The sorbothane gasket approach is entirely different from the direction that the OP discusses and most other DIYers typically take, which is to keep adding various forms of internal bracing.
There seems to be an opinion by some experts who have experimented with the bracing approach, Siegfried Linkwitz being one of them, that it doesn't do much more than move the resonances to higher frequencies. Doesn't really get rid of them by any appreciable amount.
There seems to be an opinion by some experts who have experimented with the bracing approach, Siegfried Linkwitz being one of them, that it doesn't do much more than move the resonances to higher frequencies. Doesn't really get rid of them by any appreciable amount.
Last edited:
If you stuck an accelerometer to a speaker cabinet wall ( and you played a sine wave through the speaker ) and the results were a sine wave, would that indicate that the vibration was caused be alternating air pressure, were as if the result was a none sine wave the vibration would be caused be mechanical transmission?
From Linkwitz's website, "Some Conclusions" page:
"When building boxes from 3/4 inch thick wood, then the un-braced areas should be less than 4 inch squares to obtain high stiffness and to push panel resonances into the kHz region and where they can be decoupled from the driver's structure and airborne vibration."
Last edited:
Exactly the point. Adding braces doesn't get rid of the resonances, it simply moves them to higher frequencies. They are still there and you will hear them in the room.
What SL is saying is that it will decouple them from the "driver's structure" so that they will not feedback into it and cause even more distortions.
So if you had a driver in a small enclosure, inside a larger enclosure and decoupled from it, this would greatly reduce distortion, a box in a box speaker. With tha added bonus that the sound radiating from the rear of the cone would have two walls ( like double glazing ), greatly reducing another source of cabinet vibration.
Probably would help with reducing sound into the room from the smaller box wall resonances. But don't forget about the energy that is bouncing around inside the box and comes out the front through the very thin speaker's cone. Only one way to get rid of that energy. Get rid of the box. And that solves both problems.
Additional bracing does move the resonances to higher frequencies, but not 'simply'. Magnitude reduction occurs when wood panel resonance frequencies exceed around 500 Hz, and then again above 800 Hz. I have measured, and have seen similar results in articles written several years ago. One problem to avoid is with baffle bracing spans of say around 10" x 6", numerous panels and bracing ringing at around 400 Hz, which is not good. Better to have the resonance frequencies spread out.
Along with the baffle I created to kill mechanically induced cabinet/baffle vibrations(Same thread), I also did this to kill internal acoustical reflections. This does limit it's use as a reproducer of lower frequencies, but for lower midrange and up, It's quite effective.
Midrange experiment with Auto sound panel deadener and high density acoustic felt..
Midrange experiment with Auto sound panel deadener and high density acoustic felt..
Last edited:
That's not what I found when I tested this recently here and discussion thread here. Here is an empty box vs the same box with a simple oak brace added:
I also tried a CLD brace:
On the topic of panel materials, a simple CLD of 1/4" MDF glued together with Weicon 310M Flex Classic was a huge improvement over 3/4" plywood from a big box store:
It seems that your first graph proves the point that Linkwitz made. The oak brace moved the resonance peak from 600 Hz to a higher 950 Hz, and only brought it down by about 6 db. Not really enough to get rid of it or make a significant difference.
With the CLD braces the resonance peak moves to about the same frequency and is now down about 12db. Certainly better, but not gone.
The results with the CLD panel box construction are a big improvement, although there is now a new peak of some concern at 300 Hz that is not there with conventional construction. Just not sure how many people here are going to go to all that much trouble to build a DIY speaker with CLD.
Last edited:
But what if the mid/bass box was so short front to back that a sound wave would be to long to bounce of the back? Kill three birds with one stone.
I built a prototype box few days ago with very simple technique, not sure how effective my CLD experiment is other than with knuckle test there is very noticeable difference. I used similar elastic sealant/glue as the mentioned Weicon Classic (local brand Kiilto MASA). Built the box as usual, from 15mm baltic bitch ply and then glued some scrap pieces of 6mm ply I had laying around inside the thicker panels before assembly. Was a little bit more work, but didn't ave to tape measure anything, made a bit mess but it doesn't affect the looks outside of the box etc. It is quick and easy to glue the scrap panels comparing how much work it is to put a box together in the first place. One tube of adhesive was enough for the ~90 litre box). So, can be a quickie and it makes a difference, not sure if for better but I think it could be worth it 
Last edited:
I totally agree that bracing moves the resonances to higher frequencies. In fact, that's the whole idea behind bracing! The point is that it's much easier to attenuate/dampen resonances at higher frequencies. This is why bracing is useful to reduce panel resonances overall. See here.
Excuse me? Down 6dB, a lot of noise engineers would sign for that in the blind.
classicalfan> how can you look at that plot and see a modest improvement? You seem to only want to recognize a narrow band peak as "only" 6 dB down, what about the much wider band of resonances that are down 15 dB now? I'm sorry but you are seeing what you want to see, not what most unbiased observers would see in the data.