diyAudio

diyAudio (http://www.diyaudio.com/forums/)
-   Subwoofers (http://www.diyaudio.com/forums/subwoofers/)
-   -   Variable Port Lengths/Diameters? Like a pipe organ... (http://www.diyaudio.com/forums/subwoofers/226600-variable-port-lengths-diameters-like-pipe-organ.html)

justinzane 28th December 2012 11:44 PM

Variable Port Lengths/Diameters? Like a pipe organ...
 
I've notice that just about every ported design that I've stumbled across uses one or more ports with the same tuning. I'm wondering if someone can explain why no one used multiple ports with different tunings in the same design. [For example, one 1x diameter port at 55Hz, one 3x diameter port at 27.5Hz.] I'm also struggling to find equations to model this in either Python/SciPy or Octave. Thanks.

prairieboy 29th December 2012 12:41 AM

Oh great, another opportunity to show my age ...

This has been done (there is nothing new under the sun). My apologies if I have identified the wrong speaker to have used this .... but I know it's been done and I'm quite sure it was the Hegeman model one speaker (which was an omni) which used multiple port tuning. The manufacturing of the speakers transferred to Don Morrison in Toronto, Ontario (Morrison Audio fine audio speakers and electronics.) who on his early models continued the use of multiple 1/4 wave tuned paths.

No, never heard them, can't comment on the result.

justinzane 29th December 2012 04:11 AM

Thanks for the reference. At least someone else had the idea. Hope someone can chime in with either experience with the end product or the math to figure out whether it works or sucks.

Don Hills 29th December 2012 05:50 AM

For classic ported enclosures, the resonant frequency is set by the volume of the enclosure versus the mass/velocity of air in the port(s). Having two ports of differing sizes results in a single resonance set by the combined port sizes. For example, an enclosure with a single port 15 sq cm area will have the same resonant frequency as one with two ports, one 10 sq cm and one 5 sq cm.

In the case of quarter-wave pipes or multiple coupled enclosures, the situation is less clear cut. For example, the Weems Double Chamber Reflex has two resonant frequencies, one an octave above the other. It's very much a case of modelling your ideas and see. I recommend AkAbak as a modeller. There are several threads here which discuss how to use it.

benb 29th December 2012 05:54 AM

I'm reminded of this thread from a while back (I had a little trouble finding it, I recalled it as a 1-driver-per-octave 10-way, but it's not quite that ridiculous):
http://www.diyaudio.com/forums/multi...udspeaker.html
My thoughts on making a "wider" resonance are in post #9 and there's an interesting port shown in post #17.

justinzane 29th December 2012 03:05 PM

@DonHills: Thanks. I use a FOSS software stack, so AkAbak and the other Windows based modellers are out. I'm using gspeakers + Python/SciPy + Octave, so I need the actual equations. Thinking about it, it seems like the coupled internal sub-enclosures would be necessary. Bummer.

justinzane 29th December 2012 03:19 PM

@benb: Excellent reference. I'm wondering about exactly what you refered to in post 9.

head_unit 29th December 2012 03:50 PM

Quote:

Originally Posted by Don Hills (Post 3302738)
For classic ported enclosures, the resonant frequency is set by the volume of the enclosure versus the mass/velocity of air in the port(s). Having two ports of differing sizes results in a single resonance set by the combined port sizes. For example, an enclosure with a single port 15 sq cm area will have the same resonant frequency as one with two ports, one 10 sq cm and one 5 sq cm...

Yes. Yes? Hmmm.

That's Small's treatment from a kind of lumped parameter point of view. But Dick himself would say his derivations are approximations. Thinking about it some more, it seems to me there should be SOME difference, even if tiny.

From a practical point of view, two things come to mind:
- If the ports are different lengths/sizes, they will certainly have different internal resonances (i.e. the pipe resonance).
- With the ports at different spots in the enclosure, the acoustic excitation is at least a tiny bit difference. "Theory" says not, but as I just noted that theory is an approximation, and not actual acoustics.
- The port summation won't quite follow the simple theory as the end corrections may be different (which is a bit like the point just above).

I think multiple ports are always the same simply because it's easier to calculate.

justinzane 29th December 2012 07:46 PM

@head_unit: That is what I was thinking. I'm a Python (language) fanatic and I've started porting gspeakers, but my understanding of the electrical and fluid dynamics aspects of speaker design are way to limited. I seriously need guidance on where to look to develop the model[s] needed to analyze this kind of design. If anyone could point me to some relevant resources, I'd be much indebted.

sreten 29th December 2012 08:09 PM

Hi,

Its been done, by the stupid and marketing who don't understand the physics.

Take a speaker tuned to 50Hz by one port and 25Hz by a much bigger port.

The 50Hz port depressurises the cabinet below 50Hz so the 25Hz port has
next to no effect and all it does it take up cabinet volume you could use
to tune the speaker to a somewhat overall lower alignment.

The higher tuned port prevents the lower tuned one working to any effect.

That is why, apart from clueless marketing claims/designs, you never see it.

If it worked, which it doesn't, you'd obviously see it a lot in designs.
(And be a feature in all speaker simulators, which it clearly isn't.)

rgds, sreten.

The free Basta! manual covers speaker modelling quite well and
you could add a further low tuned port to the model to see how
useless the idea is in theory and therefore in practise.
The higher tuned port "shorts" the lower tuned port.


All times are GMT. The time now is 09:38 AM.


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