Hi Dave,
About MDF and such.
I've used it for years and years, properly braced (as with anything) it works very well. I also seal inside and out so it doesn't disintegrate over time. Ever use high density fiberboard? That stuff throws sparks from a power saw. I used 1". Once it's together, it will ring if you shock it. Again, seal both sides and brace it. That stuff will hold an edge, so use a router on the corners. 1/4" round works fine. It's also really, really heavy.
Anyway, proper sound deadening material along with sound absorbent material and you're good to go. Any speaker cabinet material requires the same care as far as I know. Probably your carbon fiber material will as well. MDF at least has a higher internal damping than does HDF or plywood. Concrete is just plain heavy and rings just as bad - so no advantage there unless you enjoy molding your boxes.
That rapid 3D prototyping system is extremely cool. There will be many industrial modelers out of work in a few years. It might even revolutionize the foundry industry and reduce the costs for all castings. Less end finishing will be required as well. Cool!
-Chris
About MDF and such.
I've used it for years and years, properly braced (as with anything) it works very well. I also seal inside and out so it doesn't disintegrate over time. Ever use high density fiberboard? That stuff throws sparks from a power saw. I used 1". Once it's together, it will ring if you shock it. Again, seal both sides and brace it. That stuff will hold an edge, so use a router on the corners. 1/4" round works fine. It's also really, really heavy.
Anyway, proper sound deadening material along with sound absorbent material and you're good to go. Any speaker cabinet material requires the same care as far as I know. Probably your carbon fiber material will as well. MDF at least has a higher internal damping than does HDF or plywood. Concrete is just plain heavy and rings just as bad - so no advantage there unless you enjoy molding your boxes.
That rapid 3D prototyping system is extremely cool. There will be many industrial modelers out of work in a few years. It might even revolutionize the foundry industry and reduce the costs for all castings. Less end finishing will be required as well. Cool!
-Chris
Hi Scottmoose,
Very often, I end up using two layers of MDF, or MDF and some finish wood to build a box. They are laminated as a result. Yes, this works better as you now have three layers (count the glue). But the thing is, you aren't getting a great deal of panel movement anyway when using MDF and bracing. So many of the worries are really non-issues. The issues now become soaking up the higher frequency back-waves (or deflecting them to one side - strongly). It's the vibrational energy that is a concern now. So surface treatments that damp the box material are useful (think pliable undercoating), then something semi-dense to soak up the midrange and higher bass energy. Any further stuffing may be only to correct for too low a volume.
Right now I'm interested in why MDF is a poor construction material for speaker boxes. I don't agree with that.
-Chris
Very often, I end up using two layers of MDF, or MDF and some finish wood to build a box. They are laminated as a result. Yes, this works better as you now have three layers (count the glue). But the thing is, you aren't getting a great deal of panel movement anyway when using MDF and bracing. So many of the worries are really non-issues. The issues now become soaking up the higher frequency back-waves (or deflecting them to one side - strongly). It's the vibrational energy that is a concern now. So surface treatments that damp the box material are useful (think pliable undercoating), then something semi-dense to soak up the midrange and higher bass energy. Any further stuffing may be only to correct for too low a volume.
Right now I'm interested in why MDF is a poor construction material for speaker boxes. I don't agree with that.
-Chris
Yes, then HDF laminated inside and out with plastic laminate because i wasn't wholly satisfied. BB is better, cheaper, and easier to use,
dave
Not sure that it has as good damping as ply (ref Ron Clarke), but it does have greater energy storage and oozes a continual low level time smeared signal that buries the subtle details that give you downward dynamic range,
dave
>>> it does have greater energy storage and oozes a continual low level time smeared signal that buries the subtle details that give you downward dynamic range...
LOL... I built a few boxes for the Fostex 168s (older style with wizzer). Two were ported mdf boxes and two were ported pine boxes. The pine sounded better. They sounded livelier with better decay on instruments. I could only tell the difference when comparing directly. All of these projects sounded pretty good to me.
I built my BIBs for the same driver (but switched to the 165k eventually) from a combination of pine, mdf, particle board and ply (don't ask)... and it sounds good to me.
