Re: Re: Re: plywood/mdf/etc...
On of the best damping material to glue inside the cabinet is Pb sheet.
Some guys use to glue Pb sheet with bitumen.
Once the speakers are in place you don't move them anymore.
critofur said:
Your best bet for reducing cabinet resonances is to layer materials with substantially different Young's Modulus, perhaps making a sandwich with a visco-elastic damping material in the middle. Bitumen is my favorite, but there are a variety of damping materials you can buy, usually with double-sided adhesive, or, the material itself is adhesive. Presently, I no longer have a source for rolls of bitumen sheet.
On of the best damping material to glue inside the cabinet is Pb sheet.
Some guys use to glue Pb sheet with bitumen.
Once the speakers are in place you don't move them anymore.
>What about the claim of MDF being acoustically "dead" and anti-resonant?
====
It's not, it just resonates down around 350-400Hz depending on its 'mix'. Acoustically dead is when you rap your knuckles on something and all you hear is them cracking, such as against a concrete driveway.
====
>It is also less likely to warp and bend than normal wood, right?
====
True, but then using the most common 'normal' woods is a no-no if best long term performance is desired due to their high moisture retention causing too much expansion/contraction over time.
Since the cabs will be small, some nice low moisture retention hardwoods are an option, such as white oak, white ash, black ash, green ash, and hard maple. Then for those with machining capability or more patience than me: Lignum Vitae, Kingwood, Cocobolo, Tulipwood, Pau Ferro, Honduras Rosewood, Goncalo Alves, Bubinga, Wenge, and Purpleheart. There may be others, but this is all I could quickly find.
====
>One important point to keep in mind is that the stiffness goes up with the square of the thickness, so, 3/4" is more than twice as good as 1/2" (and 1" is 4x as good as 1/2", and so on...).
====
Possibly you are confusing stiffness gains with energy stored/dissipated? 2x as stiff, 1/4 the energy is absorbed. Anyway, stiffness increases at the cube of thickness, so 1/2" to 3/4" increases stiffness ~3.375x and 1" is 8x stiffer, big difference.
====
>The stiffer your box is, the more bass you'll get. But, even a super stiff box made out of say, steel, will ring like a bell if it isn't well damped (so it will sound like crap).
====
True, and why it is better to build a subwoofer cab out of a stiff material and add a bit of bracing since it raises its Fs above the passband. Even if used for mids/HF, since acoustic energy falls at 1/F, so does the amount of damping required with increasing stiffness.
====
>On the other hand, as the material gets thicker it resonates at a lower frequency meaning that it is more likely that there will be enuff energy to excite that resonance.
====
Yep, and increases damping (lowering its Q), spreading it over a wider BW. Double the density and Fs falls 0.707x and its Q by a factor of 4!
This is what screws you when you start with a relatively low Fs material such as MDF. Increasing stiffness through bracing doesn't raise it much unless you turn it into a maze, and increasing thickness spreads out any stiffness gain, so to get it well damped enough is a 'chasing your tail' exercise as it has to damp an ever increasing acoustic pressure. To get a subwoofer truly 'dead' with MDF requires an incredibly thick panel construction, so you are better off using concrete or brick construction.
====
>Considering all this talk about cabinet resonance, what thickness do you guys think is best for the 871: 1/2", 3/4", or 5/8"?
====
This driver can't generate much acoustic pressure and the cab is small, so 1/2" void free with a little damping is adequate, with 3/4" being plenty stiff/damped enough.
====
>Another question: if you discover that the cabinet strongly resonates at a particular frequency, is there any way you can tame it at that frequency?
====
Yeah, either add some damping material such as Deflex, or my preference, glue a diagonal brace of scrap plywood on edge to the panel to break it up into two unequal area triangles.
====
>If you all must say plywood, what kind should I buy? Birch? Maple?
====
Baltic Birch or Appleply is preferred, but if you can find some other cheaper void free plywood, then it's your call. The more plys there is, the stiffer/better damped it is.
