figuring out the proper speaker encloser size

How does a person figure this out? I have two vintage 12" (30cm) hi-efficiency (that's what I was told) fullrange speakers. The guy said they are rated for 10 watts/ 8 OHMS. That's all I know. Is it possible to figure out the proper encloser for these?

The speakers are in the US but I am in China now. But labor and materials are dirt cheap over here. I was thinking about having them made (disassembled) here and shipping them to the US.

inperson

 

Hi,

The following article will give you some rough guidelines for determining speaker enclosure size. There are many such articles at various web sites. Do some searching for more information.

Regards, HG

"Speaker Enclosures are more than just wood pieces randomly thrown together and screwing speakers into them. The volume and port size and length (if you go ported, which I HIGHLY recommend) is determined by some basic arithmetic involving some of the speakers' specifications, and your taste.

The following table gives an approximate volume in cubic feet you need to plan your enclosure for based on the woofer size. The woofer is the single most important element in calculating your enclosure as it makes 80% of the air or more.

Woofer Size --- Enclosure Volume
4" ======== .25 - .39 cubic feet
6" ======== .35 - .54 cubic feet
8" ======== .54 - .96 cubic feet
10" ======= .96 - 1.8 cubic feet
12" ======= 1.8 - 3.5 cubic feet
15" ======= 3.5 - 8 cubic feet

How you distribute that volume is pretty much up to your needs, just don't make them square! Square enclosures will make awful bouncing frequencies. You might have noticed that many sound reinforcement speakers have the sides of the box tapered in toward the back, if you can do that, do it. that helps sound move out and not bounce like it does between parallel surfaces.

To figure your enclosure size in inches, multiply all the sides as follows height x width x depth, so if your box is 16" x 12" x 8", you have 1536 cubic inches. Take that figure and divide it by 12 three times, so 1536/12/12/12=.89, so, your enclosure is .89 cubic feet.

Easy, huh?

OK, that is the rough volume, to get a more precise figure for the volume of your enclosure you will need two numbers from the specifications of your woofer. The Q (or, Qts) and the V(as) ratings. The V(as) is usually measured in Cubic feet, but some companies use liters, if this is your case, divide liters by 28.32 to get the cubic feet.

To get the volume, compare your Q rating to the chart below, take the multiplier listed for your Q, and multiply that by the V(as), and that will be the optimum volume of your enclosure in cubic feet.

Q ---- Multiplier
.2 ==== .1
.25 === .25
.3 ==== .4
.35 === .75
.4 ==== 1.1
.45 === 1.5
.5 ==== 2.0
.55 === 2.6
.6 ==== 3.5

And there ya go, the exact optimum volume for your speaker enclosure, so just figure out how to best size the enclosure, given your volume, to fit your needs."

 

But I don't know the specifications of my speakers (Q & V). That's the problem

inperson

 

I guess I could call home and ask someone to see if these figures are on the speakers somewhere. Do they mark them with this info?

 

These specs can be measured (Vas, Qms, Qes, F3), but you need the equipment to do this. (I read about this back in the 80s in Speaker Builder magazine.) You might try and see if anyone at any forums featuring speaker builders might have these specs, or know how they could be measured. I think the folks at Madisound may have a forum. http://www.madisound.com

It's not even worth trying without having the specs, and then, each cabinet design has its own formula. Likewise, speaker drivers are often designed for specific enclosure types--roughly speaking, a speaker with a loose compliance does best in a sealed cabinet, where stiffer suspensions work well in vented or transmission line systems. Some are even designed for open baflle (enclosureless) designs.

 

Not a big deal......For my homebuilt 3 ways with 12" paper cone woofers the calculated size for a ported cabinet was 3.6 cubic feet. This conforms to the chart which I provided. If you error on the high side of this chart, you won't be disapointed. If your speakers have paper cones, I would use a size of 3.5 cubic feet. Up to a reasonable point, a larger cabinet size than your speakers require will only help you.

I have seen very few 12" woofers with paper cones that require a ported cabinet size larger that 3.5 cubic feet. A check of the recommended cabinet sizes for various 12" paper woofers sold by parts express and others will verify this.

If you plan to use a sealed cabinet, you can reduce the size by up to 20% and still be in the ball park.

Good Luck!

HG

 

"Beg to differ" all you want, but selecting speaker cabinet size is not an exact science. Cabinet size is always a compromise. Take the Bozak Urbans for example (my son owns a pair). The sealed cabinet size is much larger than that prescribed by current standards for it's 12" woofer specs....It is somewhere in the neighborhood of 8 cubic feet!!! Yet they sound wonderful when driven by his MC 30's.

