Does anyone have comments on the effectiveness of bracing the woofer magnet to the rear and sides with a partial shelf brace? I did this with my last set of diy speakers and even though the I used 1" (25 mm) mdf and 3" (75 mm) baffle, and plenty of shelf braces, there was more vibration than I liked. Does bracing the woofer magnet feed more vibration into the cabinet rather than mass damping vibration? The rear panel was 2" thick and total weight 60 kg per sub cabinet. I know some commercial designs have used isolating membranes (KEF) but not everyone liked the result.
I wouldn't brace up to the magnet, mainly as I'd like to keep reflections to the rear of the cone to a minimum. I do find it odd that you were getting vibrations with your previous enclosure-did you use a foam gasket under the mounting flange of the driver? If so then the front of the driver could have had some play and that could cause the vibration you speak of-if the magnet was able to move a little on it's brace.
The rear magnet brace was a front-to-rear shelf brace and only presented a very narrow profile to woofer air movement. I used a neoprene rubber pad between the magnet and shelf brace. The vibration was enclosure, not driver chassis. With my next speakers I will run some comparative tests with and without rear bracing.
I think your enclosure was vibrating because of the moving mass of the driver. A heavier enclosure should help:
For a driver of moving mass 0.1Kg, moving 20mm p/p (about right for a subwoofer on the loud bits), a 10Kg cabinet will move 0.2mm p/p (conversation of momentum). A 20Kg cabinet will move 0.1mm p/p.
Try force-cancelling loading of the drivers - put drivers on opposite sides of the cabinet (both facing out), and the velocity of one driver in one direction is matched almost exactly (how exactly will be down to manufacturer's tolerances, how the speakers are wired, etc). The momentums cancel to produce no net force acting on the cabinet, so it will be free of vibration so long as you brace it properly - bracing is still required as there will still be pressure differences between the inside and outside of the cabinet, so it will still want to inflate like a baloon - this is much easier to brace against.
Chris
For a driver of moving mass 0.1Kg, moving 20mm p/p (about right for a subwoofer on the loud bits), a 10Kg cabinet will move 0.2mm p/p (conversation of momentum). A 20Kg cabinet will move 0.1mm p/p.
Try force-cancelling loading of the drivers - put drivers on opposite sides of the cabinet (both facing out), and the velocity of one driver in one direction is matched almost exactly (how exactly will be down to manufacturer's tolerances, how the speakers are wired, etc). The momentums cancel to produce no net force acting on the cabinet, so it will be free of vibration so long as you brace it properly - bracing is still required as there will still be pressure differences between the inside and outside of the cabinet, so it will still want to inflate like a baloon - this is much easier to brace against.
Chris
aka Active Vibration Cancelation. An elegant solution to the problem, removing an estimated 90% of the energy mechanically transferred to the box, making ballooning the biggest issue.
The majority of energy passed into the cabinet is from the mechanical attachment of the driver to the box (althou in a sub, air transmitted energy is at its highest, especially if it is sealed). If you only couple the driver to the baffle then it, the weakest part of your box, gets the brunt of the energy, i feel it is best to spread it around so that the energy is shared more evenly, bringing as much of the inherent damping of the panel material into play as possible. I rigidly couple the back of the driver to the rest of the cabinet. Your lossy gasket is not helping.
If you have vibration that you can feel it means that some of the panels have resonant frequencies in the bandpass of the woofer -- that is very low, and hard to do.
Now MDF is not very stiff. The side walls for instance may seem thick but 1" MDF is about the same stiffness as 15mm baltic birch. I'd use 15mm (we do not use mdf) for a woofer for a 7" maybe an 8", and that with bracing, anything larger i go thicker. Being not very stiff MDF doesn't make very good braces. A good place to use solid hardwood.
Have you sealed both the inside & outside of the MDF? It is leaky and should be sealed inside & out if you want it to be airtight.
Can you post some drawings so that we can get an idea of the dimensions involved?
dave
Although I have started this thread in the subwoofer forum, it is really a 3-way speaker with separate sealed bass enclosure containing 2 X !0" woofers crossing over below about 150 Hz. The mid-treble cabinet being smaller and stiffer, has no discernable vibration that concerns me. The force-cancelling rod strategy makes a lot of sense, IF it can be properly implemented. Although I am experienced in loudspeaker DIY, I don't consider this is within my abilities without the support of vibration analysis measurement equipment. My last woofer enclosures had 75 mm thick baffle and total weight approx 60 kg. The sides had the most apparent vibration even with multiple shelf braces.
