Loading principle_waves propagation issues_

[crazyhub]

Hi all,
I'm planing to build a 2-way using a 8" + tweeter; no matter about the type and features of the drivers.
I just want to drastly reduce the inner box resonances which are (IMO) perhaps as or even more important than the driver specifications in a matter of clarity and accuracy above fs. I want to go without any stuffing (or the less possible) in the way to preserve the QMS and thus the dynamic of the woofer.
As I studied the waves propagation laws, I project to build a kind of spherical enclosure ended with a vent; both ends of the vent will have conical or exponential shapes. The inner end of the vent will be the ended shape of the enclosure (see the global shape of the enclosure as a drip of water), then the tube, then the conical or exponential outer end of the vent, as wide as the enclosure . The vent will be bellow the enclosure and will benefit from the floor loading. To enhance the wave propagation, I will place the woofer at the top of the enclosure, at least 45° backward leaned. Behind I will place the tweeter.
I do not want to intend to you to speak about the filter issues, but only about this loading idea.
Sorry for my bad english but I hope you understood the matter.
What do you think about?
Thanks for your future responses.

[KBK]

Peerless CSC217 (no shorting ring, that's the CSX217) works well as a two way in a 15 liter sealed box, one of the few cheap 8" drivers that do. Keep that as a thought, as truthfully, VERY few 8" drivers do two-ways well. At all.

If going sealed, keep the Q up, 0.8 or higher. With the Peerless in a sealed 15 liter box, the Q ends up being near 1, which helps with midbass and up resolution. Just my 0.02 cents.

Oh. I see. Read the rest of your post.

Open baffle 'ish' design or ported? That's an incredibly tough task. The (almost) inherenetly 'higher motional mass' of what is in a 'port-designed' or 'free air' driver in the 8 inch range does not at all agree with what is needed to get midrange resolution out of a 8" driver. See my earlier bits.

[sreten]

Originally posted by crazyhub

I just want to drastly reduce the inner box resonances which are (IMO) perhaps as or even more important than the driver specifications in a matter of clarity and accuracy above fs. I want to go without any stuffing (or the less possible) in the way to preserve the QMS and thus the dynamic of the woofer.

Hmmm.......

talk about creating your own problems and then complicatedly trying to fix them.

The "dynamics" of a woofer are loosely related to Qms, a better way
would be to describe Qms as "hysteresis" of the driver, and as far
as I know box stuffing does not affect this behaviour to any degree.

Angling the midbass unit 45° is asking for trouble at the top of the midband.

/sreten.

[crazyhub]

Hi,
Please,
KBK:

The (almost) inherenetly 'higher motional mass' of what is in a 'port-designed' or 'free air' driver in the 8 inch range does not at all agree with what is needed to get midrange resolution out of a 8" driver.

What do you mean by 'higher motional mass'? mms of the driver? air mass inner the enclosure? the Q of the mass/spring system?

sreten:

The "dynamics" of a woofer are loosely related to Qms, a better way would be to describe Qms as "hysteresis" of the driver, and as far as I know box stuffing does not affect this behaviour to any degree.

I agree that other specs mainly affect the dynamics of a woofer (i.e. inductance, BL...) and I agree that you can describe Qms as "hysteresis". But IMO, we can also see 2 issues:
- Qms is a data in a free air environment, obviously a free field environment doesn't introduce any dampening to the mechanical moving parts of the driver. So the hysteresis is fixed.
- Dampening stuff behind the driver and walls of the enclosure (thus reflexions) do not act linearly because frequencies aren't affected at the same degree. So hysteresis isn't fixed and thus the moving parts do not work linearly in time and space. Thus dynamics aren't equaly reproducted in the bandwidth. IYO, am I wrong?

KBK:

If going sealed, keep the Q up, 0.8 or higher. With the Peerless in a sealed 15 liter box, the Q ends up being near 1, which helps with midbass and up resolution.

Regarding that Q value is related to the driver/enclosure load at the resonance frequency, I don't understand why it does affect the midbass and up resolution. Please explain.

sreten:

Angling the midbass unit 45° is asking for trouble at the top of the midband.