The Urbans were built during the era when big was considered better, and therefore, for marketing reasons, a larger cabinet size was more "saleable."

Today, just the opposite is true. Therefore, you will find many speakers in cabinets much smaller than their specs would suggest.

Much more important than exact cabinet size is making certain to add adjustable L-Pads to your mids and tweets to get the drivers balanced properly. If you have ever built a pair of homebrews, you know what I mean!

It is my contention, that most speakers, both home built and store bought, are not balanced properly. This imbalance is caused by differences in the speaker drivers themselves, the crossover components, amplifier channel output, etc.

If you don't believe me, I would suggest that you buy a sound level meter at Radio Shack and check your own speakers. Play a pink noise recording and measure the outputs of each of your left and right drivers and compare them. Hold the meter 3-5" from the cone of each driver to take the measurements. You will be unpleasantly surprised!

HG

 

Thanks, all of you, for the input

 

Obviously, you haven't spent much time listening to Vintage Bozaks....they are particularly noted for their their ability to deliver tight, intense bass with supurb accuracy.

Have a nice day,

HG

 

I like your approacch to things here. Everything doesn't have to be by the book, it can still end up nice

Thanks, inperson

 

Well yes there is an exact science to enclosure design.

The first thing is to determine the complete Thiele-Small parameters, Qts, Qms, Qes, Vas, Fs, Cms, BL, Rm etc… you can determine these parameters by doing complete measurements, in a chamber or there are a few products that specialize as a T&S woofer tester. This is a must, from this data you will determine if the woofer will work in a sealed cabinet, vented, and total volume of the enclosure.

There are set criteria for bass alignment:
Sealed:
Qts greater than 0.3
Low Fs, high mass, and high compliance(Vas,Cms)
Medium BL
EBP less or = to 50

Vented
Qts 0.2-0.5
Moderate Fs, medium mass, and high compliance (Vas,Cms) optional
Strong BL
EBP greater than or = to 100

To determine EBP or “efficiency bandwidth product” divide Fs by Qes if Qes is not
known Fs/Qts.

All of this is just to determine sealed or vented. From there you will need to decided if sealed the Qtc of the box:
0.5 Critically damped
0.577 Bessel
.707 Butterworth or B2
All of this effects the sound and size and the amount of stuffing.

Vented you must determine the tuning Fb, this can only accurately be determined by measurements. You will want to align your tuning near the Fs of the woofer, not always some designers will chose to play with this. The only way to truly determine the tuning is to load the woofer guess a port diameter and length and width and measure if it does not measure at the tuning frequency you were aiming for, then you will need to change box volume or the port length and width or maybe add or remove stuffing which changes the internally volume the woofer sees, more you stuff larger the volume.

Sealed boxes are a bit easier to design without test equipment.

This is a very rudimentary start to determine box volume and alignment, you can guess and sound will come out of it, but to do it correctly, it takes time and science. After this start is where the hard part come in, voicing making small micro adjustments in tuning, the volume, woofer to floor distance, stuffing and damping, port size and location or adjust the Qtc. That is the magic but there is some science that is required to choose a box.

 

Wow!!! With one posting you just scared the beJesus out of 100 members who were thinking about build their own speakers.

Lets be realistic, a speaker driver is a very simple mechanical device that pumps air. As much as speaker manufactures try to mystify the process, and pimp their own products, anyone with a little common sense can build a pair of speakers to rival most of the junk on the market today. And, I must say, at a much lower cost!!!

Please go back and read my first post in this thread. I still maintain that by using the "rule of thumb" information provided in this post, I can build a speaker box that will perform as well as any that you can build using whatever calculators you like. Actual experience has shown me that within +/- 15%, box size differences are almost impossible to detect by the human ear.

Regards,

HG

 

Once you figured out the size of the box you have to design the crossover. And that’s even harder...

 

Interesting! As I recall, the CM 911 had an output rating in the range of 75 - 100 wpc rms. If this is correct, who in there right mind would use this amp with speakers designed to be used with tube amps in the 20 - 35 wpc range. The Bozaks had SPL's of 92-96dB. Any amp over 40 wpc would seriously overpower the speakers. Hmmm.....................

HG

 

Regards
Robert

 

I agree, and I have a few years of Speaker Builder magazines, with many plotted response graphs and impedance curves, that show the Thiele-Small parameters ensure you get the most accurate response out of any given driver. The formulas are not difficult by any means. Heck, one of the first computer programs I wrote was in BASIC, and it was a T-S calculator.

