Questions about Enclosures

[RickX]

...and I thought designing and building speakers would be easy ! That was until I started doing some research into the science of loudspeaker design; an exercise that raised more questions than it answered !

I've been lurking around the Mission Possible DIY section for a few weeks now and have been totally amazed by the designs and craftsmenship of all the knowledgable forum members. I've learned a lot by reading the various threads, but I've also amassed a list of many questions that I haven't really found answers for.

This thread is all about my questions regarding enclosures. Some of them already have answers based on my "research", but most of them are still unanswered.

I would really appreciate any help with answering these questions I'll update this orignal post with answers as they come in. Hopefully other beginners like me will find the information in here useful. Thanks !

Questions

1. What does it mean to "tune the enclosure/box to 37Hz" ?
Not really sure. I'm guessing here, but I think its an exercise in setting the output frequency of the port. If so, does that mean that the Port will always output 37Hz regardless of the frequency being emitted by the woofer ?

1a. How do you do that ?
Increasing the port length decreases the tuning frequency. There's a formula here: http://www.clubknowledge.com/Car_Audio_FAQ/?t16
However I also noticed that reducing the port diameter decreases the tuning frequency as well. I would have thought that the opposite is true. Do any issues arise by reducing the port diameter ? Does it decrease the amplitude of the sound waves (lower volume) ?

1b. Does it apply to ported and sealed enclosures ?
I believe it only applies to ported enclosures.

1c. Why would you do that ?
[insert answer]

1d. Does it only apply to dedicated subwoofers, or can it apply to regular speakers as well (i.e. 3 way speakers) ?
I assume it applies to both subs, and regular speakers.

1e. Some speakers have 2 ports, and I assume they're tuned to different frequencies. What effect does that have ?
[insert answer]

2. What are the basic good enclosure design principles ?
- Box should be well braced
- Internal walls should be lined, especially woofer areas.
- Baffle should be have smooth round edges
- Anything else ? I'm sure there are a few more.


2a. Is there an ideal H x W x D ratio for enclosures ?
One suggested ratio from "Designing, Building, And Testing Your Own Speaker System, Chapter 3" is 0.37:0.68:1 (internal). Width x Depth x Height for imaging OR Depth x Width x Height for bass response. Although to me that seems a little too extreme. Most of the DIY designs here follow a 0.5:0.55:1 (WxDxH) ratio. Side Note: I haven't read that book, just the excerpt from Google Books.

3. Bass response or lower frequency response. With all other things being equal, will the bass response of a single woofer in a 20L enclosure be equivalent as 2 woofers in a 10L enclosure ? This question is trying to ascertain whether extra woofers can compensate for lower enclosure volume.
[insert answer]


4. Why do ported enclosures have a lower frequency response than sealed enclosures, when both are using exactly the same woofer ?
I'm not sure, but I think it's because the waves generated by the negative movement of the woofer are channeled through a port which forces the sound waves to travel a farther distance, in essence "stretching" them out, resulting in longer wavelengths = lower frequencies. That's why longer port lengths lead to lower Tuning Frequencies (according to the formula). Can someone please point out if my "analysis" is wrong !

5. I've noticed some 3 way designs (TMWW) have the WW sealed away from the TM part. Other designs have an "open" chamber design ? What's the difference ? Why would you choose one over the other ?
[insert answer]

6. What is baffle step compensation ?
Took me a while to understand get my head around this, but this is what I know so far. When the wavelength of sound is smaller than the baffle width, the sound is directional. When the wavelength is greater than the baffle width (generally from woofers), it becomes omni-directional at those frequencies and it seems those frequencies are attenuated (usually by 3dB). To compensate for this effect, the frequencies from the tweeter and midrange are attenuated to match the output of the woofer.

However, if you have your speaker placed on the wall or very close to a wall, you have a sort of infinite baffle scenario, so BSC is not required.

Judging when you need BSC seems to me to be more art than science. But a general rule is that the further you place your speaker from the wall, the more likely it is that you will need BSC circuitry.


6a. What if you have a woofer that is crossed over at 500Hz. According to your calculations, frequencies below 250Hz will be attenuated because they're longer than your baffle width. However if you go ahead with BSC circuitry to attenuate the tweeter and mid sound, won't you end up with a spike between 250Hz and 500Hz in the frequency response curve ? What do you do in a situation like this ?
[insert answer]

7. I've read that it's best to scallop the back of the front baffle to allow sound waves to exit the rear of the driver smoothly. I'm not sure what that means, and what problem "scalloping" aiming to resolve ?
[insert answer]

8. Passive woofers: Side firing vs Front firing. What are the things that need to be taken into consideration when designing a speaker ?
Bass has some directionality, so frequencies between 100Hz and 250Hz will be better represented by a front firing woofer. Below 100Hz, sound is more omni-directional and a side firing woofer will work just as well. Any other considerations ?

