Building a non-rectangualer enclosure - What shape?

[peter_m]

Hello everyone,

I want to build a pair of Zaph's ER18 and wanted to push the project a little further... While getting the garage and my self messy, why not make it a unique project I can enjoy for many many years?

So instead of a rectangular enclosure, what about an enclosure that has the sides closing in, or a shape where the top is not parallelled to the ground, but rather sloped down toward the back? As I understand it, purpose of non-rectangular enclosures are to minimize the standing waves so the first questions is: Which shape minimizes the standing waves best? The second question is: How strictly do I have to adhere to this magical shape?


peter

[relder]

Do you have the Loudspeaker Design Cookbook book? It somewhere within has shapes of various enclosures and measurements (or simulations, can't remember which) of their effectiveness. I think both in terms of diffraction (outside shape) and internal waves (internal shape.) A sphere is the ultimate if I recall.

[peter_m]

Do you have the Loudspeaker Design Cookbook book?

Yikes, no I don't have the book. I was hoping to benefit from someone's experience. I know... I'm a freeloader.... I think I want something that is relatively easy to build into a floor-standing speaker... Maybe a sphere is a little bit of a stretch for my woodworking skills....

[Dennis H]

I don't think slightly non-parallel walls do much for standing waves. That's more a job for stuffing. So build whatever you like the looks of. I like the new PE enclosures but they would be a bit of a challenge to DIY.

http://www.partsexpress.com/pe/showdetl.cfm?&Partnumber=302-737

[Jed]

If you are using Zaph's design for the ER18, I wouldn't change the baffle layout because it will change the FR for the worse. As for standing wave management, just play around with different stuffing to get the tuning right.

[peter_m]

Ok, so you guys are all jealous of my wood working skills eh??? Just kidding! I have none and I like to play with tools till the nurse at the emergency rooms says... not you again!!!

But seriously now, are you all saying that rectangular shapes are just as good?

[augerpro]

I say build it and see. But-keep the baffle dimensions and layout the same, make sure no dimensions is an even multiple of any other, tell us how difficult it is to cut all those angled edges and clamping them up when you glue them!

[Jed]

But seriously now, are you all saying that rectangular shapes are just as good?

When the design was measured with a rectangular baffle, yes.

Of course there are better solutions out there. One is the Avalon faceted front that minimizes diffraction etc, but that would totally change the design and the crossover won't be optimum, given the fact that it was designed for a rectangular baffle.... etc.

[peter_m]

Well, the baffle will stay rectangular, I'm talking about whats behind the baffle...

[Andy_G]

deleted

[peter_m]

Assuming I can manage to keep the same baffle width, is this site any usefull?

http://vikash.info/audio/standing_wave_calc/index.asp

[gmed]

Why dont you do what I did? A lot of work but it does what youre asking for!

http://www.htguide.com/forum/showthread.php4?t=23805&page=1&pp=35

[peter_m]

Really nice.... The time you put into this really shows and I am sure it was worth it.

How did you come up with the shape?

[chrismercurio]

http://www.magico.net/?d=03_Products/02_Mini_II

It is a tribute, not a clone, to the Magico Mini.

I agree with others in the thread that if you want to build Zaph's design, build it as is or the crossover will not work correctly.

[tpremo55]

Pete,
GMED has a very nice clean construction there. I considered this on my curved-side Statements build but the 60" tall hight required 80 layers per cabinet. now I don't consider myself lazy, but that is a LOT of sawdust. For an alternative, take a look here:

http://picasaweb.google.com/tpremo55/Statements_Web_Album

As I recall, Zaphs design is much smaller, so this may not be a big issue. my only concern with this design is the potential for resonance in the sides. In the end, I find the cabinet really quite dead and do not find this conern a reality. In the end, I made only slightly less sawdust.

I'm looking to do a traditional stack laminate on another smaller project that I am working.

Here are some other great example sites I picked up while doing the same search a number of months ago:

Some interesting speaker designs:

http://members.optusnet.com.au/~gradds55/curved_loud.htm

http://www.seventh-veil.com/nonsuchconstruction.htm

http://www.hiendfi.com/modules.php?name=News&file=article&sid=361

http://www.hifi-notes.com/sonusfaberamatifabrikage-nl.htm

http://www.sound-e-motion.com/en/catalogue.shtml

http://www.wilson-benesch.com/curve/curve.html

And a couple basic woodbending references:

http://www.allwoodwork.com/article/woodwork/methods_of_bending_wood.htm

http://www.tai-workshop.com/english/tech-2(b)-e.html

Good luck! ...and keep us posted on your project!