====
>If I choose to use veneered ply, external fasteners (i.e. screws) should not be used at all; I don't want to ruin the finish. A good option then is to use dovetail joints (which are held together by glue alone) instead of the common butt and miter joints. Is this feasible for larger cabinets?
====
Yes, but FWIW, I built up to 30ft^3 cabs with butt or mitered joints reinforced with glued and/or screwed in rabbitts made from scrap. Got two 20ft^3/~345lb (including drivers) cabs built in '69 in the living room right now that haven't cracked a joint yet, though they've been moved around quite a bit over the decades. Due to their size/weight though, I used 2x4s for the rabbitts.
GM
====
It's not, it just resonates down around 350-400Hz depending on its 'mix'. Acoustically dead is when you rap your knuckles on something and all you hear is them cracking, such as against a concrete driveway.
====
>It is also less likely to warp and bend than normal wood, right?
====
True, but then using the most common 'normal' woods is a no-no if best long term performance is desired due to their high moisture retention causing too much expansion/contraction over time.
Since the cabs will be small, some nice low moisture retention hardwoods are an option, such as white oak, white ash, black ash, green ash, and hard maple. Then for those with machining capability or more patience than me: Lignum Vitae, Kingwood, Cocobolo, Tulipwood, Pau Ferro, Honduras Rosewood, Goncalo Alves, Bubinga, Wenge, and Purpleheart. There may be others, but this is all I could quickly find.
====
>One important point to keep in mind is that the stiffness goes up with the square of the thickness, so, 3/4" is more than twice as good as 1/2" (and 1" is 4x as good as 1/2", and so on...).
====
Possibly you are confusing stiffness gains with energy stored/dissipated? 2x as stiff, 1/4 the energy is absorbed. Anyway, stiffness increases at the cube of thickness, so 1/2" to 3/4" increases stiffness ~3.375x and 1" is 8x stiffer, big difference.
====
>The stiffer your box is, the more bass you'll get. But, even a super stiff box made out of say, steel, will ring like a bell if it isn't well damped (so it will sound like crap).
====
True, and why it is better to build a subwoofer cab out of a stiff material and add a bit of bracing since it raises its Fs above the passband. Even if used for mids/HF, since acoustic energy falls at 1/F, so does the amount of damping required with increasing stiffness.
====
>On the other hand, as the material gets thicker it resonates at a lower frequency meaning that it is more likely that there will be enuff energy to excite that resonance.
====
Yep, and increases damping (lowering its Q), spreading it over a wider BW. Double the density and Fs falls 0.707x and its Q by a factor of 4!
This is what screws you when you start with a relatively low Fs material such as MDF. Increasing stiffness through bracing doesn't raise it much unless you turn it into a maze, and increasing thickness spreads out any stiffness gain, so to get it well damped enough is a 'chasing your tail' exercise as it has to damp an ever increasing acoustic pressure. To get a subwoofer truly 'dead' with MDF requires an incredibly thick panel construction, so you are better off using concrete or brick construction.
====
>Considering all this talk about cabinet resonance, what thickness do you guys think is best for the 871: 1/2", 3/4", or 5/8"?
====
This driver can't generate much acoustic pressure and the cab is small, so 1/2" void free with a little damping is adequate, with 3/4" being plenty stiff/damped enough.
====
>Another question: if you discover that the cabinet strongly resonates at a particular frequency, is there any way you can tame it at that frequency?
====
Yeah, either add some damping material such as Deflex, or my preference, glue a diagonal brace of scrap plywood on edge to the panel to break it up into two unequal area triangles.
====
>If you all must say plywood, what kind should I buy? Birch? Maple?
====
Baltic Birch or Appleply is preferred, but if you can find some other cheaper void free plywood, then it's your call. The more plys there is, the stiffer/better damped it is.
====
>If I choose to use veneered ply, external fasteners (i.e. screws) should not be used at all; I don't want to ruin the finish. A good option then is to use dovetail joints (which are held together by glue alone) instead of the common butt and miter joints. Is this feasible for larger cabinets?
