, and why not take advantage of some of the resonances :)


Christopher said:
I'm sure it's just not that simple, or else you would see no rectangular boxes and yet they abound.

You don't see a lot of non-rectangular boxes because they are more expensive to make, and to make shipping boxes for.

dave
 
At frequencies where the waves stand the fiberglass will be pretty much invisible.
I think I need to understand this in a simple way:

a) By damping (reducing enclosure vibrations), resonating frequencies are effectively raised, and these higher frequencies can then be tamed somewhat by...stuffing? (ie. fiberglass/dacron etc.)

b) But standing waves are a function of enclosure dimensions, and not directly related to (a).

I now understand (please confirm) that these are seperate things to consider in enclosure design.

So first I can brace to reduce large panel vibrations into smaller area, higher frequency vibrations. This can be aided by damping materials (roofing felt, dynamat?). I can then absorb these higher frequencies with stuffing material.

What can be done about standing waves (apart from minimise the effects via well chosen enclosure dimensions)? I can't alter the dimensions without losing precious enclosure volume (and I'm already 500+ litres below the optimal ;) - (so I guess a little more won't hurt - but is it worth it?)).
 
frugal-phile™
Joined 2001
Paid Member
Vikash said:
I now understand (please confirm) that these are seperate things to consider in enclosure design.

Yes, one is the box vibrating, and the other is the air in the box vibrating.

So first I can brace to reduce large panel vibrations into smaller area, higher frequency vibrations. This can be aided by damping materials (roofing felt, dynamat?). I can then absorb these higher frequencies with stuffing material.

roofing felt or dynamat will pull the resonant frequencies of the panels back down so to some extent negates what you are trying to do with the bracing. If you push the frequncies that the panel vibrates at and lower the Q of those resonances with angled braces, then hopefully they don't get excited. The major contributor to exciting the panels to vibrate is the motion of the woofer. One of the most effective ways of minimizing the energy the woofers put into the box is to use 2 tightly coupled woofers in a push-push configuration (a side-effect is that your speaker should image better vrs the MTM config you now have)

What can be done about standing waves

The amount of bracing you have will help somewhat, damping helps and is most effective in the middle of the box.

Another is turn the standing wave into an advantage by doing a quarter-wave design (ie TL).

dave
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.