I built a back horn for the same driver from mdf and thought it sounded bad. It was probably a combination of mdf and poor design... but i learned from it.
I built my OBs using the new TB 8" with ply and can't imagine using mdf. Just a feeling. I wanted to use pine but could not find big enough boards that were flat/straight.
I must comment when working with pine or ply... sanding, screwing, glueing... the box resonates a tone when you knock on it even when the cabinet is half complete. There is a life to these boxes more like a musical instrument.
Imagine a piano made from mdf compared to a piano made from ply, spruce and other choice woods... piano's are built with wood that gives them life! Imagine a violin from mdf... I only use mdf for speaker bases.
LOL... I built a few boxes for the Fostex 168s (older style with wizzer). Two were ported mdf boxes and two were ported pine boxes. The pine sounded better. They sounded livelier with better decay on instruments. I could only tell the difference when comparing directly. All of these projects sounded pretty good to me.
I built my BIBs for the same driver (but switched to the 165k eventually) from a combination of pine, mdf, particle board and ply (don't ask)... and it sounds good to me.
I built a back horn for the same driver from mdf and thought it sounded bad. It was probably a combination of mdf and poor design... but i learned from it.
I built my OBs using the new TB 8" with ply and can't imagine using mdf. Just a feeling. I wanted to use pine but could not find big enough boards that were flat/straight.
I must comment when working with pine or ply... sanding, screwing, glueing... the box resonates a tone when you knock on it even when the cabinet is half complete. There is a life to these boxes more like a musical instrument.
Imagine a piano made from mdf compared to a piano made from ply, spruce and other choice woods... piano's are built with wood that gives them life! Imagine a violin from mdf... I only use mdf for speaker bases.
A higher Q resonance which is less audiable & less likely to get excited by music... the higher in frequency the better. Once excited it should die pretty quickly.
dave
I think you have this backwards. High Q = low loss = long time ringing. Just as in a tank circuit in radio theory. Electrical equivalent is LC resonant circuit where the coil has low R.
Plywood, being laminated, is lossy as it dissipates energy as heat in the layers. Hence low Q resonance, which dies away quickly. Wider bandwidth though.
I think MDF is less lossy and therefore likely to have high Q resonances that hang around longer. But high Q = narrower bandwidth.
Better to have any panel resonances below the midrange frequencies and well damped (low Q)
e.g. BBC thin wall cabinets use thin ply and bituminous damping to increase the loss factor and lower the panel Q.
Plywood, being laminated, is lossy as it dissipates energy as heat in the layers. Hence low Q resonance, which dies away quickly. Wider bandwidth though.
I think MDF is less lossy and therefore likely to have high Q resonances that hang around longer. But high Q = narrower bandwidth.
Better to have any panel resonances below the midrange frequencies and well damped (low Q)
e.g. BBC thin wall cabinets use thin ply and bituminous damping to increase the loss factor and lower the panel Q.
Agreed re the plywood (or any material made from many layers) as they exhibit significant internal boundary losses, so inherently possess a greater degree of self-damping than a material with fewer or a single layer. However, I disagree with the concept that a panel's resonant BW should be located below the midband.
Assuming the desire is to minimise panel excitation, that means pushing its Fs (& thus its resonant BW) into a region where less energy is present, i.e. above or below the functional BW of the enclosure. For bass cabinets, pushing it below this region is almost impossible, given that each doubling of a panel's thickness (& therefore mass) only drops Fs 0.707x that of the original thickness, and worse, as GM notes elsewhere, as Fs decreases so the resonant BW and amplitude of the panel increases. The other alternative is to push panel Fs up, above the functional BW of the cabinet. Birch / marine / apple ply etc. as sheet materials go are quite good at this trick, esp. when suitably braced; as well as possessing considerably more in the way of internal losses, they are substantially stiffer than MDF for a given thickness, so it is much easier to lift their resonant BW up out of a zone where it may be subject to excitation. A reasonable gauge of aluminium or CRS would be even better; they ring like a bell to be sure, but not in any place where they're likely to be excited, thus making for a very efficient box as panel damping can be reduced to a minimum. Either way, for me, over-reliance on panel damping is essentially sloppy design; providing you've used extremely stiff walls, suitably braced where required to get the panel resonant BW well up, out of the way, then what little audible resonances are left (if any) are easily damped with a minimum of material.
