I guess I would recommend you discuss that with Vance Dickason. His book states that raising the resonance is better.
How does placing a naked panel right behind the driver is a good thing ? Hell im going nuts deciding what and how to brace 
How can NOT bracing be a good thing ? most commercial speakers I know off do have braces and the higher their cost the more well braced they are ( magico mini ... B&W 800 series etc... )
How can NOT bracing be a good thing ? most commercial speakers I know off do have braces and the higher their cost the more well braced they are ( magico mini ... B&W 800 series etc... )
That is very true. The tough part is getting the panel resonance so high up. Might have to incorporate some type of ultra HDF materials to build the box.
IF we built an ultra stiff midbass enclosure which rings at 2000Hz, and set the crossover point at 1500Hz 4th order, don't you think the panel resonances will be inaudible ?
Edit : I'm still new to the forum, and haven't got the quote right yet. Sorry for that.
Huh?
Poor analogy.
Cross bracing on a bridge is that way for load transfer, not rigidity. A load bearing structure that is too rigid would break apart under stress. It needs to flex in order to absorb the energy and not snap under pressure.
Ordinary box stuffing (fiberglass, acousta-stuff, etc.) does a much better job absorbing acoustic energy at higher frequencies than it does a lower frequencies.
If your box walls resonate at 100 Hz you will have a hell of a harder time taming that resonance than you would if it was 1 kHz.
It is wrong to look at this as simply individual parts. It is a system and the components act together to create a working system, that includes damping materials like fiberglass, wall treatments, foam, and so on.
Yes, and several responses seem to have a similar problem.
The primary emphasis should be on PANEL DAMPING (at least with regard to bracing). Not structural integrity, and not dominant mode freq. shifting.
Structural integrity is for the most part a moot point.
The emphasis on shifting modes is there so that you can shift them high enough in freq. so that:
1. They are easier to damp,
2. They interact less with the driver's frame (..and resulting operation of the driver).
Last edited:
This is another area where I am confused. If you run a brace from one wall to another, what keeps the two walls from resonating in the same direction?
In other words, the left wall moves right and the right wall also moves right because there is a brace connecting the two walls. It is much more difficult for the two walls to move either away from each other or toward each other, but they can still oscillate side-to-side.
What I did was run several braces along the surface of the cabinet walls to subdivide the walls into smaller panels. Then I tied those braces to the opposite wall with cross braces. The back wall has a cross brace that attached to the side walls' cross brace.
I also made a horizontal shelf brace higher up in the cabinet that tie all four walls together and another shelf that ties the top of the cabinet to the horizontal shelf.
The shelfs have large cutouts to open the box up internally and the midrange/tweeter enclosure is attached to the horizontal shelf.
So, my enclosure is a mix of braces and shelves for rigidity.
It's all about trade-offs - higher frequencies may be easier to damp, but higher frequencies are also more noticeable to the ear, so I personally would rather have an enclosure with a lot (relatively speaking) of low frequency resonance over an enclosure with a little high frequency resonance.
A braced cabinet might not resonate it's panels very easily, but a braced cabinet has the disadvantages of A) having standing waves reflecting off the bracing into the panels, and B) having standing waves reflecting off the bracing back into the driver's cone, and C) there's the real possibility that the bracing itself resonates, which would amplify points A) and B), and resonating braces would more than likely add to panel resonances at the brace/panel attachment points. And seeing as most bracing is rather small/short in relation to the panels of the enclosure, brace resonances would be of the more offensive higher frequency type.
So it would seem to me that any gains that are made by the stiffening of enclosure panels with braces are to a large degree nullified by the bracing itself.
I haven't measured it, but my intuition tells me that multiple layer damping sheets along with egg-crate foam will damp out the majority of standing waves before they even reach an enclosure's panels, and thus thick panels will be sufficient enough by themselves to resist resonating, and any small residual resonances that are left in the panels will be of low frequency in nature, and thus not offensive to the ear. And the added bonus is that there isn't any bracing to reflect standing waves into the panels or the driver cone(s).
A braced cabinet might not resonate it's panels very easily, but a braced cabinet has the disadvantages of A) having standing waves reflecting off the bracing into the panels, and B) having standing waves reflecting off the bracing back into the driver's cone, and C) there's the real possibility that the bracing itself resonates, which would amplify points A) and B), and resonating braces would more than likely add to panel resonances at the brace/panel attachment points. And seeing as most bracing is rather small/short in relation to the panels of the enclosure, brace resonances would be of the more offensive higher frequency type.
