Box design

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

I've used WINISD to design a box for a pair of 2-way speakers (11cm woofer), and well ,I finaly decided that the box should be about 30L in bass reflex configuration to get some "good" bass.

I want to make a box like the one in the picture, but I`d like to know your opinion. Will it have resonance problems with such a "tall" box? Will it affect much the position of the port? (for example in the botom or in the middle)

An externally hosted image should be here but it was not working when we last tested it.


I've also thought about making a TL, but I can´t afford any of the software that is avariable for this single desing...(Is there any free version of the Martin´s mathcad models or TLwrx???)

thanks a lot!!!!
 
frugal-phile™
Joined 2001
Paid Member
The image you posted is really small... but it is large enuff to tell that it is tall enuff that it is going to be a TL, but with the port up by the driver it is going to be strange. The port would need to be near the bottom to take full advantage of the 1/4 wave resonance.

It also looks like a 2-way, which may be why you got little response.

What driver(s)?

You can download Martin's tables and the accompaning spreadsheet to model simple TLs.

dave
 
...alternatively, you can buy a year's personal DIY license of his worksheets for the rather minor sum of $25. If you can afford to buy drivers, material etc., then it's not a vast expenditure. If you need to save up for a month, do so. You'll be better off in the long run.

Re the lack of response, as Dave points out, your query was regarding a 2 way, and this is the section of the forum about wide-band drivers. FWIW, I would expect you to run into more or less severe resonance problems with a box of that aspect ratio if the design does not take these into account.
 
Thanks for your replies!!

Well... I think I better start in the right way...telling you the whole history.

My drivers are Hertz ESK130, which is a mid/low cuality car-audio two-way system:

http://www.hertzaudiovideo.com/Doc/pdf_esk130.pdf

The woofer is EV 130,and his electro-acustic parameters are:

An externally hosted image should be here but it was not working when we last tested it.


As you see, is a 11mm woofer, and low-bass response is very limited (its supposed you willl use a sub-woofer for that in a car...).

The thing is that this system is the one I have,and I can not change it... I just want to adapt it for use at home the better way it could be done (I mean...I do know these drivers were designed for car-audio, installed in a door... not for home use)

Anyway...I used WinISD, and made some tests. First, although I get an EBP near 90, i tested the closed box. This was one of the results, with a 10L box:

An externally hosted image should be here but it was not working when we last tested it.


Then, I did some tries with vented box (simple bass reflex). I think the best configuration is around 20L and tuned near 50Hz:

An externally hosted image should be here but it was not working when we last tested it.


(i got a group delay peak as high as 28ms around 45Hz....but i got similar results with any box volume or tuning frecuency)

The thing is... if i use optimum dimensions for 20L, the box looks very big and "ugly" for such an small drive, and because of that I wanted to build a tall and slim box, which looks better for home. Here is the picture again (good sized xD)

An externally hosted image should be here but it was not working when we last tested it.


So, planet10, it is just a "tall" bass reflex box, not a TL. But I don't know if it will work fine, because, as Scottmoose sais, that aspect ratio could generate problematic resonances... and that is what I was asking xD

But, I also thought about building a TL, because the size of the box will be similar.

I tried using Martin´s tables and the Excel worksheet, (I would do a classic TL), but those tables just can be used until Fs of 70Hz, and the EV130 have an Fs of 89Hz.

And well, I have very limited resources (the drives were free, a gift from a friend who changed his car woofers)...I know 25$ is not much for a year license...but for a single design...is too much. Well... after the license expires (no more updates), I can use the worksheets I downloaded until I want, isn't it? That's why I asked if there were no free old versions or something..

Thanks!
 
frugal-phile™
Joined 2001
Paid Member
Anyway...I used WinISD, and made some tests. First, although I get an EBP near 90, i tested the closed box. This was one of the results, with a 10L box:

...

Then, I did some tries with vented box (simple bass reflex). I think the best configuration is around 20L and tuned near 50Hz:

So, planet10, it is just a "tall" bass reflex box, not a TL.

The Q of your driver is quite high, any sealed box, no matter how large will have a Q higher than i really like to see, your sim has what i would consider a big bump at the bottom. The BR sim you show is worse, you want a smooth roll-off not a rippled one.