Assuming the desire is to minimise panel excitation, that means pushing its Fs (& thus its resonant BW) into a region where less energy is present, i.e. above or below the functional BW of the enclosure. For bass cabinets, pushing it below this region is almost impossible, given that each doubling of a panel's thickness (& therefore mass) only drops Fs 0.707x that of the original thickness, and worse, as GM notes elsewhere, as Fs decreases so the resonant BW and amplitude of the panel increases. The other alternative is to push panel Fs up, above the functional BW of the cabinet. Birch / marine / apple ply etc. as sheet materials go are quite good at this trick, esp. when suitably braced; as well as possessing considerably more in the way of internal losses, they are substantially stiffer than MDF for a given thickness, so it is much easier to lift their resonant BW up out of a zone where it may be subject to excitation. A reasonable gauge of aluminium or CRS would be even better; they ring like a bell to be sure, but not in any place where they're likely to be excited, thus making for a very efficient box as panel damping can be reduced to a minimum. Either way, for me, over-reliance on panel damping is essentially sloppy design; providing you've used extremely stiff walls, suitably braced where required to get the panel resonant BW well up, out of the way, then what little audible resonances are left (if any) are easily damped with a minimum of material.
Last edited:
The BBC did do that, something i gave up on a while ago. Adding mass without stiffness is usually counter-productive
The ringing in a properly done ply cabinet may ring for a number of cycles, but because they are placed at very high frequency, it does not last long. Because it is at high frequencies, and is high Q (energy has to be injected over a very narrow bandwidth and for quite a few cycles), it is very unlikely that sufficient energy ever enters the panel to cause it to ring. In essense giving a non-resonant box.
There is much more energy available at low frequencies and as Scott points out you end up with Ouroboros.
Toole shows that high Q resonances are less audiable than low Q resonances.
I agree on plywood... this is what Ron Clarke pointed out, and being in his field to know these things (lives depend on it) i cite him.
dave
Hi Dave,
I generally agree with you on high Q, high resonance for the box. That's where the HDF stuff really excelled. They sounded the very best of any construction method.
The secret to a good box is simple. Stiff as heck and braced. Now there won't be much vibration to begin with, and poor coupling with an exciting force. Add surface damping to that (like undercoating) and you have a pretty dead system. Add something to absorb midrange frequencies and now you don't have any other sound bouncing back out through the woofer cone material. The first coating applied to the box inside serves to seal the wood against moisture, making MDF last forever without crumbling. The HDF has high glue content I think, but I coat it anyway, why take a chance?
If you do make a box this way, the panels will not contribute to the sound output. It will be an inert box. Isn't that what we want anyway?
I have built with void less birch ply before. Yes, it sounded livelier. The reason is simple though, you can feel the sides vibrate a great deal more than MDF or HDF. The energy is stored and released, not damped. You could make a good box this way, but it would require more bracing and careful damping. On the plus side, the box looks a whole lot better, and it's way lighter! Because of the lower mass, solid wood or even plywood is easier to excite than a stiff, heavy material. This simply means there is more work required to make those boxes as inert as MDF or HDF types.
Now, how inert do you think an MDF box (untreated with anything) would be if saturated with moisture? That's just before it proceeds to disassemble itself.
-Chris
I generally agree with you on high Q, high resonance for the box. That's where the HDF stuff really excelled. They sounded the very best of any construction method.
The secret to a good box is simple. Stiff as heck and braced. Now there won't be much vibration to begin with, and poor coupling with an exciting force. Add surface damping to that (like undercoating) and you have a pretty dead system. Add something to absorb midrange frequencies and now you don't have any other sound bouncing back out through the woofer cone material. The first coating applied to the box inside serves to seal the wood against moisture, making MDF last forever without crumbling. The HDF has high glue content I think, but I coat it anyway, why take a chance?
