Everything resonates. Surely it's a question of where you put these resonances.
I believe it's far too easy to muffle the midrange with MDF or medite. Many specialist manufacturers purposely avoid the use of MDF for their cabinets, or at the least prefer some form of MDF composites.
Harbeth (along with Spendor?) is the exception, but then their entire cabinet construction techniques are also an exception.
|
@georgehifi your post about the surfboard hits the nail on the head. It's easy to think that stiffer/stronger/more solid must be better, yet all the strength in the world is no good if the cabinet becomes a soundboard at a critical frequency, especially in the midrange.
It's possible for loudspeaker cabinet materials to become almost transparent to the sounds coming from within the cabinet at certain frequencies again.
I think experiments have demonstrated that MDF untreated is more likely to act in this fashion.
And then there's the issue of a cabinet being too rigid and causing back pressure through the driver materials themselves.
Hugely complicated business is cabinet construction. Open Baffle gets around most of them but has its own issues. |
"It helps to hear, literally, what is going on, by listening to the cabinet’s walls through a stethoscope, playing music and test tones. You find two things:
- Flexing of the cabinet walls allows low bass to come through. Out in your room, this adds in phase to the direct sounds coming from your woofer, making the overall presentation warmer, and adding bass ’ambience’ below 80 Hz. The bass is less tight, less defined, of course. Probably makes the hifi sound ’better’ at soft volumes.
For low-bass flexing, braces help, granite helps, thick materials help, thin plywoods do not. Thin carbon fiber does not. Solid hardwood will split given time. Think here about maximizing panel ’stiffness’ or rigidity, not its ’strength’. Cement, concrete? Sure! FYI, paint the inside of wood cabinets with thinned-out wood glue-- as the surfaces of MDF and plywoods are porous, absorbing bass pressure whenever the woofer fires off.
- The other stethoscope discovery is that a cabinet lets voice range sounds, in the 200 to 300 Hz range, come right through. This makes for ’scratchy’ sounds in wood boxes and for ringing sounds in granite and metal. These vibrations do not come from the SPL inside the cabinet (unless there is a large, undamped (loud) standing wave inside). Those 200-300Hz vibrations come from the direct excitation of the cabinet material via the screws mounting ’that’ driver. When you loosen all its screws, all of a sudden, the walls go silent!
One fix is to use rubber-mounted screws, but this makes bass impacts ’rubbery’. KEF and others tried this in the 80’s, giving it up after sales tanked from all that loose bass. The driver could be mounted to a regular inner cabinet with a vibration-isolated cabinet wrapped around it. Unfortunately, this leaves those inner-cabinet vibrations undamped, which get back to the driver, making its cone vibrate (= noise).
When these tones get into the cabinet material via those screws, there is no way to damp them, and no way to brace against them, because they travel inside the cabinet material, not on the surface. If those screws cannot be rubber-isolated, the cabinet material needs to have ’high internal damping’, which is not a property of metal, cement, nor most woods. Cabinet thickness does not matter here, at least for making something to fit in a home.
For a home constructor, I recommend 3/4" Baltic birch plywood cabinets, or at least for its front panel, with braces inlaid 6-8 inches apart, center-to-center. But BB plywood is such a tough wood to work with and to make pretty! Two 1/2" BB layers glued up with Elmer’s Carpenter’s Glue made a front panel that was a bit more dead to those midrange tones getting into it from the screws, but not enough to justify the extra work. It made far more difference to put 1/4" wool felt on the front baffle to suck up the tweeter’s reflections before getting on with designing the tweeter’s crossover." Roy Johnson Green Mountain Audio @royj, Sorry for almost missing your post. Every point you made/quoted seems to concur with what others have discovered through long arduous painstaking experiments. Thanks for that, Roy, I will print it off and keep it for reference. No (esp solid sealed) cabinet can do it all (great bass, mid and treble), but the heavily damped thin walled approach championed by the BBC (Harbeth, Spendor, etc) seems to do the least damage in the precious midrange. So many of those speakers LS3/5, BC1s plus innumerable Spendor or Harbeth models are renowned for the purity of their midranges. The use of stethoscopes and accelerometers sure has a funny way of saving us all a lot of time and energy! |
There are some who still feel that using MDF alone in a traditional cabinet is a tonal midrange killer. So naturally enough many other options have been explored, all usually costing far more than the readily available and cheap to work with MDF. Some cynics might argue that they were the only reasons for MDF becoming almost ubiquitous in the last 2/3 decades. More recently composites have become increasingly popular. Peter Comeau, formerly of Heybrook, now working with IAG-owned brands like Wharfedale, Quad, Audiolab and Castle, has been studying cabinet resonances for decades, and now uses chipboard/MDF composite panels for his loudspeakers. Harbeth stick with MDF / bitumen damping panels but their cabinets use a unique ’lossy’ construction which flies in the face of all those who favour high rigidity above all else. Harbeth also firmly believe in using veneer on both sides of the speaker panels - presumably to ensure long term stability. Nobody yet has claimed to have been able to make those pesky cabinet resonances totally inaudible, and it’s difficult to see how they could, but the choice of where to put them will always partially depend upon what material you use to build your box. Fans of panel or open baffle speakers will have to look for other things to worry about. https://www.google.com/amp/s/zstereo.co.uk/2018/10/20/peter-comeau/amp/ |