====
Yes, but FWIW, I built up to 30ft^3 cabs with butt or mitered joints reinforced with glued and/or screwed in rabbitts made from scrap. Got two 20ft^3/~345lb (including drivers) cabs built in '69 in the living room right now that haven't cracked a joint yet, though they've been moved around quite a bit over the decades. Due to their size/weight though, I used 2x4s for the rabbitts.
GM
Doesn't the frequency depend on the size of the panels? Larger panels resonate at some frequency, bracing "breaks up" the big panel into smaller panels that all resonate at higher frequencies... this was my limited understanding of how mdf cabinets respond, but I've only built subwoofers, where the higher the cabinet panels resonate the better.
Dan
>Doesn't the frequency depend on the size of the panels?
====
Correct. Anytime specs are used in these types of threads, all you can ~accurately compare is manufacturer's specs, i.e. MOR/MOE/J-ball/density/etc.. The effects of joints, etc., have a big effect on the finished cab's performance and computing them is no trivial pursuit.
====
>Larger panels resonate at some frequency, bracing "breaks up" the big panel into smaller panels that all resonate at higher frequencies...
====
Yes, this is what I said, but the lower the material's Fs is, the more stiffening required to be of sufficient benefit to justify the additional cost/labor/size/weight. I would rather start with a high Fs material, use minimal bracing/damping, spend less time building and more time enjoying them.
GM
====
Correct. Anytime specs are used in these types of threads, all you can ~accurately compare is manufacturer's specs, i.e. MOR/MOE/J-ball/density/etc.. The effects of joints, etc., have a big effect on the finished cab's performance and computing them is no trivial pursuit.
====
>Larger panels resonate at some frequency, bracing "breaks up" the big panel into smaller panels that all resonate at higher frequencies...
====
Yes, this is what I said, but the lower the material's Fs is, the more stiffening required to be of sufficient benefit to justify the additional cost/labor/size/weight. I would rather start with a high Fs material, use minimal bracing/damping, spend less time building and more time enjoying them.
GM
Re: Re: Re: Re: plywood/mdf/etc...
The frequency depends on the size of the panel.
I think concrete actually has a higher Q than MDF, it certainly rings at a higher frequency. It's actually a rather poor material for most speaker cabinet applications other than subwoofers.
Regardless, your example is not a very fair comparrison because just about anything will be VERY dead sounding if it's damped with tons of earth under it the way a driveway is. Go find a nice flat patch of moist dense ground, press a piece of MDF into it, and you'll find it' doesn't ring much after you wap it.
*chuckle* Rosewood and Cocobolo are two of the most prefered woods for xylophones/marimbas because they resonate so well! Exactly what you DON'T want in a speaker cabinet. (unless you have very good damping)
Brick is substantially better than concrete, if you could get custom cast brick (or even better, porceline) panels they could make superb subwoofer cabinets. Build bracing into the design so the outside is flat but inside has braces so you get the stiffness but sacrifice less cabinet volume. Some custom installers assemble marble speaker enclosures onsite, works very nicely.
If you use 3/4 MDF + 3/4 BB ply it will be fine with just minimal bracing. But that's not the subject of the thread...
For small speakers with small drivers like the W3-871 I say use 1/2 baltic or danish birch (more layers than domestic birch) plywood, damp it with a couple layers of bitumen/asphalt based "roof repair" tape and use lots of 1/4" plywood triangular braces. If you build the box without the back first, it's quick and easy to put the braces in using hotmelt glue, besides, you want to damp/brace the front of the speaker the most.
GM said:
The frequency depends on the size of the panel.
I think concrete actually has a higher Q than MDF, it certainly rings at a higher frequency. It's actually a rather poor material for most speaker cabinet applications other than subwoofers.
Regardless, your example is not a very fair comparrison because just about anything will be VERY dead sounding if it's damped with tons of earth under it the way a driveway is. Go find a nice flat patch of moist dense ground, press a piece of MDF into it, and you'll find it' doesn't ring much after you wap it.