If you do make a box this way, the panels will not contribute to the sound output. It will be an inert box. Isn't that what we want anyway?
I have built with void less birch ply before. Yes, it sounded livelier. The reason is simple though, you can feel the sides vibrate a great deal more than MDF or HDF. The energy is stored and released, not damped. You could make a good box this way, but it would require more bracing and careful damping. On the plus side, the box looks a whole lot better, and it's way lighter! Because of the lower mass, solid wood or even plywood is easier to excite than a stiff, heavy material. This simply means there is more work required to make those boxes as inert as MDF or HDF types.
Now, how inert do you think an MDF box (untreated with anything) would be if saturated with moisture? That's just before it proceeds to disassemble itself.
-Chris
And then there's always the blind listening test - which we did several years back at a DIY get together - two pairs of identical speaker enclosures of a bipole MLTL with CSS FR125 drivers. Some of the audience certainly preferred the "weight" and "bloom" that the MDF cabinets added, but not all.
I'm among those who chose to ignore the masturbatory arguments that others so enjoy over the technical arcana of any specific aspect of the the audio (re)production chain, and concentrate on the subjective "smile" test.
The plywood and occasional solid wood enclosures / OB panels that I've heard during the past decade certainly pass that with flying colours for me
As this is hobby is DIY (i.e. for yourself ) not DIMWIT (Do It My Way - Ignorant Tyro) that is implied by the tone to which these discussions sometimes devolve, shouldn't that be good enough for any of us?
I'm among those who chose to ignore the masturbatory arguments that others so enjoy over the technical arcana of any specific aspect of the the audio (re)production chain, and concentrate on the subjective "smile" test.
The plywood and occasional solid wood enclosures / OB panels that I've heard during the past decade certainly pass that with flying colours for me
As this is hobby is DIY (i.e. for yourself ) not DIMWIT (Do It My Way - Ignorant Tyro) that is implied by the tone to which these discussions sometimes devolve, shouldn't that be good enough for any of us?
Hi Chris,
Well, the problem with that listening test is simply, "how was each cabinet finished inside?". I'm not arguing one way or the other here. I'm only pointing out that the test may not have had all the variables controlled. Your comment about MDF cabinets adding "bloom" to the sound clearly shows that they were not prepared properly.
Once you attempt to compare things that are not completed properly, those tests are unfortunately a waste of everyone's time - or worse. Once "lessons learned" are agreed upon (or not), a falsehood has been created. Audio lore.
If you read what I said earlier, and perhaps this point wasn't made clear enough. Any cabinet material (within reason) can be used to make very good enclosures. Some will require more care than others to attain the expected level of performance. What that means is that you can use solid light pine with no knots to make a box. You'll probably have to make it really thick, or laminate many sheets together to achieve the same performance of another material.
Your listening test point is valid, but only if all the factors are controlled and known. Also, it's important to note that comparing imperfect items, the consensus doesn't break down into one performs better than the other. Instead you begin selecting the item along the lines of which shortcomings can you live with. That's an enormous difference from what may be considered "better" than something else.
The other factor that isn't considered when listening tests are performed is that your subjects may be choosing what is closest in sound to what they live with on a day to day basis. The factors that can affect how a decision is made are many, poorly understood by many organizers of these tests and completely uncontrolled. Therefore, unless the listening test is rigorously set up and controlled by people who know how to do this, they are little more than a curiosity. Often the results may show more about how the test is run than the intended subject. The results are only valid for that specific situation. Too often the results are propagated across the web as if the testing was done in complete agreement with a scientific procedure. Not too many are, and even tests conducted by professionals always include statements to show uncontrolled variables and their expected effects. They also take the results and run statistics on them to validate whether or not the outcome seems truly valid. With ad hoc testing, the results are normally quoted as absolute fact with no checks run on the data. Yipes!