If you want to get down to the REAL basics: a driver in too small of a cabinet has a huge midbass hump and no low bass response. Too large of a cabinet and you will get response that reaches lower, but it will still be weak AND you risk driver damage as the driver is acting "unloaded" in that large of a volume. In fact, true acoustic suspension speakers like those in the Advents were made with very loose suspensions; the small cabinet acted as the suspension (or "air spring", if you will) that supported the driver and kept it from having a meltdown.

Ported/vented speakers aren't a case of, "Let's stick a tube in the cabinet to shoot more bass out of it." Far from it. Ports are for tuning the cabinet, nothing more. The resonance of the cabinet volume, the driver's free air resonance and the resonant frequency of the tube allow the designer to tune the cabinet to get the smoothest response. My main speakers for the past 27 years are vented, and I still have yet to hear similar sized speakers that sound like them. Passive radiator systems are the same, except that they use a specifically weighted cone to tune the bass response. A transmission line works on a different principle (usually, the line length is a quarter-wavelength design, based on the driver's free air resonance). And horn speakers are based on formulas too complex for my little brain to handle. Dual bandpass enclosures are also based on different formulas but, even there, the driver specs are always used to get the most response out of the driver.

I had a pair of old speakers that I'd blown the woofers in--they were originally 8" acoustic suspension drivers. Madisound had some Philips 6-1/2" woofers on sale, and I bought eight of them in anticipation of a project I never got around to building. But since I had the T-S parameters, I worked backward using the cabinet volume to arrive at an ideal port length to turn these into ported speakers. And it came out very nice! The cabinets were a little undersized, but not by much--there was a slight mid-bass rise to it, but nothing unpleasant. Thing is, the port helped me tune it properly. And to run these sealed, I would have needed a much larger enclosure, according to the T-S formulas. For awhile I just had them slapped in the cabinets as-is, and they sounded like dog doo.

And this isn't even taking into account the crossover design...

 

Actually my response came from you scaring the “beJesus” out of me with overly-simplified statements that allude to the fact that you can take any woofer, regardless of TS parameters, and load it in an enclosure that you have determined via simple charts and consider the overall product to be well-designed and a rival for reputable speaker designs. Your charts represented an idea that the size of the woofer was the only determining factor in the enclosure volume, this leads me to a few questions for you; without knowing the Fs & Qts how did you determine bass loading, tuning frequency, port length and width? If you made no effort to determine these vital parameters, then all you have done is shoved a driver in a box with a port…

Your statement of, “Lets be realistic, a speaker driver is a very simple mechanical device that pumps air.” has left me a little speechless (though not unable to write!). You are acting as if drivers do not have impulse response, waterfall cumulative decay, group delay vs. speaker response, distortion, overall frequency response, and huge differences in overall sonic characteristics and behavior; plus, you don’t even want me to get started on the differences between lowQ, high-damped compared to hiQ, ultra rigid drivers or motors with magnetic non-linearities, SD motors, under-hung voice coils, voice coil materials, Le of the voice coils, etc… Not to mention that driver matching is a key ingredient to speaker design and without knowing how to do rudimentary measurements and understanding of TS parameters you could never get close to the best on the market of speaker designs, so the price should dramatically lower with much lower investment in time, knowledge, and measurement equipment. This means your quoted statement at the beginning of this paragraph shows me that “you are so far behind in the race that you think you are winning!”

 

Because most modern speakers are designed with SPL's below 90db!!!

HG

 

So when you stated, “As much as speaker manufactures try to mystify the process, and pimp their own products, anyone with a little common sense can build a pair of speakers to rival most of the junk on the market today. And, I must say, at a much lower cost!!!” I agree with you that what you said could match most of the junk, which comes from designers doing the same thing you are talking about (namely, shoving drivers in a box, crossing their fingers, and hoping for the best), but real speaker designs are much more detailed and well-thought out than what you discussed. But, on your other part of the statement, I disagree with you completely. The stuff I discussed was simple bass loading and I am not trying to mystify and considered what was discussed to be some of the more basic, rudimentary elements of speaker design; the truly mystifying aspects of speaker design would be the transfer function, time and phase relationships, on and off axis, baffle step compensations, impedance equalization, and maximizing power response…all of that is the truly scary stuff, not bass loading and volume of enclosures! Drivers and speaker design done correctly is a very complex task and, in opposition to what you believe, I heartily welcome any and all interested parties to try their hand at speaker design, so they can get a true working knowledge of the beauty and complexities associated with it.