Todd

[Dennis H]

Assuming I can manage to keep the same baffle width, is this site any usefull?

http://vikash.info/audio/standing_wave_calc/index.asp

Not really. You need to keep the height and width of the baffle as well as the internal volume the same as Zaph designed them. That means the only dimension of his design that can change is the depth to keep the volume the same if you change the shape of the rear part of the box.

I think you are creating a problem (standing waves) where none exists. It's really a non issue with proper stuffing and/or wall lining. The lowest-frequency standing wave that can form in Zaph's enclosure is about 500 Hz given its longest dimension of 14". He is lining the walls with Whispermat which can easily absorb that low so it ain't broke and you don't need to fix it.

Edit: okay, that calculator does tell you that the lowest mode is 487 Hz. I just did it in my head and got close enough for guvmint work: 560/(height in feet).

[tpremo55]

One additional note about the modification of the enclosures of established designs. The things to keep in mind when modifying the enclosure as as follows:

1. Baffle Width
2. Tweeter Height
3. Enclosure Volume

Keep all the same and you are usually good to go.

I found google sketchup pretty handy when designing odd-shaped cabinets (unless you like differential equations) to establish the cross-sectional area. It's honestly very easy to use. It's been years since I've been active with AutoCAD and I would say sketchup is easier to pick-up than AutoCAD was. ... not to mention it's free.

Todd

[ch83575]

I think Dennis H is right that you probably don't really need to loose sleep over the standing wave issue. But, from what I have seen most of the special shape cabinets are extremely well braced and well built. Build quality and bracing effect cabinet resonances, not standing waves, but I think that it is a much more real consideration in the sound of a speaker. I am amazed at how much of the boxes resonating character can be heard, even if it doesn't look like it is showing up in the measurements. My intuition is that while you don't need to theoretically worry about standing waves in this design that if you choose to go with a more esoteric box design you could very well end up with an audible difference. I think it mostly comes down to construction and execution as much as anything else.

-Chad

[peter_m]

tpremo55:
Really nice links, thanks mate! My favourite is this one: http://www.hiendfi.com/modules.php?n...rticle&sid=361 Mindbogeling complexity and precision to make it all fit.... wow, just guys in a garage... Amazing! I will keep google sketchup in mind for when the time comes. Thanks.

DennisH:
You might be right. My only concern is I want this to be the best I can make it. If I can find proof that a curved side box can help and I can find the proper shape (not based on looks) I think it would be worth it for me.

ch83575
Then if I just made my walls thicker, with more bracing, then the same will be accomplished? If so, then 1" for side walls, is that enough or should I go thicker?



I've copied a picture of the inside of the magico mini... could be a good starting point. But the question remains.... What shape is best to minimize the standing wave? Remember reding Zaph braging about his TransLam design... show us some plans Zaph... You had posted it a few months ago in your blog...

Peter_M

[Andy_G]

deleted

[peter_m]

Andy,
have you experimented between trapezoid and rectangular, have you heard differences in sound?

[Andy_G]

deleted

[JonP]

Assuming I can manage to keep the same baffle width, is this site any usefull?

http://vikash.info/audio/standing_wave_calc/index.asp

A more detailed, and standalone, calculator is Bob Collo's Boxnotes:
(be sure to check out the rest of his site, lots of good stuff there!)

http://www.subwoofer-builder.com/freesoft.htm

It will help calculate the port volume as well, but you still need to add in however much the bracing will add.

But, don't get too hung up on size... like everyone's been saying, it's a relatively minor effect, and you can mostly eliminate it with good damping materials.

And just in case it's not totally clear, the important things about a designs box are, the front baffle dimensions and the driver's spacing and position... (have an affect on the mid-high frequency response, so the crossover design is built around it) the volume of the box and tuning of the port... (affects the low frequency response, and is chosen for an optimum tune for that woofer)

These things are set by "Mother Physics", keep them the same and you're free to get creative with other dimensions...