The one source I've found says for a given freq, if boundary is within 1/8 wavelength, then driver is "officially" boundary-loaded at that freq.
Have never read anything definitive, however. Is this rule true?
Have never read anything definitive, however. Is this rule true?
Hi,
at 1/8 wavelength its certainly true but not that helful.
The issue is at what wavelength the transition takes place,
but this is complicated by the inevitable reinforcements and
cancellations through the transition region.
🙂 /Sreten.
at 1/8 wavelength its certainly true but not that helful.
The issue is at what wavelength the transition takes place,
but this is complicated by the inevitable reinforcements and
cancellations through the transition region.
🙂 /Sreten.
I'm designing an active two-way system with full range at ear level and 12" (sub) woofer floor-loaded. I'd like to choose a xover freq (100-200Hz) so that the woofer is "officially" floor-loaded at all its freqs, and the fullrange up top is not floor-loaded at any of its freqs.sreten said:
I'm sort of honing in on 120 to 140Hz as a good place.
Maximal boundary loading with respect to sound pressure level gain occurs within the length of the respective freq..
Here is a Java calculator:
http://www.sengpielaudio.com/calculator-wavelength.htm
Here are some other things to think about:
1. room gain (also known as the "Collom's Curve")
2. floor bounce (also known as the "Allison Effect")
3. baffle loss
4. Acoustic center as it relates to imaging.
If I were you I'd be more concerned with the last 3, and unfortunetly problems 2 and 4 are in-opposite to each other. In otherwords imaging cues extend down to the lower mids and the acoustic center of the driver should be well away from a boundry (like the floor) to achieve good imaging, but elevation off of the floor creates a reverse phase reflection from the floor that causes a degree of cancelation and a resulting loss in spl.
Of course then you "add-in" the loss in spl due to baffle loss and you could experience a substantial dip in the midrange's response (with this loss and floor bounce). Example (here is a standmounted speaker and subwoofer with a bit of baffle-step compensation, though not quite enough - and still it has a fairly substantial "suck-out" because of the floor bounce):
http://www.enjoythemusic.com/magazine/equipment/0905/escalantedesign.htm
Here is a Java calculator:
http://www.sengpielaudio.com/calculator-wavelength.htm
Here are some other things to think about:
1. room gain (also known as the "Collom's Curve")
2. floor bounce (also known as the "Allison Effect")
3. baffle loss
4. Acoustic center as it relates to imaging.
If I were you I'd be more concerned with the last 3, and unfortunetly problems 2 and 4 are in-opposite to each other. In otherwords imaging cues extend down to the lower mids and the acoustic center of the driver should be well away from a boundry (like the floor) to achieve good imaging, but elevation off of the floor creates a reverse phase reflection from the floor that causes a degree of cancelation and a resulting loss in spl.
Of course then you "add-in" the loss in spl due to baffle loss and you could experience a substantial dip in the midrange's response (with this loss and floor bounce). Example (here is a standmounted speaker and subwoofer with a bit of baffle-step compensation, though not quite enough - and still it has a fairly substantial "suck-out" because of the floor bounce):
http://www.enjoythemusic.com/magazine/equipment/0905/escalantedesign.htm
The plan is trapezoidal "wall-hugging" enclosures. I'd mount in wall if feasible. So by design, baffle loss and diffraction should not be a problem.ScottG said:
Now I'm thinking I should just cross as low as the mains will go, which is about 100 to 120 Hz, and place the bass drivers as low as possible in the enclosure. I was toying with the idea of giving the bass drivers a little more of the spectrum for added "umph", but am seeing it might cause more problems than solve. The subs will already have to cover at least two octaves.
Actually, now am thinking of building the subs and mains as "matched pairs". Then I could switch between woofers floor-loaded, or close to mains, just by flipping over the enclosure (would have to make feet removable). And of course could toy with xover freq, although it won't be a real-time adjustment.
Both mains and subs, btw, will be in roughly .6Q sealed enclosures.
Dumbass said:
Oh.. well yes that's quite different.. and your plan "sounds" good to me. Just make sure you have the ability to adjust the phase of the subwoofer at crossover (i.e. a variable phase adjustment as opposed to a simple 180 deg. switch.)
Note thought that the "allison effect" still holds true unless your loudspeaker operates as an "infinite line" line source at that freq..
Am not sure if the xover I have in mind (the basic active marchand kit) allows continuous phase adjustment. (I guess that's the great advantage of using digital xover.) The 4th order L-R is apparently "phase coherent" but that says nothing about how the drivers react at xover freq. Xover hopefully low enough where phasing issues aren't "crucial".ScottG said:
Allison effect . . . hopefully will minimize it by (1) putting a rug and thick rug pad between listening seats and speakers and (2) keeping seats and mains relatively low, i.e. speakers and ear level around 30"-32" rather than typical 40". (Bean bags? 😀 )
- Status
- Not open for further replies.