*chuckle* Rosewood and Cocobolo are two of the most prefered woods for xylophones/marimbas because they resonate so well! Exactly what you DON'T want in a speaker cabinet. (unless you have very good damping)
Brick is substantially better than concrete, if you could get custom cast brick (or even better, porceline) panels they could make superb subwoofer cabinets. Build bracing into the design so the outside is flat but inside has braces so you get the stiffness but sacrifice less cabinet volume. Some custom installers assemble marble speaker enclosures onsite, works very nicely.
If you use 3/4 MDF + 3/4 BB ply it will be fine with just minimal bracing. But that's not the subject of the thread...
For small speakers with small drivers like the W3-871 I say use 1/2 baltic or danish birch (more layers than domestic birch) plywood, damp it with a couple layers of bitumen/asphalt based "roof repair" tape and use lots of 1/4" plywood triangular braces. If you build the box without the back first, it's quick and easy to put the braces in using hotmelt glue, besides, you want to damp/brace the front of the speaker the most.
Thanks everyone for the numerous responses.
Based on what I've read, I will decide on either 1/2" or 5/8" baltic birch plywood with minimum bracing.
But I have another idea for a more advanced project: a cabinet with non-parallel sides on the inside but is absolutely rectangular on the outside. This means non-uniform-thickness walls (i.e. a side panel 3/4" thick in the front that tapers to 1/2" in the back). Does this make much of a difference (as far as reducing standing waves)?
Based on what I've read, I will decide on either 1/2" or 5/8" baltic birch plywood with minimum bracing.
But I have another idea for a more advanced project: a cabinet with non-parallel sides on the inside but is absolutely rectangular on the outside. This means non-uniform-thickness walls (i.e. a side panel 3/4" thick in the front that tapers to 1/2" in the back). Does this make much of a difference (as far as reducing standing waves)?
coolkhoa said:
2nd time in a month i've had this idea proposed. It won't do much for internal standing waves, but will be kind off interesting in what it does to the panel resonances.
Scott Dunn has such a construction in the finishing stages -- he laminated very fine grain doug fir to MDF and ended up with a minimum 3/4" thickness.
dave
After reading through this thread twice, im still a tad confused over what the true optimum volume is. Ive read several different values ( 0.59, 0.95, 0.1, etc) and im just wondering which of them is correct. My brother is building a center channel speaker for the HT system with this driver. Box size doesnt really matter to me, so even if the 0.95 one is correct, thats fine by me. Sorry if this is a dumb question...
Chris8sirhC said:
No enclosure volume is "correct." It all depends on personal taste. Based on what others have said, the optimum sizes (internal) for the W3-871 range from 0.1 cu ft to 0.2 cu ft. The minimum size offers the best power handling but with a slight peak in the bass response (higher Q), whereas the maximum size gives you less power handling but a truer response. A 0.13 cu ft box is a good compromise, with a good balance of power handling and flatness of response.
It's all documented in John's Audio Projects Page.
>The frequency depends on the size of the panel.
====
So I said already.
====
>I think concrete actually has a higher Q than MDF, it certainly rings at a higher frequency.
====
Hmm, the density of MDF is ~50lbs/ft^3 Vs concrete's ~150lbs/ft^3..... nah, I don't think so.
====
>It's actually a rather poor material for most speaker cabinet applications other than subwoofers.
====
From a purely technical POV, wrong again.
====
>Regardless, your example is not a very fair comparrison because just about anything will be VERY dead sounding if it's damped with tons of earth under it the way a driveway is. Go find a nice flat patch of moist dense ground, press a piece of MDF into it, and you'll find it' doesn't ring much after you wap it.
====
I wasn't comparing the two, only to provide a reference for what defines 'dead'. What most folks claim as 'dead' because it sounds dull really has a Fs right in a sub's upper BW/typical satellite's lower BW. Anyway, MDF will still ring more than the concrete due to their base "Q" difference.
