We first came across the issue at KEF after the launch of the original Reference 105. (The older unit with the wooden head assembly) It was noticed that a coloration was always there on voice, something in the middle hundreds. As the Fourier computers came into use there was a program for sampling acceleration all over the cabinet and then showing an animated plot of the cabinet at any modes found. There was a strong one where the cabinet front and 12" woofer went one way and the rear panel the other at a strong combined woofer and cabinet resonance. Rather than trying stiffening the better solution was found to be to decouple the woofer from the cabinet. The KEF 12" trochoidal unit was developed with rubber issolation mounts and a soft foam gasket behind. The resonant frequency of the suspended woofer was low and the vibration delivered to the cabinet greatly reduced. The coloration went away.
A cheaper version of issolation for the B200 8" was developed with rubber grommets and a steel sleeve that prevented you from overtightening the assembly. That was always a problem. If you could tighten down some screws then you would short circuit the suspension and negate the benefit. Mike Gough told me you could be in the anechoic chamber with a repeating impulse going to the woofer and as you tightened the screws "tick, tick tick" would become "boink, boink, boink". Loosen the screws slightly and the cabinet resonance would go away.
Maybe its a British thing to worry about cabinet resonances.
David S.
A cheaper version of issolation for the B200 8" was developed with rubber grommets and a steel sleeve that prevented you from overtightening the assembly. That was always a problem. If you could tighten down some screws then you would short circuit the suspension and negate the benefit. Mike Gough told me you could be in the anechoic chamber with a repeating impulse going to the woofer and as you tightened the screws "tick, tick tick" would become "boink, boink, boink". Loosen the screws slightly and the cabinet resonance would go away.
Maybe its a British thing to worry about cabinet resonances.
David S.
Many forums don't set a time limit. I tend to sometimes make corrections or small additions even after days or sometimes weeks - I don't think that's a problem, as long as you don't completely change the post. At diyaudio I sometimes get a bit annoyed when I discover I made some stupid little mistake and I'm too late to change it.
Davids post would become easier to understand if he could correct it. I had to read said sentence several times to see that the sentence was wrong, not me.
Yes, but we have other threads for beating up on those subjects. 🙄
My experience is this: I have a friend who builds replica Fender amps with 1/2" pine sides. He installs speakers that do +100dB with one watt, and there is definitely a large contribution from the box compared to a more modern MDF enclosure.
Now, if you listen to string quartets at levels that allow conversation, feel free to discount this data as inapplicable. However, I suspect at least a few here have to shout, "SOUNDS GOOD, DOESN'T IT?!?' 😎
Member
Joined 2009
Paid Member
Cal,
I still like your box with the opening behind the mid-driver. I think it deserves more attention - do you have more to share on it ?
p.s. I heard a nasty rumour you are moving over to small speakers. This isn't true is it ?
I still like your box with the opening behind the mid-driver. I think it deserves more attention - do you have more to share on it ?
p.s. I heard a nasty rumour you are moving over to small speakers. This isn't true is it ?
Last edited:
It's ok to ask the moderators to make changes for you. Just hit the report post button. The reason the editing is set at 30 minutes is there have been many instances where the poster has changed the content to the point where a quote or question from someone seems to come out of thin air and is now irrelevant, not to mention confusing.
Hi Bigun,
Not really. This thread has strayed a long way from what I was describing and why it appeals. I was speaking of the box's effect on the driver and this thread seems to be looking at it from the other side. All I know is that planet10 Dave opened my eyes to a problem I didn't know I had but once I learned, there's almost no going back.
No no, that's just a nasty rumour. 🙂
Constrained Layer damping
What they are showing in their graphs is very little ringing, mostly at high frequencies due to very stiff material AND very small distances between braces and a significant level of attenuation of said HF ringing from the constrained layer damping in their "Q5" material. No loose screws need apply.
As for being wasteful... maybe so, it may be overkill. Certainly very expensive to make. But it sure looks like an all-out attempt to make a really inert cabinet.
To just diss them without further looking into their documentation seems a little unfair. It really does look like they have dramatically reduced box colorations. The Magico approach could also be applied by DIYers, albeit with different (read: less expensive) materials. Certainly, the B&W matrix boxes of yore, along with additional constrained layer damping are something to have a look at.