To address another concern you may have, scientific tests with instruments are not a complete test in audio. Listening tests are required as well. The cool thing about measuring speakers / cabinets / rooms is that what you actually can hear can be explained by those measurements. Study the information at the NRC labs on audio about correlation between measurements and predicted listening panel results. Floyd Toole did some great work there, and he was correct as the empirical tests showed. Something to have pride in as this was a Canadian effort.
-Chris
Well, the problem with that listening test is simply, "how was each cabinet finished inside?". I'm not arguing one way or the other here. I'm only pointing out that the test may not have had all the variables controlled. Your comment about MDF cabinets adding "bloom" to the sound clearly shows that they were not prepared properly.
Once you attempt to compare things that are not completed properly, those tests are unfortunately a waste of everyone's time - or worse. Once "lessons learned" are agreed upon (or not), a falsehood has been created. Audio lore.
If you read what I said earlier, and perhaps this point wasn't made clear enough. Any cabinet material (within reason) can be used to make very good enclosures. Some will require more care than others to attain the expected level of performance. What that means is that you can use solid light pine with no knots to make a box. You'll probably have to make it really thick, or laminate many sheets together to achieve the same performance of another material.
Your listening test point is valid, but only if all the factors are controlled and known. Also, it's important to note that comparing imperfect items, the consensus doesn't break down into one performs better than the other. Instead you begin selecting the item along the lines of which shortcomings can you live with. That's an enormous difference from what may be considered "better" than something else.
The other factor that isn't considered when listening tests are performed is that your subjects may be choosing what is closest in sound to what they live with on a day to day basis. The factors that can affect how a decision is made are many, poorly understood by many organizers of these tests and completely uncontrolled. Therefore, unless the listening test is rigorously set up and controlled by people who know how to do this, they are little more than a curiosity. Often the results may show more about how the test is run than the intended subject. The results are only valid for that specific situation. Too often the results are propagated across the web as if the testing was done in complete agreement with a scientific procedure. Not too many are, and even tests conducted by professionals always include statements to show uncontrolled variables and their expected effects. They also take the results and run statistics on them to validate whether or not the outcome seems truly valid. With ad hoc testing, the results are normally quoted as absolute fact with no checks run on the data. Yipes!
To address another concern you may have, scientific tests with instruments are not a complete test in audio. Listening tests are required as well. The cool thing about measuring speakers / cabinets / rooms is that what you actually can hear can be explained by those measurements. Study the information at the NRC labs on audio about correlation between measurements and predicted listening panel results. Floyd Toole did some great work there, and he was correct as the empirical tests showed. Something to have pride in as this was a Canadian effort.
-Chris
Not all speaker designs are easy to finish inside after they are built and applying finish during construction just ads to the complexity. Secondly, building in MDF usually implies applying an external finish like veneer, or high quality paint job, which ads to the cost.
I personally don't doubt there are differences in sound between MDF and BB, good or bad, but I suspect a lot of DIY types will choose the most economical and easiest option to meet their objectives and not worry about subtle sound differences in the finished product.
My $.02
The key question is of course, where are the resonance occuring?
For e.g., one of the reasons I regard the knuckle-rap test so beloved of reviewers etc. as deeply misleading is that if the box sounds 'dead' to the aforementioned belt with the bones, that almost invariably means its resonant BW is way too low, slap-bang in the middle of the box operating region. You whack a plywood (or a metal) box in the same way, yep, it rings. Substantially above the region the cabinet is functioning over, and therefore the energy available to cause excitation is minimal.
MDF / HDF is not actually all that well damped; it's high mass to be sure, but just high enough when coupled with its modest MOE / stiffness value to cause some headaches for most bass boxes. Typically, it requires substantially more panel damping than, say, the infamous BB ply because of this lower Fs; if you don't apply enough, then the low level, later timed panel emmitance is what tends to 'suck the life' out of the sound; a not uncommon observation about MDF boxes. For me, this is dodgy engineering though; give me a more efficient approach any day. Admittedly I'm biased as my engineering idols (Sir William White, Sir Stanley Hooker, Sir Frank Whittle, Sir Sydney Camm, Malcolm Sayer and Colin Chapman) all tended to take this approach.