====
>*chuckle* Rosewood and Cocobolo are two of the most prefered woods for xylophones/marimbas because they resonate so well! Exactly what you DON'T want in a speaker cabinet. (unless you have very good damping)
====
*chuckle* is right. Thanks for proving my point/indirectly endorsing my recommendation, though that obviously wasn't your intention. They resonate so well because they are stiff, which means that they are easily damped due to their high Q/low acoustic amplitude.
In the case of a separate mid/tweeter, if they resonate above the mid's passband then they won't be excited since the tweeter has a sealed back.
====
>Brick is substantially better than concrete, if you could get custom cast brick (or even better, porceline) panels they could make superb subwoofer cabinets.
====
Not really, since the joints are the limiting factor and they are pretty weak. Brick construction with a bonding coating works extremely well though, and the preferred construction for large subterranean basshorns.
====
>Some custom installers assemble marble speaker enclosures onsite, works very nicely.
====
Yes, marble, slate, etc., make great, if stationary, speaker cabs. I normally recommend that either a top plate made from one of them or a heavy potted plant be placed on top of the cab to mass load it somewhat. helps with the SAF too.
====
>If you use 3/4 MDF + 3/4 BB ply it will be fine with just minimal bracing. But that's not the subject of the thread...
====
Yep, this is way overkill.
====
>For small speakers with small drivers like the W3-871 I say use 1/2 baltic or danish birch (more layers than domestic birch) plywood, damp it with a couple layers of bitumen/asphalt based "roof repair" tape and use lots of 1/4" plywood triangular braces. If you build the box without the back first, it's quick and easy to put the braces in using hotmelt glue, besides, you want to damp/brace the front of the speaker the most.
====
Nice, but way overkill for a driver with such a modest acoustic pressure potential. The excess doesn't hurt though, so whatever you're comfortable with.
GM
====
So I said already.
====
>I think concrete actually has a higher Q than MDF, it certainly rings at a higher frequency.
====
Hmm, the density of MDF is ~50lbs/ft^3 Vs concrete's ~150lbs/ft^3..... nah, I don't think so.
====
>It's actually a rather poor material for most speaker cabinet applications other than subwoofers.
====
From a purely technical POV, wrong again.
====
>Regardless, your example is not a very fair comparrison because just about anything will be VERY dead sounding if it's damped with tons of earth under it the way a driveway is. Go find a nice flat patch of moist dense ground, press a piece of MDF into it, and you'll find it' doesn't ring much after you wap it.
====
I wasn't comparing the two, only to provide a reference for what defines 'dead'. What most folks claim as 'dead' because it sounds dull really has a Fs right in a sub's upper BW/typical satellite's lower BW. Anyway, MDF will still ring more than the concrete due to their base "Q" difference.
====
>*chuckle* Rosewood and Cocobolo are two of the most prefered woods for xylophones/marimbas because they resonate so well! Exactly what you DON'T want in a speaker cabinet. (unless you have very good damping)
====
*chuckle* is right.
Thanks for proving my point/indirectly endorsing my recommendation, though that obviously wasn't your intention. They resonate so well because they are stiff, which means that they are easily damped due to their high Q/low acoustic amplitude.In the case of a separate mid/tweeter, if they resonate above the mid's passband then they won't be excited since the tweeter has a sealed back.
====
>Brick is substantially better than concrete, if you could get custom cast brick (or even better, porceline) panels they could make superb subwoofer cabinets.
====
Not really, since the joints are the limiting factor and they are pretty weak. Brick construction with a bonding coating works extremely well though, and the preferred construction for large subterranean basshorns.
====
>Some custom installers assemble marble speaker enclosures onsite, works very nicely.
====
Yes, marble, slate, etc., make great, if stationary, speaker cabs. I normally recommend that either a top plate made from one of them or a heavy potted plant be placed on top of the cab to mass load it somewhat. helps with the SAF too.
====
>If you use 3/4 MDF + 3/4 BB ply it will be fine with just minimal bracing. But that's not the subject of the thread...
====
Yep, this is way overkill.