Yes, I know, but it depends mainly of the woofer type; I will use a 8" with an extanded top range (bump) and a relative high slope filter near 2khz.

I would add: What I want to try is a kind of the "B&W sphere ended with a tapered tube" but replace the tapered tube by a vent, allowing to get the bass-reflex properties with the less harmonic resonances inside the box. I believe that if I choose the correct shapes, this will act as a kind of waves guide, allowing the waves energy to smoothly escape from the enclosure.

Thank you for your next replies.

[pocketman]

have you had any luck with designing the B&W sphere/tube design with a vent?
Dean

[Few]

I'm not sure "hysteresis" is really the right term for what's being discussed here. Hysteresis implies a "history" effect. For example, piezoelectric devices exhibit hysteresis in the sense that if you apply zero volts to a piezoelectric device and then increase the voltage to 1 volt, the final displacement is not the same as if you had started at 2 volts and then decreased the voltage to 1 volt. The previous history (0 volts vs. 2 volts) affects the subsequent behavior. Is that what you mean to be talking about? You can damp a vibration (lower its Q) without introducing hysteresis. In fact I would think acoustically absorptive materials in an enclosure would do just that. I'm not trying to be picky about the language, I just want to be sure I understand which effects you're trying to address with your design idea.

Few

[crazyhub]

Hi pocketman and others,

I didn't build the ported sphere mainly because this design doesn't allow a good impedance match between inner and outside of the enclosure...so I believe this principle will not work well.

These last months I was occupied with other things, so my 2-way project is for the moment at the choice of the drivers.

BUT, I designed an acoustical diodes network that, I believe, will work fine. The principle is to build several diodes with the same shape but with different volumes, each behind and connected to the others in a network surrounding the driver cone.
This network finaly will be ended with a small amount of dampening stuff and is adapted to sealed or ported enclosures. The critical point is to find a shape allowing the wave to enter each diode without returning to the driver or to the previous one...not so difficult if respecting impedance by optimising the angles and sections.
Different volumes of each diode will smooth the resonance of each other.
I'm not really sure but the main resonance of the entire enclosure will probably remain the same than a classical enclosure because of same volume of air, but with a lowered spl level.

The main reason of this network is to drastly reduce the amount of energy returning to the driver and secondary cancel the enclosure main resonance(s).
Your thoughts please.

Some raw drawings:

[pocketman]

crazyhub

The accoustic diode system looks even more complicated than the sphere/tube. I found a site where someone has tried it

http://home.wanadoo.nl/dezaire/ultiem/ultimate.htm

I think I am going to give it a shot, I just wished I could find more info on the theory of it

Dean

[crazyhub]

I just wished I could find more info on the theory of it

the theory of the sphere tube is very simple indeed, the opposite of a horn: progressively enhance the impedance of the wave travel by 1) optimising the energy transfer from cone to tube, 2) reduce the energy inside the tube by reducing its section and 3) finaly damp the rest of the wave with an amount of stuffing. The challenge is to got a smooth travel without break-up, reflexions and on. Conical shapes are perfect in this way.
I guess the longer and the more stuffed the tapered tube, the best dampening result.

[crazyhub]

Hi all,

Finally completed!

Don't see it as a labyrinthus, it is not...much more a real acoustical diodes network.
Vs my previous design, one optimisation: diodes are no more closed allowing no resonances inner each of them; standing waves should only appear at very small lengths and will be killed with very slight damping material, mainly at the top and bottom where vent and filter will be placed.
The main working principle is to propagate waves from the cone to the extremities of the enclosure. No angles more than 45° is the key.

This enclosure will be fitted with two mid-bass and a tweeter in a vented TMM 2.5-way. The drivers:
http://www.lautsprecher-vertrieb.de/...hauptrubrik=24
http://www.lautsprecher-vertrieb.de/...&hauptrubrik=8
For those who are interested, see the mid-bass parameters, fit them in Unibox...amazing driver! as for the tweeter...Hope a little bomb,