In their web-site, Magico show the use of constrained layer damping and show the effects in one of the waterfall graphs.
What they are showing in their graphs is very little ringing, mostly at high frequencies due to very stiff material AND very small distances between braces and a significant level of attenuation of said HF ringing from the constrained layer damping in their "Q5" material. No loose screws need apply.
As for being wasteful... maybe so, it may be overkill. Certainly very expensive to make. But it sure looks like an all-out attempt to make a really inert cabinet.
To just diss them without further looking into their documentation seems a little unfair. It really does look like they have dramatically reduced box colorations. The Magico approach could also be applied by DIYers, albeit with different (read: less expensive) materials. Certainly, the B&W matrix boxes of yore, along with additional constrained layer damping are something to have a look at.
Last edited:
Speaker should always be much larger than your wife. That usually ensures you can - over time - upgrade to even bigger speakers .... 😉
Member
Joined 2009
Paid Member
I've seen Open Baffle builders do this, trying to reduce coupling between the driver and the baffle. I've tried using this approach with power transformers - keeping the hum out of the chasis using rubber grommet based mounts. I had only limited success. But I like the idea wrt speaker drivers. What is considered state-of-art for compliant speaker mounting ?
For a full-range range, there would be the trade-off that the speaker driver will be vibrating in it's compliant mount which could introduce Doppler distortion as the high frequencies are now being generated from a moving source.
Perhaps it comes from the Navy, we don't want the boats shaking around too much when we're shooting lots of canons ?
Well in my books what you were talking about is still relevant to the title of this thread. The trouble is, my next build is a single-full range, I can't have an opening behind the mids-section without also having an opening behind the bass. So I have to tradeoff bass extension (box) against open back clarity - or I have to do something really clever.
glad to hear it, you are my inspiration for going big with the next build!
I'll note this as a design requirement 😀
We could always make cabinets out of 1/2 granite sandwich with dynamat, making 1" thick cabinet walls..perhaps even with threaded rods through it with adjustable tension.
It seems like it might be easier to tune and control resonances...perhaps even to our advantage rather than viewing them as a nuisance to eliminate...
Blame the whiskey, I'm thinking outloud.
If I can extrapolate from studio design ie the room-whithin-a-room approach I'd say if you use a sandwich material for the box it could be important/beneficial if the driver is mounted on only one layer of granite or whatever but not bolted through both hard layers.
In studios even a single bolt or other solid/stiff connection between inner and outer shell can ruin things and knock 10dB off the insulation.
It'd also be conceivably better than soft-mounting the driver to my mind.
In studios even a single bolt or other solid/stiff connection between inner and outer shell can ruin things and knock 10dB off the insulation.
It'd also be conceivably better than soft-mounting the driver to my mind.
I am trying to understand the mechanism that is being implied.
1. is the "coloration" due to a direct coupling of the speaker to the frame. IOW, the energy near the system resonance is "diverted" to shaking the cabinet rather than producing SPL
Or
2. The SPL is produced, then the air vibrates cabinet (ignore the huge loss due to impedance mismatch) and the SPL is then converted to cabinet
vibration.
If it was mostly caused by #1, then a rubber washer or gasket around the speaker frame should work miracles. As for #2, wouldn't there be a huge loss of energy due to impedance mis-matches (wood-to-air or air-to-wood)
1. is the "coloration" due to a direct coupling of the speaker to the frame. IOW, the energy near the system resonance is "diverted" to shaking the cabinet rather than producing SPL
Or
2. The SPL is produced, then the air vibrates cabinet (ignore the huge loss due to impedance mismatch) and the SPL is then converted to cabinet
vibration.
If it was mostly caused by #1, then a rubber washer or gasket around the speaker frame should work miracles. As for #2, wouldn't there be a huge loss of energy due to impedance mis-matches (wood-to-air or air-to-wood)
My samples of daycron and long hair wool came, so maybe this weekend I can see which does a smoother job. With the daycron, I really only have a little bump above 500 that is more likely an edge diffraction artifact. These are built to the SR-72 dimensions so drivers are symmetrical. The full radius on all edges had a rather dramatic effect. I have seen both sides of the decoupled driver argument. Guess I have to try it myself to know.
No evidence? Speakers built using this philosophy vrs Spendor BC1 and Rogers LS3/5A in my listening room. But you would have had to have been there.
dave