With two 10" woofers in the cab why not mount them on opposing faces as suggested? That way their opposite motion cancels the bulk of the vibration.
As I stated, without the proper instrumentation it is virtually impossible to make the force cancelling strategy work as intended. Each driver has slightly different moving mass, the effective wall stiffness's are slightly different, the drivers have minor differences in frame stiffness, mass etc. All this means a major tuning exercise for each implementation. There must be a good reason why I can't recall any current no-limit speaker uses this technology. My experience is that there is no getting around the effects of the three basic parameters of mass, stiffness and damping. Of these it is the damping that is hardest to get right. I am just seeking users experience on whether the woofer bracing strategy just succeeds in transferring vibration from the front baffle to the entire enclosure. If so, my inclination is to try deal with induced woofer vibration with constrained multilayer damped baffle, with the outer baffle layer isolated from the inner layers with a damping layer.
You are overthinking it...
There are quite a few makers taking advantage of push-push active vibration cancelation. KEF, Naim, B&W, Paragon, Martin Logan to pull a few off the top.
It is a no-brainer & a well established technique.
A stiff cabinet (preferrably not MDF) and suitable braces tightly coupling the magnets of the drivers.
dave
I am not saying that KEF, B&W, Martin Logan, Naim and the other makers mentioned, do not make excellent speakers, but the force cancellation technique seems most popular for their stand alone sub-woofers whereas I am will be using it as a woofer section (<150 HZ) of a two box 3-way loudspeaker. Since the mid-treble enclosure sits on top of the woofer section, I want to minimize any woofer-to-upper cabinet vibration transfer as well as the woofer walls. When I mentioned no-limit designs, I am talking about Rockport, Magico, YG Acoustics, Meridian etc, who all use exotic materials and construction techniques to eliminate cabinet vibration. I only have access to the usual diy materials, but I want to make the best use of them. I have already invested in top class Scan Speak drivers (Revelator mid-range, tweeter and twin 10" woofers). The basic construction is constrained layers of 18 mm mdf with a damping layer between. I will post some pictures as construction proceeds.
We use this technique in bi-pole FR boxes, so there is no frequency limit (althou as frequency goes up there is less energy to load into the box), besides reducing cabinet resonances it also aids in increasing DDR.
Commercail use extends far beyond subwoofers.
If you want to eliminate vibration, push-push is very effective.
MDF is a bad start if you want a box with little compromise.
dave
Commercail use extends far beyond subwoofers.
If you want to eliminate vibration, push-push is very effective.
MDF is a bad start if you want a box with little compromise.
dave
You can suspend the mid (treble) cabinet with rubber bands ; on two columns if above the woofer ,or, if hanged down , just swinging .
(treble) the tweeter may be also decoupled from the midrange enclosure .
I did some testing before I settled on mdf. Like many others, I have been suspicious of the use of mdf, so I made up some 30 X 30 cm (1 ft square) test panels of constrained layer construction. The damping layers were 1) Sikaflex-Pro polyurethane sealant adhesive, 2) Bostick glazing silicone, 3) 3 mm rubber sheet, 4) polyurethane wood adhesive, 5) Fiberglass resin. All the damping layers were 3 mm thick. The constraining layers are 18 mm mdf or 18 mm birch ply. The objective tests was to drop a 20 mm steel ball bearing from a fixed height onto the test panels, which were either clamped at two edges or solidly held down onto a 150 mm (6") thick concrete slab. The height of the rebound was measured on a calibrated scale. The lower the bounce, the more energy absorbed. To cut a long story short, the best performer was mdf+Sikaflex-Pro , with the mdf+Bostik a close second. The worst was mdf+fibreglass resin and close behind mdf+polyurethane wood adhesive. In these last two cases, the high bounce was accompanied by a definite "ping". Most unpleasant! The mdf+Bostik glazing was the best choice for a couple of reasons. a) Cost; half the price of the Sikaflex, b) ease of extrusion; the Sikaflex was very hard to pump out of the gun, a real issue for large areas. The subjective test was listening to the signature of the ball bearing impact. All of the plywood panels sounded distinctly boxy, woody and coloured, and were rejected. As for isolating the mid-treble from the woofer, I have used sorbothane hemispheres between the enclosures with excellent results.
