diyAudio (
-   Subwoofers (
-   -   Unconventional subwoofer design (

BAM 14th December 2003 12:08 AM

Unconventional subwoofer design
Ok, here is what's up...I am working on a design for an 8th-order bandpass subwoofer. Tentatively this enclosure will use two Peerless 833599 6.5" woofers with rubber surrounds, parts express #297-628. What do you think about this?

I will probably use Isaac's Subwoofer Simulator for this.

kelticwizard 15th December 2003 05:24 AM

2 Attachment(s)

Just figured I would include a diagram of an eighth order bandpass in case anyone is familiar with the configuration, but not the name.

Also, I was in a thread where someone had simulated a tremendously low F3 for a bandpass box. There was a post by someone who seemed to be a professional who said that bandpass simulations, especially those of configurations with multiple ported chambers, do not conform to real life output.

I could not comment one way or another, having never built a bandpass. But I thought I would pass it on.

Is there any particular advantage of this enclosure style? Does the 8th Order bandpass give a lower F3 than other enclosures, such as the vented box?

I have heard that the more resonators, (ported chambers), involved, the greater the group delay, and possibly worse transient response. But again, I do not speak from experience on this.

Here is the configuration for anyone who has seen this enclosure.

Good luck with your project. Please let us know how it turned out.

paulspencer 15th December 2003 06:12 AM


I'd have to say I'm not in general a fan of bandpass subwoofers, although it does depend on your application. I'd say they are suitable for non critical applications, where you are willing to sacrifice some quality to get more output - say in a PA setup, or for some boom boxes with small drivers that otherwise wouldn't really be up for the task - especially computer speakers if you want to add a little boom to your computer games. A 4th order bandpass, OTOH, is perhaps more feasible as it doesn't compromise sound quality as much. Jamo once had such a sub, with a 10" driver in their flagship speaker - actually they had two of them in isobarik configuration inside a curved speaker.

It really depends on what you want it for. I believe bandpass boxes are very sensitive to the parameters of the driver, so it may prove hit and miss unless you actually measure the parameters to be sure they are what they are supposed to be.

Subwoofer simulator is a good one to use. However, be aware that if the parameters of your driver are a little different to the published specs, it could mess up your design.

Ok, so why use a bandpass? ...

Basically it allows you to trade efficiency and bandwidth. They gain output by narrowing the bandwidth. So, compared to a vented box, a 6th order bandpass will have a higher F3 and higher efficiency. They also tend to have higher power handling due to the increased excursion control. This is particularly useful in PA, where low extension is not really useful, and high output is desired. If you design a bandpass to have a lower F3 than a vented box, you start to loose efficiency. If you want a lower F3, this is probably not the way to go. If you want a lower F3 than you can achieve with a vented box, your options include:

transmission line
horn loaded
sealed box with eq (with high excursion drivers, at the expense of output)

Bose used an 8th order bandpass with small midbass drivers in their older acoustimass modules. They go low enough for most music and roll off rapidly on the upper end, so that they are more flexible placement wise - distortion is less audible as it has to get out through the ports, and the rolloff helps also to prevent being able to localise the sub. If you aren't fussy about sound quality, it's quite clever, and will easily satisfy most people. But it will satisfy very few who are likely to be found browsing this forum!


paulspencer 15th December 2003 06:17 AM


one more comment - my suggestion is to get hold of some scrap MDF or whatever you have available and have a go at a box - compare vented to bandpass, but don't put in too much effort to a nice box at first - make a quick dirty prototype and be ready to experiment a little.

Another thing that would be interesting to try is to model with two versions of the driver - put in a second version of the driver with slightly different parameters. Also, make sure you break in the driver as the parameters are likely to change as the suspension gets a little looser.

It's best to make a decision after you have experimented yourself, and made a quick prototype. Let your own ears be the judge.


Bull 16th December 2003 10:38 PM

I could be wrong but.... 6 1/2 inch drivers in a bandpass box,the drivers wouldn't have enough x max.

mwmkravchenko 17th December 2003 02:01 AM

BE very kind to your woofers with this type of a setup. These boxes can be woofer eaters. I won't warrantee them if I build them. People tend to turn them up to loud because they have such a small passband. They "want to hear the tunes" and if you over drive the woofers by excess eq or wxcess power the flap around and melt, die , blowup etc. Not for the faint of heart!!!


kelticwizard 17th December 2003 02:58 AM

2 Attachment(s)
Well, I just modeled one of those Peerless 833599s. Subwoofer Simulator will not model two.

The power delivered to the woofer is 96 watts.

The sensitivity seems to go down, but the bandwidth seems to go down very far for a speaker with an Fs of 37. Speaker normal sensitivity is 88dB.

Group delay is a very large 45 ms.

Green line=output without room gain

Red line=group delay

Purple line=cone excursion

Next post is the box dimensions.

kelticwizard 17th December 2003 03:07 AM

2 Attachment(s)
Here are the box dimensions and tunings.

I would certainly make the Av of the vent larger than 1 inch, but I did not have time to calibrate a bigger vent for the same tuning.

But I think everyone gets the idea.

The box seems to have the potential to give smooth tuning for frequencies well below the speaker's resonances-in fact, about a whole octave below.

GM 17th December 2003 03:44 AM

>Subwoofer Simulator will not model two.
Sure it will, just input the composite specs of two drivers.


paulspencer 17th December 2003 06:02 AM

mwmkravchenko and bull are right - 6.5" drivers don't go far as subs BUT a bandpass box is the best you can do with a smallish box and these drivers as far as output and power handling go.

A bandpass box should not reduce sensitivity at all! Not unless you increase the bandwidth too much. Compare it to a vented box with similar extension and you should be able to get slightly more efficiency.


All times are GMT. The time now is 07:45 PM.

vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2