8a. Why would you choose one over the other ?
One reason to choose side firing would be to keep the width of the speaker narrow. But that's really a superficial issue. I'm more interested from a perfomance perspective.

9. Why do some speakers have the tweeter and the midrange off-axis, and others have them in straight line ? What are the advantages / disadvantages of the off-axis design.
[insert answer]

10. Is there an absolute minimum amount of space that should be left between a woofer and the sides of the enclosure ? E.g. If I had a 7" woofer, would it be problematic if I mounted it in an enclosure with an internal width of 8" ?
[insert answer]

11. Why are tweeters usually placed above midranges, and midranges above woofers ? Is it all about having the higher frequencies closer to ear level ?
[insert answer]

12. What is floor bounce ?
Floor bounce in an interference problem that occurs when sounds wave reflected from the floor cause destructive interference with sound waves travelling directly from the bass / midrange driver to the listener's ear.

12a. What should you listen for when trying to discern if your speaker suffers from floor bounce ?
[insert answer]

12b. Does it only apply to hard floors ? Or does it affect carpeted areas as well ?
[insert answer]

12c. How do you prevent it from happening ?
[insert answer]

13. What effect do Acoustastuff and similar products have ?
The basic premise is that stuffing a box with Acoustastuff or something similar (e.g. wool) will lead to a smaller enclosure exhibiting characteristics of a larger enclosure, because it absorbs and releases air molecules when the cone moves in and out respectively. This leads to lower than usual air pressure when the cones moves in, and higher than usual air pressure when the cone moves out - much like a enclosure with a larger volume.

13a. How do you know when you need to use it ? Is it only used for sub enclosures ?
Haven't been able to find a formula out there that dictates when to use it. It seems to be a subjective matter (like many things in speaker design). If you'd like a lower frequency response from your speaker, that's when you'd use the stuff.

13b. How do you know how much to use ?
[insert answer]

14. Isobaric design versus Bipole design. What are the differences in sound characteristics produced by each design ?
[insert answer]

14a. Under what circumstances would you use an Isobaric or Bipole design ?
[insert answer]

[J-Dub]

Hi Rick, Seems you have a lot of questions.
I am sure I will not be able to answer them all but i will do my best to answer a few.

question 1. : No it will not always put out 37Hz. Google "portedxls" and there is some very simple software that will help explain this if you play with it for a while.
You are tuning the enclosure much like a pipe organ wich will "boost" the lower frequency so that the driver will find it easier to perform at frequencies lower than its FS. This will create a curve in the frequency response pattern that you will want to use to determine how the driver will perform in such hypothetical enclosure.

1a: decreasing the port diameter Will lower the tuning but you will need to consider "Chuffing" the sound of the air blowing through the pot tube. It is recommended to use the largest port diameter that will fit in your enclosure considering realistic lengths.

1b: ported only applys to ported enclosures however the "Q's" of the driver will determine if it will work well in a ported or sealed enclosure.

1c: depends on the driver

1d: usually you are porting the box "tuning it" for the lower frequencies, therefore it applys to subs and the woofers of a 2 way or three way enclosure.

1e: 2 ports are used in the same effect of one larger port. IE: 2 times the diameter of the smaller port size. Not recommended except for special situations. again beware of port noise or chuffing.

2: Yes
Yes
depends on how you have accounted for baffle step in your crossover.
also, as a general rule i usually double the thickness of my side walls for the front baffle. IE: side wall = .75 inches, front baffle = 1.5 inches.

2a: This could get pretty comlpex but i dont think it is as necessary as it seems to be. 1st make sure front baffle will accomodate largest driver. then find a suitable depth. adjust hieght and depth to accomodate port length and make sure to have plenty of room for air to move around the port inside the box. Never put the port directly behind the driver and try not to put the port in the corner of a box or in a situation where air cannot move through it freely. I try to make sure that the space between the open end of the port and the nearest wall is at minimum the port diameter. IE 6" port equals minimum 6" from the wall it is firing into.

3: Generally 2 woofers in the SAME sized enclosure will give you a 3 decibel gain. i dont think reducing the volume by half and adding a woofer will do much to help without a significant increase of the amplifiers power. remember that every doubling of the amplifiers power will give aprox 3 db gain. IE 500W amp = 100db, 1000W amp in same situation = 103db. Reducing box volume will also affect the "Q" of the box. This will be explained on the same website mentioned before. Again "Google" "portedxls" you will need microsoft excell to run the program but there is a lot of usefull info there as well.

ok I'm done for now. Hope that helps and I will let others continue to answer your questions. Good luck and welcome to Mission Possible!!

