strange resonance in loudspreakerbox

[keyser]

I've recently built a pair of Scan-Speak Reference Monitors.
I've deviated a bit from the original design: the back and front are both 50 mm (2 inches) thick and I used cross-bracing.
The box feels rock solid. However, with a tone-sweep I noticed an audible box resonance around 300 hz. The enclosure can be felt to resonate around that specific frequency! This resonance causes a buzzing or rattling sound. To other frequencies the box is practically dead.
I've tried damping materials of different kinds between the driver basket and the baffle, yet the problem remained and is still unsolved.
Has any of you an idea of what might cause this increasingly annoying resonance?

[keyser]

a few pictures

bracing and box

bracing and box 2

finished box

finished box 2

woofer hole with a look inside

[kevinkr]

I assume that the front and back panels are separate pieces of plywood or mdf that you glued and screwed together?

Have you been able to localize the physical location of the resonance at all?

You might have coincidentally chosen a set of dimensions where a panel naturally resonates at 300Hz? Is it a very high q resonance or is there a range for frequencies around 300Hz that also trigger it, if very high q it might or might not be as big an issue.

I noticed that you used biscuit joinery, my understanding is that this technique can give rise to really bad buzzes if all surfaces were not glued and tightly clamped. If you did not glue along all the edges of the panels all may not be lost, you may be able to apply hot melt along the edges of all adjoining panels and this should help.

I assume the edges of the braces that touch are glued everywhere.

Glue and screws are the preferred assembly technique to minimize buzzing at joints.

Hope this is helpful to you.

[tinitus]

Nice box, shouldnt make such trouble - I wonder if that interior "foam" destroys box volume ??

Have you tested both enclosures ??

[keyser]

Originally posted by tinitus
Nice box, shouldnt make such trouble - I wonder if that interior "foam" destroys box volume ??

Have you tested both enclosures ??

that specific foam (called PriTex) is also used in the original version. Shouldn't be much of a problem.

[redunzelizer]

I'd also recommend getting a stethoscope, but probably not an expensive type like this one: http://www.stethoscope.com/index.cfm...&ProductID=279

You then also can use it on turntables, heatsinks, cd-players, general audio-equipment housings, etc. etc. From my experience: In most cases you will find out that you will get a lot to "do something" about unwanted "sound", both emitted, and induced.

regards

[keyser]

Originally posted by kevinkr
I assume that the front and back panels are separate pieces of plywood or mdf that you glued and screwed together?

The double 1” MDF panels are indeed glued together, with glue spread over the full surface.

Originally posted by kevinkr
Have you been able to localize the physical location of the resonance at all?

The full cabinet can be felt vibrating. However, it is strongest near the back of the woofer. Note that just here is the biggest ‘not-braced’ surface of the box.

Originally posted by kevinkr
You might have coincidentally chosen a set of dimensions where a panel naturally resonates at 300Hz? Is it a very high q resonance or is there a range for frequencies around 300Hz that also trigger it, if very high q it might or might not be as big an issue.

As you can see on the pictures, this speaker is very heavily braced considering its size. The resonance is in al small range just above 300 hz. From around 300 to 300 hz. This can be considered pretty high Q, I guess.

Originally posted by kevinkr
I noticed that you used biscuit joinery, my understanding is that this technique can give rise to really bad buzzes if all surfaces were not glued and tightly clamped. If you did not glue along all the edges of the panels all may not be lost, you may be able to apply hot melt along the edges of all adjoining panels and this should help.

I assume the edges of the braces that touch are glued everywhere.

Glue and screws are the preferred assembly technique to minimize buzzing at joints.

Hope this is helpful to you.

I think every biscuit joint is glued well enough as far as not to make it rattle. All braces are glued on all sides. I have used no screws.

I simply made this box as rigid as possible. Although the rest of the sound is very good, this rattle came as a big disappointment.

Thanks for the help so far. Still got any ideas?

[mazurek]

just a thought, you got no feet on these, what surface do you have these on when you are testing them? It appears your boxes are ridiculously solid, I would look at interactions with the environment.

[owdi]

If you put the entire speaker in your lap while you test, does it still only happen at 300hz? Try hugging it close to your chest (yes I know this will look silly), does it still vibrate at 300hz?

If it doesn't, I would suspect your speaker stands. Some additional weight should drive the resonance down and reduce the magnitude. Maybe you have found the first legitimate use for those audiophile slabs of granite you are supposed to place on top of your speakers.

Dan

[keyser]

Originally posted by keyser
As you can see on the pictures, this speaker is very heavily braced considering its size. The resonance is in al small range just above 300 hz. From around 300 to 300 hz. This can be considered pretty high Q, I guess.

Of course this should not have been from 300 to 300 hz. It is actually about 300 hz to 330 hz.

My speakers stand on pretty solid speaker stands. I’ve already tried the speakers while holding them in my arms. That doesn’t help.
Yesterday night I screwed the woofer of one of the speakers out, and put it on a pillow. This time it sounded perfectly clean. After that I lay the woofer back into the baffle, but didn’t put the screws in. Strangely, the resonance had not returned!
Somehow the screws must be passing energy from the woofer basket to the box. I still don’t really know what parts exactly cause the resonance inside the box, but decoupling the woofer basket from the box certainly helps.
Gluing might be an option, but is a rather permanent one. Has any of you a better idea?