cabinet vibrations

[Fitts]

I was wondering if anyone has any advice on the following: If a cabinet has a panel that resonates (but does not rattle) well above the crossover frequency of the cabinet does one really need to worry about sound deadening that panel further?

Thanks,

[GM]

If there's no signal strong enough to excite it.......... That said, cabs that have eigenmodes extending down near/at the cab's intended BW can be a problem, so if this is the case, then on those walls some additional damping is advised.

GM

[Todd W. White]

I agree - the panel CAN start wiggling if excited by upper harmonics generated by fundamental frequencies in the lower bands.

My preference is to brace it well, rather than add damping material only, as the brace stops the vibration, while the damping tries merely to absorb it.

[acre]

Hi Dave!

Another thought: Suppose the offending frequency is around 1,500 Hz or so; well beyond what the driver can do in the cabinet. I would think that in theory, now there is an "orphan" frequency whose phase could be a problem for the HF driver. If that frequency is 100% in phase, then no problem but if it isn't, you have another frequency that could mess up the sound of the horn instead of the LF unit. I agree, brace the daylights out of it if you can. I don't know if that thought is right but it would make sense from a physics point of view......Lee

[Fitts]

Thanks everyone, I had forgotten about the 2nd, 3rd and higher harmonics. The horn curves of my A7 cabinets resonate at a pretty high frequency but now I see why I need to figure out a good way to dampen them.

Regards,

[GM]

>My preference is to brace it well, rather than add damping material only, as the brace stops the vibration, while the damping tries merely to absorb it.

====

While I agree that in theory a cab can't be too rigid, bracing doesn't stop vibration per se, but raises the panel's resonance. Since acoustic energy falls with increasing frequency, at some point of adding bracing the mass and damping properties of this system finally damps it down to insignificant levels, but depending on where on the frequency curve you start, you could wind up with a wooden version of a Birdcage Maserati's frame.

So I take a balanced view to achieving a well damped/rigid cab by starting with a fairly rigid material such as 3/4" 13 ply Baltic Birch and brace it enough that there's no large unsupported panels and all six sides are tied together at an intersecting point so it can't 'breathe', then if there's still some minor resonances, use damping material to finish them off.

====

>Another thought: Suppose the offending frequency is around 1,500 Hz or so; well beyond what the driver can do in the cabinet. I would think that in theory, now there is an "orphan" frequency whose phase could be a problem for the HF driver.

>If that frequency is 100% in phase, then no problem but if it isn't, you have another frequency that could mess up the sound of the horn instead of the LF unit. I agree, brace the daylights out of it if you can. I don't know if that thought is right but it would make sense from a physics point of view.

====

Actually, then whether in phase or not it would be an issue since it will sum either positively or negatively as a distinct separate source offset in time by some amount. This is what the 825's flares do. The 828's flares are made of a lower resonance material and are sealed off, so 'drum' at some lower multiple of the horn's frequency. I haven't measured it, so not sure what it is.

On a related note, consider a long vent on a sub. It has pipe harmonics that unless well damped will comb filter with the mains. And folks wonder why they can often locate the sub source off in the corner though it's only reproducing WLs that are so large they swamp the distance from the corner to the listening position.

Ditto the large vents on the A7. It comb filters with rest of the speaker's output to ~1kHz before it finally peters out. Modeling it in MJK's ported worksheet, it accurately predicts all the unevenness of its measured response.

GM