So it would seem to me that any gains that are made by the stiffening of enclosure panels with braces are to a large degree nullified by the bracing itself.
I haven't measured it, but my intuition tells me that multiple layer damping sheets along with egg-crate foam will damp out the majority of standing waves before they even reach an enclosure's panels, and thus thick panels will be sufficient enough by themselves to resist resonating, and any small residual resonances that are left in the panels will be of low frequency in nature, and thus not offensive to the ear. And the added bonus is that there isn't any bracing to reflect standing waves into the panels or the driver cone(s).
True, but I can't imagine such an asymmetrical force developing in a speaker cabinet. Front to back, perhaps, but then I'd suspect the baffle of being inadequate.
Really! And which of the adjusted parameters involved in raising the resonance of a panel suggests that to you?In fact it's the opposite, actually more energy is needed!
Can you give me a link(s) to support those claims? I would be interested in reading more about that and seeing the supporting data.
If you'll go back over my posts you'll see that I made it pretty clear that I'm stating opinion and preference, not fact.
Last edited:
That's okay.
You stated earlier that, "To my way of thinking bracing creates more problems than it solves."
You stated that was your theory. Surely you have some basis for that theory and that is what I am trying to understand; what makes that theory tick.
Far too much misinformation.
Basically the situation is very complex and just cannot be simplified to "right vs. wrong"
Damping is good, and stiffness is good, but too much of either is a waste. Lowering a panels resonance DOES NOT MEAN that there aren't any higher resonances! In fact the most total resonances with any given frequency band will be for that panel with the lowest resonance. That means that the resonance density is lowest for the stiffest panel - thats why we stiffen them. Damping is more effective at Hf, but then the acoustic coupling from the driver drops off quite a bit, BUT the structural coupling increases with frequency. However the structural driveing force from the driver drops at HFs. Then there is the radiation efficiency of the whole stryucture which is vastly more complicated than all of the rest. All in all its such a complex problem that it really cannot be tackled with simplistic discussions like this.
Bottom line - after vast amounts of experimentation with everything discussed here - Simple bracing works best and a "good" amount of damping of the acoustic interior. Well damped "joints" are more effective than panel damping. Going to extremes of any of this appears to yield ever decreasing effectiveness. Do whats reasonable and leave it at that.
You are confusing "yield strength" with "rigidity" - they are not the same thing. "Load transfer" and "rigidity" are the same thing. Something that is not rigid cannot transfer load while something that is rigid does. Neither case tells us IF the stress will exceed the yield strength of the material as that is an entirely different thing.
Basically the situation is very complex and just cannot be simplified to "right vs. wrong"
Damping is good, and stiffness is good, but too much of either is a waste. Lowering a panels resonance DOES NOT MEAN that there aren't any higher resonances! In fact the most total resonances with any given frequency band will be for that panel with the lowest resonance. That means that the resonance density is lowest for the stiffest panel - thats why we stiffen them. Damping is more effective at Hf, but then the acoustic coupling from the driver drops off quite a bit, BUT the structural coupling increases with frequency. However the structural driveing force from the driver drops at HFs. Then there is the radiation efficiency of the whole stryucture which is vastly more complicated than all of the rest. All in all its such a complex problem that it really cannot be tackled with simplistic discussions like this.
Bottom line - after vast amounts of experimentation with everything discussed here - Simple bracing works best and a "good" amount of damping of the acoustic interior. Well damped "joints" are more effective than panel damping. Going to extremes of any of this appears to yield ever decreasing effectiveness. Do whats reasonable and leave it at that.
You are confusing "yield strength" with "rigidity" - they are not the same thing. "Load transfer" and "rigidity" are the same thing. Something that is not rigid cannot transfer load while something that is rigid does. Neither case tells us IF the stress will exceed the yield strength of the material as that is an entirely different thing.
Last edited:
Here is a link where one individual ran some quantitate tests.
Loudspeaker construction
His results are interesting in that he found that bracing's main contribution was at 100 Hz and lower. Indeed, this is an important find. Consider that with average music, 50% of the music power is below 500 Hz.