Mate. You put a rod bracing between speaker magnet and the back panel of You cab to dampen the panel resonance? How it perform , please ? Thx.
Bon,
With all the love in the world, do yourself a favour and build a few test boxes. Have a listen and get "sawdust dirty" to make these calls.
All the ball bearings in the world on such a complex test will tell you..... how high a ball bearing bounces.
I see people throwing pearls out here - take them and have a try.
Balanced drivers and bracing is abut as close as you will ever get to a "free kick" if you are fussed about cabinet flex.
Build some of these and post your results!
With all the love in the world, do yourself a favour and build a few test boxes. Have a listen and get "sawdust dirty" to make these calls.
All the ball bearings in the world on such a complex test will tell you..... how high a ball bearing bounces.
I see people throwing pearls out here - take them and have a try.
Balanced drivers and bracing is abut as close as you will ever get to a "free kick" if you are fussed about cabinet flex.
Build some of these and post your results!
Inconceivable anyone could hear the improvement in blind testing between good construction and obsessive construction.
Just maybe remotely possible you could measure a difference in actual sound in the room except maybe with trick techniques.
On those big notes, your woofer is moving great gobs of air (and producing 20% harmonic distortion) and your cab walls are moving on a far far tinier scale (albeit deceptively easily felt with your fingers).
B.
This sure is a Lazarus thread ...
I must have built at least a dozen pairs of speakers since I asked for advice. To cut to the chase, the most effective technique I have found for controlling box vibration is opposed reaction cancelling woofers, as suggested above by planet 10 and chris661. It is spookily effective. I have completely abandoned my reservations about this technique.
https://www.diyaudio.com/forums/subwoofers/304882-force-cancellation-woofers-3.html#post5706556
That's because the topic never dies! Back then you said "mdf+Bostik glazing was the best choice for a couple of reasons. a) Cost; half the price of the Sikaflex, b) ease of extrusion; the Sikaflex was very hard to pump out of the gun,"
--> Is that still your best? Or any revision since then?
I kept a constrained layer sample about 60x65mm sitting on top of my speakers as a conversation piece for 8 years now. The silicone constrained layer is still pliable, definitely not hardened. Despite this, I can't pull it apart. The only time I managed to separate a sample, I had to put it in the vise and with extreme force the mdf failed rather than the silicone interface. This is still the best readily available damping layer I have found. Just make certain it is glazing standard silicone. This is designed to be flexible, durable and uv stable.
Despite my success with cld speaker construction, I am a huge fan of reaction cancelling woofers. My preferred strategy is separate enclosures for the mid/treble (cld) and bass (reaction cancelling opposed woofers).
Despite my success with cld speaker construction, I am a huge fan of reaction cancelling woofers. My preferred strategy is separate enclosures for the mid/treble (cld) and bass (reaction cancelling opposed woofers).
Attachments
KEF mounted the B200 magnet and isolated the baffle from frame to reduce distortion, not really to reduce box resonances. You might investigate that anyway when trying to build the best of the best. Yes, opposing woofers with the magnets bolted together works great for subs. I wish my current sub project could support that concept.
Now is should be very easy to mount your mid cabinet on some little feet. Sobothane maybe. felt dots, stick on rubber feet, but just mounting on feet on the edges of the lower box should do fine. Or on a post from behind. Or, well many ways to isolate.
It is my viewpoint Exotic material are pretty much marketing jibber-jabber. MDF, PB and Ply are all quite suitable if used well. Hard to charge $20K for a MDF box even if it works. For woofer sections, I prefer birch ply over MDF as it is more rigid, mid cabs I prefer MDF for the low Q. Then there was always the sand cavity folks like Tannoy used. I have seen tensioned rods, and just super thick silastic dampening to lower the resonance and Q.
Now is should be very easy to mount your mid cabinet on some little feet. Sobothane maybe. felt dots, stick on rubber feet, but just mounting on feet on the edges of the lower box should do fine. Or on a post from behind. Or, well many ways to isolate.
It is my viewpoint Exotic material are pretty much marketing jibber-jabber. MDF, PB and Ply are all quite suitable if used well. Hard to charge $20K for a MDF box even if it works. For woofer sections, I prefer birch ply over MDF as it is more rigid, mid cabs I prefer MDF for the low Q. Then there was always the sand cavity folks like Tannoy used. I have seen tensioned rods, and just super thick silastic dampening to lower the resonance and Q.