Isosceles ( A speaker project)

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

After some studying I see it now. You have moved the resonance modes above the driver bandpass and created a longer return path to weaken reflections.

Question. Would you use Accuton drivers again? I guess what I am really asking is was the end result worth the effort?
I grew up on beryllium drivers and I thought the Accuton would be a move forward. I do like the speed of the hard cones.
 
The size of a cell as you describe is the area between two panels?

It looks like the diode in that configuration would be more effective at reducing the standing waves in the vertical direction.
Or does the effect of the cells breaking the waves work in the horizontal directions as well.
 
crazyhub said:
using damping material thus reducing efficiency thus reducing dynamic range.


Hi,

I'm sorry but IMO none of the above necessarily follows, if anything
correct damping of resonances increases dynamic range in terms of
the mush floor to actual levels, and the alleged "reduced efficiency"
is poor design, it come from stuffing increasing effective volume, in
a proper design the effective volume is the wanted volume, in real
terms the box should be smaller than the undamped box would be.

e.g. the most "efficient" alignment is Chebychev equiripple.
No-one argues that this is better than the more common and less
efficient alignments, and more to the point the higher damped
alignments have higher mid-band reference efficiency.

:)/sreten.
 
sreten said:

Hi,

I'm sorry but IMO none of the above necessarily follows, if anything
correct damping of resonances increases dynamic range in terms of
the mush floor to actual levels, and the alleged "reduced efficiency"
is poor design, it come from stuffing increasing effective volume, in
a proper design the effective volume is the wanted volume, in real
terms the box should be smaller than the undamped box would be.

e.g. the most "efficient" alignment is Chebychev equiripple.
No-one argues that this is better than the more common and less
efficient alignments, and more to the point the higher damped
alignments have higher mid-band reference efficiency.

:)/sreten.

Hi sreten,

I measured the effects of stuffing a box to determine the maximum quantity before the stuffing added volume. It is easy to observe with near field measurements and increasing the stuffing in measured increments.

What I encountered was that stuffing the box to maxmize apperant volume and reduce the midbass resonace reduced the overall SPL output around a db or so.

If damping absorbs sound energy and converts it to heat then a reduction in SPL follows.
 
ttruman said:
Crazyhub ,

After some studying I see it now. You have moved the resonance modes above the driver bandpass and created a longer return path to weaken reflections.

Question. Would you use Accuton drivers again? I guess what I am really asking is was the end result worth the effort?
I grew up on beryllium drivers and I thought the Accuton would be a move forward. I do like the speed of the hard cones.
The photos I posted are from the speaker I'm currently working on, filter voicing stage these last days.
I use C220NT6 (underhung Titanium VC, neodymium magnet) up to approx 800hz, C79 up to 2700hz and an Utopia beryllium tweeter.
3-way config, bass //, mid+tweeter series.
Vs the standard C220, the NT6 gives an astonishingly tighter sound and better impacts, transients...in my cabs...
Whit this tweeter it's hard to fing a good mate (oneness) with the C79 but I'm on this way, almost day after day: higher orders work best...
So you can see that me too I like hard cones more than anything else, particularly Accuton ones.
 
ttruman said:
The size of a cell as you describe is the area between two panels?

It looks like the diode in that configuration would be more effective at reducing the standing waves in the vertical direction.
Or does the effect of the cells breaking the waves work in the horizontal directions as well.
Yes, for the size of a cell.
Yes, the shape of the diodes network in my current speaker does indeed act mainly in the vertical (top to bottom) direction with the main purpose being to reduce the lenght of the wave; but this is because I needed only 18liters load for the bass driver, thus creating a tower.
But the diodes also do act in the front to back direction because no wave has the same distance from the opening of a cell to the first rebounce on the rear wall.
Sure, the more reflexions, the more efficient the dampening effect but we have to deal with WAF sizes, isn't it?
 
I have run some ripple tank simulations on various enclosures shapes including the isosceles. Sreten was dead on in that the resulting standing waves are quite like a rectangle box.

--

Crazyhub
Interestingly modeling your diodes in the ripple tank resulted in a dramatic improvement. Using just the "startburst" pattern directly behind the driver resulted in very similar results.

If it's ok with you I would like to try the diodes in my next design.

Let me know how you make out with the latest WIP. Very interested in the outcome. You’re working with my two favorite driver materials.

--

Based on these findings I don't really see a reason to pursue the Isosceles design further other than the cool factor of the shape and the design challenge. It’s all about the sound after all.


Thanks all for the advice.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.