[brianpowers27]

4. Think of the port as a 2nd speaker. THis spekers is tuned to compliment the natural rolloff of the main speaker. The frequencies this driver will play are usually centered around a very low frequncy and are band limited.

5. Most woofers/mids radiate the same sound to the front and back of the speaker. An isolated chamber attempts to reduce the interferance of the backwave from the mid to the woofer. I couldn't tell you how audible this is. Zaph has a short article explaining this effect on a 2.5 way speaker at his site.

6. I can't add much, since your explanation is good. Generally BSC is 6db, not 3db of loss. It is important to note that the only speaker requiring no BSC is in wall. Even small amounts of cabonet depth will affect the amount of BSC required. I generally design my speakers about 10"d. I also place them out from the wall between 6-24". I find that full 6db BSC is right for this. WHen I use less BSC I find that there is a midbass hole between 80-300hz.

7. Scalloping removes some of the compression that occurs when the mdf interferes with the speakers ability to push the cone backward. THe front reaction is as important as the rear reaction due to its AC nature.

8. Front firing gives you the benefit of better control over directivity. THeories differ on how directional we can perceive. My anecdote: I placed a sub behind the listening position crossed at 80hz lr2 and it seemed directinoal. When I placed it at 60hzlr2 I couldn't find it. Keep in mind that the slope is very important in this consideration, since it isn't the low frequencies we hear.

9. Offset tweeters attempt to diversify the pathlength from the radiating source of the tweeter to the edges. This gives benefits in the form of smoother diffraction response. It is easy to try in most modeling programs...

10. NO absolutes...
11. 2 reasons, ear level and aesthetics.
12. Floor bounce frequency is dependent on the path length from the speaker to the floor to your ear. This bounce is affected by the characteristics of the surface. See NRC coefficients for more information regarding surfaces.

13.

[J-Dub]

12. Brian is right look up absorbtion and reflection coeficients of different materials. Ideally absorbing the wave from the floor should be better. IE: carpet. Hardwood will reflect the wave without much absorbtion. It is important to know, however, that waves will be reflecting off of almost every square inch of surface in the entire room. Axial, Tangential, and Oblique. Walls, ceiling, floor, tables, chairs etc... It is good practice to have absorbtion on the floor, diffusion on the rear wall sometimes the front, and absorbtion on the side walls. Diffraction can be used on side walls and ceiling as well. This is not an absolute must for your home setup by any means but most studios are set up this way. To really make things confusing a normal small bedroom generally has enough absorbtion through the carpet, bed etc.. to make a nice listening environment.

13. If you want to go this far you should read the master handbook of acoustics and Kings work on transmission lines. Absorbers in the enclosure do a lot of things but mainly turn the backwave into heat and absorb some of the wave. The heat is probably not very noticable but it is happening through friction over the fibers. Too much to try and explain the rest.

13a. Different materials will have different effects for different frequencies. R13 insulation works well for lower frequencies in the box. Acoustafoam and tighter denser materials will help with higher frequencies. Again you will have to read the handbook to get all of this. There are chapters on it.

14. go to www.diysubwoofers.org to find out some basics on different types of enclosures.

[RickX]

Thank You !

J-Dub and Brian, thank you very very much for taking the time to answer my questions. I really appreciate it, and have learnt a lot from your answers.

I understand that mastering acoustics is a life long endeavour, but I feel I'm well on my way now to understanding the basics. Thanks again.

I downloaded Ported.xls and Unibox, and I'm learning how to use them to select the right type of enclosure.

1b: ported only applys to ported enclosures however the "Q's" of the driver will determine if it will work well in a ported or sealed enclosure.

I've read that drivers with low Qts produce tight/punchy bass, while high Qts drivers give slow/heavy sound with lots of LF output. What I don't is that magic number that determines whether a driver is more suited to sealed or ported enclosures. Is it Qts=0.5, where Q > 0.5 = sealed and Q < 0.5 = ported ? I'm just guessing now...

Never put the port directly behind the driver and try not to put the port in the corner of a box or in a situation where air cannot move through it freely.

Why would you not put the port (and I assume you're referring to the port opening) directly behind the driver ? I assumed that would be a good idea because the sound waves generated from the rear motion of the driver would exit through the port more easily. Not sure if I've explained myself too well there...

5. Most woofers/mids radiate the same sound to the front and back of the speaker. An isolated chamber attempts to reduce the interferance of the backwave from the mid to the woofer. I couldn't tell you how audible this is. Zaph has a short article explaining this effect on a 2.5 way speaker at his site.