And because of the high aspect ratio of the box it will be a (non-optimal) quarter-wave resonantor (a TL) and not a BR, so the sim does not accurately represent what is happening.

What i would do with this driver, is to build as large a sealed box as you can live with (ie the one you have), then place a series of aperiodic vents at the very bottom on the back, and then damp appropriately to make it aperiodic.

You can see an examnple of what i mean by aperiodic vents in this picture:

4L-aperiodic-FR125.jpg


and a basic configuration of how to arrange the damping in the attached cross-section of the PEARL PR-2.

dave
 

Attachments

  • perkinsPR-2-sm.gif
    perkinsPR-2-sm.gif
    38 KB · Views: 80
*Edit -Dave beat me to it*

You're saying you don't think your music is worth $25 / the price of a couple of CDs? Oh well, to each their own.

What Dave is pointing out is essentially the same thing as I was: a tall slender box will end up with a powerful vertical standing wave and end up as an MLTL (Mass Loaded Transmission Line), or MLQWR as I prefer to call them (Mass Loaded Quarter Wave Resonator). You can try to break it up a bit internally, but that doesn't work all that well. Bass reflex is based purely on helmholtz resonance theory, which assumes uniform air particle density in the enclosure and no standing waves. Soon as you move away from that, things start to change; the transition from one load to the other occurs when you need to change the vent dimensions compared to what BR principles would suggest in order to maintain the same Fb.

Anyway, looking at the driver specs. you've detailed, it certainly wouldn't be my first choice for a BR; sealed or aperiodic would be far better bets. Alternatively, you could go with a simple, untapered Alpha TL (straight pipe, sealed at the top, open at the bottom). See attached. This one is set for flattest impedance; 29in long, pipe CSA 48.5in^2 (make it any shape you like providing the area remains the same), driver mounted as close to the top as possible & stuffed 2.2lbs ft^3 with dacron hollow-fibre material or similar. Nice damped response, F10 about 50Hz.
 

Attachments

  • Alpha TL.GIF
    Alpha TL.GIF
    5.6 KB · Views: 88
Last edited:
Aperiodic Speaker Enclosure Design

Mmm, very interesting, well... as you saw Im a newbie (the "newestbie" you can find :D), and I have never heart about this kind of enclosure...I'n not very familiarized with the electro-acustic parameters or the response/impedance curves...so I can not get inmediate conclusions just by seeing those numbers, as you do :p (many thanks for your help!!!)

The main problem is that I have to maintain a small sized box seeing from the top, but not beeing a problem how tall it could be (well... max1.70m xD). If I build a tall box with aperiodic vents, won't i get anyway resonance problems?

What i would do with this driver, is to build as large a sealed box as you can live with (ie the one you have), then place a series of aperiodic vents at the very bottom on the back, and then damp appropriately to make it aperiodic.

Do you mean I could use my desing ratio???

On the other hand:

Alternatively, you could go with a simple, untapered Alpha TL (straight pipe, sealed at the top, open at the bottom). See attached. This one is set for flattest impedance; 29in long, pipe CSA 48.5in^2 (make it any shape you like providing the area remains the same), driver mounted as close to the top as possible & stuffed 2.2lbs ft^3 with dacron hollow-fibre material or similar. Nice damped response, F10 about 50Hz.

Like This?:

An externally hosted image should be here but it was not working when we last tested it.


Thanks Again
 
Don't want to be annoying... honestly xD

If I do something like this, so the cross section remains constant until the open end, and considering the center line of the cross section the total length, will it still be an Alpha TL? (And then, your calculations still valid)

An externally hosted image should be here but it was not working when we last tested it.


PD: Is that the way it should be done in any geometric or in case of different cross-sections? I'm seriously thinking about buying the worksheets license...for future projects...cause in this one I'll use your numbers :p
 
Right, providing the cross-section of the line & its length are preserved, within reason a bend won't have a negative impact. There wouldn't be much benefit in bending it forward though; there's no output from the terminus to speak of (it's a purist TL). Personally, I'd just get it on some substantial spikes. Easier to build too. YMMV as ever of course.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.