This supports, in my mind, the idea of raising the cabinet resonances as high as possible. If you consider a resonance with a sharp Q and compare that resonance power at say 300 Hz and 1,000 Hz, with both resonant peaks of equal width, you can see that the lower resonance will occupy a wider percentage of the region from 20 Hz to 500 Hz (the region where 50% of the music's power resides) than the 1,000 Hz resonant peak, which has a much, much wider spectrum (500 Hz to 20 kHz).
If the resonant peak was arbitrarily 50 Hz wide, then the lower resonance would occupy 10% of the lower spectrum (0 to 500 Hz). Compare that same resonant peak to a spectrum from 500 Hz to 20 kHz and it is only .25%. Even if the resonant peak was 500 Hz wide it would only occupy 2.5% of the spectrum above 500 Hz.
I think this is one reason why we want to push the resonance points of the cabinet as high as possible. While ringing is undesirable, the narrower that spike the less intrusive it would be.
The above link also shows that a 50% fill of box stuffing effectively reduced internal panel resonances by 20 dB. Interestingly enough, the red and blue curves of the author's graph shows that 20 dB of attenuation occurs at the higher end of the frequency spectrum (above 500 Hz).
Again, this would support the idea that higher frequency resonances are easier to control than lower ones using box stuffing or even wall damping.
The author's take away seems to be that box bracing is not as important as box stuffing. The curves in the author's graphs seem to support this conclusion and it also demonstrates that there multiple successful paths to reducing box resonances. Again, there are commercial examples of these multitudes of methods and all seem to work successfully with marginal differences.
Loudspeaker construction
His results are interesting in that he found that bracing's main contribution was at 100 Hz and lower. Indeed, this is an important find. Consider that with average music, 50% of the music power is below 500 Hz.
This supports, in my mind, the idea of raising the cabinet resonances as high as possible. If you consider a resonance with a sharp Q and compare that resonance power at say 300 Hz and 1,000 Hz, with both resonant peaks of equal width, you can see that the lower resonance will occupy a wider percentage of the region from 20 Hz to 500 Hz (the region where 50% of the music's power resides) than the 1,000 Hz resonant peak, which has a much, much wider spectrum (500 Hz to 20 kHz).
If the resonant peak was arbitrarily 50 Hz wide, then the lower resonance would occupy 10% of the lower spectrum (0 to 500 Hz). Compare that same resonant peak to a spectrum from 500 Hz to 20 kHz and it is only .25%. Even if the resonant peak was 500 Hz wide it would only occupy 2.5% of the spectrum above 500 Hz.
I think this is one reason why we want to push the resonance points of the cabinet as high as possible. While ringing is undesirable, the narrower that spike the less intrusive it would be.
The above link also shows that a 50% fill of box stuffing effectively reduced internal panel resonances by 20 dB. Interestingly enough, the red and blue curves of the author's graph shows that 20 dB of attenuation occurs at the higher end of the frequency spectrum (above 500 Hz).
Again, this would support the idea that higher frequency resonances are easier to control than lower ones using box stuffing or even wall damping.
The author's take away seems to be that box bracing is not as important as box stuffing. The curves in the author's graphs seem to support this conclusion and it also demonstrates that there multiple successful paths to reducing box resonances. Again, there are commercial examples of these multitudes of methods and all seem to work successfully with marginal differences.
Last edited:
I thought I pretty much spelled out in my posts the reasoning behind my opinions in regard to bracing. If there's something specific that I could clarify for you I'd be happy to oblige.You stated earlier that, "To my way of thinking bracing creates more problems than it solves."
You stated that was your theory. Surely you have some basis for that theory and that is what I am trying to understand; what makes that theory tick.
The problem that I have with this link, and all tests that do vibration measurements on the panel (ala stereophile) is that not all vibration radiate the same. Some vibration modes radiate very efectively and others less effectively. It is the radiated sound at the listener that matters and NOT the vibration itself. When one actually measures the far field sound effect from various methods, the results are not at all in line with the vibration measurements themselves.
The most effective radiation will be the lower frequencies where all the walls are moving in the same direction - the monopole or breathing mode. This mode is a very efficient radiator and needs to be controlled. But the higher frequency modes all tend to cancel in the far field because there is always about the same total motion out as there is in. This is why the "cross brace" works so well - it restrains the breathing mode by tieing all the panels into a common center point which is rigidly held, by the mutually perpendicular braces.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.