I'm struggling to understand the concept of interference within the enclosure. Why does it matter ? I mean if you have an enclosure that is thick walled, well sealed, braced and damped, what effect does internal inteference have ? I mean I could have two drivers backwaves interfering with each other internally, but that shouldn't matter since I really wouldn't be hearing any of that. Or does internal interference color the sound externally somehow ? Is it because of the port opening ?

9. Offset tweeters attempt to diversify the pathlength from the radiating source of the tweeter to the edges. This gives benefits in the form of smoother diffraction response. It is easy to try in most modeling programs...

I need to learn more about diffraction response because that explanation went a little over my head. What modelling programs were you referring to ?


Thanks again for your replies

[J-Dub]

You are welcome

1b.. http://www.diysubwoofers.org/sbc/sbc5.htm

Quote from this page: "Note: if you are using an existing driver, the driver's T/S parameters will typically make the choice for you. Low Q (<0.3) drivers are generally more suited to higher-order bandpass and horn-loaded systems (which are not covered here). Drivers with Qts between 0.3 to 0.4 are usually best used in vented systems, and drivers with higher Qts are usually best suited for sealed systems. 4th order bandpass systems can usually work with drivers with any Qts between 0.3 to 0.6. In the case of car audio systems, you will almost always get the best results with a simple sealed system"

Why would you not put the port (and I assume you're referring to the port opening) directly behind the driver ? I assumed that would be a good idea because the sound waves generated from the rear motion of the driver would exit through the port more easily. Not sure if I've explained myself too well there...

You dont want to do this because the idea is to control the backwave of the driver, creating pressure in the enclosure forcing waves in and out of the port by use of the enclosures volume. If you put the port opening directly behind the driver you will get a little use of the enclosure volume but also some backwaves will be traveling axially "directly" out of the port. It will just defeat a lot of what you are trying to do.

I'm struggling to understand the concept of interference within the enclosure. Why does it matter ? I mean if you have an enclosure that is thick walled, well sealed, braced and damped, what effect does internal inteference have ? I mean I could have two drivers backwaves interfering with each other internally, but that shouldn't matter since I really wouldn't be hearing any of that. Or does internal interference color the sound externally somehow ? Is it because of the port opening ?

The idea of an enclosure is to control the backwaves of drivers so that they either dont interfere with the front waves or help the front waves. A quarter wave tube is an example of the backwave helping the front. Ported enclosures are kind of the same way. A sealed enclosure is attempting to hold back the backwave as much as possible.

Now to answer your question, No enclosure is perfect. even with sufficient dampening and sturdy walls there will still be waves bouncing back into the drivers creating unwanted or colored sound. It is our job to do as much as possible considering our budget and space, to isolate, control or use those waves in the most productive manner.


9. I will let others share with you what programs they use, but diffraction is best described as "The distortionof a wavefront caused by the presence of an obstacle in the wavefield" [Master Handbook of Acoustics 4th edition, F. Alton Everest, glossary.]

I am not an expert on baffle step theory but the best I can gather is this: If a wave is traveling and hits a wall with a rectangular opening in it (imagine a brick wall 10 feet high 50 feet wide and an opening like a door in the middle of it) the wave will bend around the walls on the opposite side through the opening in a hemicylindrical fashion. The opening now becomes the NEW source of sound to the listener on the opposite side of the wall. Given that sound is capable of bending 90deg even from a speaker mounted on its baffle the waves traveling along the baffle may be amplified according to the baffles width and the length of the wave. The wave comes out of the speaker is diffracted along the front baffle untill it is diffracted again off of the side of the baffle. The higher a frequency is, the shorter it is, so it makes sence that the short width of a speaker baffle will have some effect on shorter waves.

That was the best I can do for that one hopefully someone else can double check my work for you or give you an easier explaination.

[brianpowers27]

I'm struggling to understand the concept of interference within the enclosure. Why does it matter ? I mean if you have an enclosure that is thick walled, well sealed, braced and damped, what effect does internal inteference have ? I mean I could have two drivers backwaves interfering with each other internally, but that shouldn't matter since I really wouldn't be hearing any of that. Or does internal interference color the sound externally somehow ? Is it because of the port opening ?

The backwave will do 2 things in a sealed box. Neither are desirable.

1. resonate the box walls
2. return back to the driver and oscillate the cone.

In a ported system the backwave can do 3 things. Point 1 is less than that of its sealed cousin, due to the pressure amounts involved.

1. Resonate box walls
2. Exit through the port unfiltered () Yuck
3. Return back to the driver and oscillate the cone.