 

By the way, tell me about your homebuilts. What drivers, what box size, and what xover points did you select? Are they 2 way or 3 way, and, what order of xover did you sellect? Also, what xover points did you finally arrive at? Finally, what L-pads did you use to balance them and what method did you use to balance them? If possible, please post a picture of them. Thanks

HG

 

Your posts are typical of someone trying to show how much they know rather than trying to help the person who started the thread!

He did not ask about ports, he asked about box size. If you have ever built a set of homebrews, you should know that port length, width, tuning, and location can be altered to suit your tuning preference. For God's sake, he hasen't even indicated yet if I wants to build ported or non ported boxes. Why are you talking about ports???

Yikes!!!

HG

 

I'm a big proponent of DIY. There's nothing wrong with building something and then if it does not make you happy figure out why. This is top down learning - nothing wrong with that. Many people learn by getting their hands dirty first 'cause chalkboard stuff bores them. THAT'S how the first triode amplifier was made!!!!

Take you best stab at a plan and go for it!

DIY is the only path to audio salvation
Regards
Robert

 

Forgive my ignorance but do I need a crossover for a single unit speaker?

 

I, too, was trying to help the original poster by not letting him build an enclosure in China, have it shipped to the US without even knowing TS parameters, which is a fundamental underpinning of speaker design.

“he hasen't even indicated yet if I wants to build ported or non ported boxes. Why are you talking about ports”…
I did not mention ports, in reference to anything other than the fact that without knowing the TS of the previously mentioned woofer any guess of the enclosure size would be futile. I only mention that the driver will tell him what enclosure it needs, not the other way around; likewise, although you highly recommended ported (without knowing anything about the driver’s TS parameters) being used, you then went on to make a blanket statement that if ports were not used that you could just shrink the box for sealed… are you joking? For what Qtc? (You seem to be quite reckless in your recommendations.) I mentioned ports to you about your design (not, in reference to any other), which you said was ported, how did you come up with the proper measurements to determine tuning frequency and Impedance curve (questions that you have yet to answer)? So, again I will ask how did you determine the tuning and port dimensions?

“If you have ever built a set of homebrews, you should know that port length, width, tuning, and location can be altered to suit your tuning preference.”…
NO, you cannot just change volume and port dimensions to suit preferences, that would make it a tone control. They are chosen by the woofers TS parameters, Fs (in particular), and box tuning; this has to be determined by measurements or you can cause damage to the drivers by significant reduction in electrical impedance at the box resonance frequency. This is not something to guess at, it is based on the TS of the driver. Two drivers may be the same size and both may have the proper Qts for vented boxes, but if they have very different Fs and Vas parameters, then they would need to be in entirely different boxes with different port dimensions and tuning, as they are not able to be generically “mixed & matched” according to size. According to the advice you gave about design, only the size of the woofer is the predominant factor…wrong again.

I do not make homebrews, but I have made many designs in the lab that turned out both good and bad, so I do know what I am talking about. I was not trying to show how much I know, just as much as I can assume that you were not trying to show how much you know, I wanted the original poster to realize that knowing the drivers specs was a bit more critical than just the size of the woofer…you were the one who said that disagreed saying your “rule of thumb” method would allow for a market-worthy enclosure and if sealed just make it 20% smaller and a more detailed, thoughtful approach is just marketing and pimping?

To answer the questions you posed, here are the specs for a speaker that I worked on, of course it was a team effort as I am not audacious enough to think I have all of the answers, the transfer function was done in conjunction with Phil Bamberg who has over 20 years experience and has work with large companies, such as Klipsch, and small companies.

By the way, Lpads are only the beginning of driver balance, you will never get a flat amplitude response without careful transfer analysis, but they are a good place to start.

Transducers
18cm woofers x2 Audio Technology
28mm tweeter Scan-Speak 9900

Impedance
Nominal 4 ohms
Minimum(@145/ 20 kHz) 2.4 / 5.2 ohm
Phase (@80 / 12k Hz) 24 /+23 degrees

Amplitude response
To 20kHz +1.0/-2.9

Sensitivity
2.83v@1-meter full space 88.5dB

Bass alignment
Vented HPF4; F3@45Hz, F6@35Hz

Inter-driver Phase response
600Hz to 6kHz Less than 15° phase difference

Max linear output
105dB 80w HPF2 @80Hz all frequencies, no compression
102dB 40w No HPF, all frequencies, no compression

Crossover
Alignment Linkwitz-Riley 4th
Elements 20, 6 of which are in the 3 drivers, series signal path
Phase compensation 3 elements in one circuit

 

No.