====
>For small speakers with small drivers like the W3-871 I say use 1/2 baltic or danish birch (more layers than domestic birch) plywood, damp it with a couple layers of bitumen/asphalt based "roof repair" tape and use lots of 1/4" plywood triangular braces. If you build the box without the back first, it's quick and easy to put the braces in using hotmelt glue, besides, you want to damp/brace the front of the speaker the most.
====
Nice, but way overkill for a driver with such a modest acoustic pressure potential. The excess doesn't hurt though, so whatever you're comfortable with.
GM
>But I have another idea for a more advanced project: a cabinet with non-parallel sides on the inside but is absolutely rectangular on the outside. This means non-uniform-thickness walls (i.e. a side panel 3/4" thick in the front that tapers to 1/2" in the back). Does this make much of a difference (as far as reducing standing waves)?
====
Not enough taper. You need at least a 2:12 taper ratio to spread out the standing wave amplitudes sufficiently. I recommend you build a box within a box and fill the space with kitty litter, sterilized sand, Portland cement, or similar to ensure that there's no resonances of any consequence.
GM
====
Not enough taper. You need at least a 2:12 taper ratio to spread out the standing wave amplitudes sufficiently. I recommend you build a box within a box and fill the space with kitty litter, sterilized sand, Portland cement, or similar to ensure that there's no resonances of any consequence.
GM
GM said:
First of all, your 2:12 taper ratio can be reduced to 1:6. The example I gave was just to make it easier to "visualize" the idea. If I did taper, I'd probably go with 1.5" tapering to 0.5", with a ratio of 1:3 of 3:1, whichever way you look at it. It can be built like a "normal" cabinet with at least some reduction of standing waves. 1:6 or 6:1 is obviously better but less practical, as far as construction goes.
Much thanks.
>First of all, your 2:12 taper ratio can be reduced to 1:6.
====
True, I'm used to thinking in terms of feet for designing rooms.
====
>The example I gave was just to make it easier to "visualize" the idea. If I did taper, I'd probably go with 1.5" tapering to 0.5", with a ratio of 1:3 of 3:1, whichever way you look at it. It can be built like a "normal" cabinet with at least some reduction of standing waves.
====
It won't reduce standing waves, only attenuate their amplitude a tiny amount. Really, you're better off just using a good damping material and skip the taper.
GM
====
True, I'm used to thinking in terms of feet for designing rooms.
====
>The example I gave was just to make it easier to "visualize" the idea. If I did taper, I'd probably go with 1.5" tapering to 0.5", with a ratio of 1:3 of 3:1, whichever way you look at it. It can be built like a "normal" cabinet with at least some reduction of standing waves.
====
It won't reduce standing waves, only attenuate their amplitude a tiny amount. Really, you're better off just using a good damping material and skip the taper.
GM
A little off topic...
I posted this question some time ago but received no replies, so here it is again:
Would I need to use a Zobel circuit in conjunction with a notch filter for best performance from the 871's? Or is the least electronics the better (performance-wise)? If it matters, I will be using the TB full-range and almost down to its resonant frequency (where the impedance peaks).
I posted this question some time ago but received no replies, so here it is again:
Would I need to use a Zobel circuit in conjunction with a notch filter for best performance from the 871's? Or is the least electronics the better (performance-wise)? If it matters, I will be using the TB full-range and almost down to its resonant frequency (where the impedance peaks).
Sloping front baffle
How about an 871 cabinet with a sloping front baffle? Not for phase alignment (obviously) but for reducing standing waves...
Such a cabinet would be much easier to construct than one of the tapering concept, though you still have to recalculate the volume. The question is should I place the driver near the top of the slope or the bottom?
How about an 871 cabinet with a sloping front baffle? Not for phase alignment (obviously) but for reducing standing waves...
Such a cabinet would be much easier to construct than one of the tapering concept, though you still have to recalculate the volume. The question is should I place the driver near the top of the slope or the bottom?
thanks for your help coolkhoa, I was asking for my younger brother whos building a center channel and I think he decided to go with a 4 driver center channel with an internal volume of 0.8 cubic feet. I'm guessing it would be best to mount the drivers in a close square configuration to get good dispertion on both the x an y axis?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.