Which isn't to say I think it's impossible to use MDF; I think we can all be a bit excessive sometimes, and if it's a choice of that or nothing, I'd take it like a shot. Although I'd rather it was cut by machine, surrounded by extractors, while I am elsewhere. It beats not having any sound, even if it might not be ideal.
For e.g., one of the reasons I regard the knuckle-rap test so beloved of reviewers etc. as deeply misleading is that if the box sounds 'dead' to the aforementioned belt with the bones, that almost invariably means its resonant BW is way too low, slap-bang in the middle of the box operating region. You whack a plywood (or a metal) box in the same way, yep, it rings. Substantially above the region the cabinet is functioning over, and therefore the energy available to cause excitation is minimal.
MDF / HDF is not actually all that well damped; it's high mass to be sure, but just high enough when coupled with its modest MOE / stiffness value to cause some headaches for most bass boxes. Typically, it requires substantially more panel damping than, say, the infamous BB ply because of this lower Fs; if you don't apply enough, then the low level, later timed panel emmitance is what tends to 'suck the life' out of the sound; a not uncommon observation about MDF boxes. For me, this is dodgy engineering though; give me a more efficient approach any day. Admittedly I'm biased as my engineering idols (Sir William White, Sir Stanley Hooker, Sir Frank Whittle, Sir Sydney Camm, Malcolm Sayer and Colin Chapman) all tended to take this approach.
Which isn't to say I think it's impossible to use MDF; I think we can all be a bit excessive sometimes, and if it's a choice of that or nothing, I'd take it like a shot. Although I'd rather it was cut by machine, surrounded by extractors, while I am elsewhere. It beats not having any sound, even if it might not be ideal.
well, to answer anatech's question:
Chris; I don't finish the inside surfaces of my speaker boxes, and frankly I can't remember seeing it on any of the commercial enclosures I've had occasion to poke inside.
So in regards to the aforementioned and very casual listening session (that would not pass anyone's standards for a rigorous scientifically controlled test), both pairs were equally handicapped by this deficiency.
If I was to consider doing so on any future builds, would you anticipate any problems in using different material types as sealant and exterior top /finish coat(s)? I tend to apply paper backed veneers with the iron-one wood glue method, with 2 or 3 coats of sprayed nitrocellulose lacquer - with or without solvent based stain colorants.
edit: and to all - apologies for the pizzy tone of my earlier post - I'm currently on an interesting mix of meds for separate health issues of sciatia and "male plumbing".
Chris; I don't finish the inside surfaces of my speaker boxes, and frankly I can't remember seeing it on any of the commercial enclosures I've had occasion to poke inside.
So in regards to the aforementioned and very casual listening session (that would not pass anyone's standards for a rigorous scientifically controlled test), both pairs were equally handicapped by this deficiency.
If I was to consider doing so on any future builds, would you anticipate any problems in using different material types as sealant and exterior top /finish coat(s)? I tend to apply paper backed veneers with the iron-one wood glue method, with 2 or 3 coats of sprayed nitrocellulose lacquer - with or without solvent based stain colorants.
edit: and to all - apologies for the pizzy tone of my earlier post - I'm currently on an interesting mix of meds for separate health issues of sciatia and "male plumbing".
Last edited:
Right now Im building 3way monitor speakers using good ol fashioned particle board
And they are to be placed in a bookshelve
So much for "high end"
The money saved can be used on other projects
Actually I would say its all in the construction technique
Lots of ways to improve a construction using multiple materials
Even the glue, and the way its used, is important
Instead of plywood, I would make my own
19mm particle board in the middle
6mm MDF on one side(inner)
6mm plywood on other(outer)
And thats just one option
And I would suggest to "paint" all lthe inside with rubber sealing
On the outer side(finish), most will choose by the looks
And its already compromised from the very first choise, looks/design
What would be better for the sound wouldnt look good enough
All in all, noone will go all the way anyway
And they are to be placed in a bookshelve
So much for "high end"
The money saved can be used on other projects
Actually I would say its all in the construction technique
Lots of ways to improve a construction using multiple materials
Even the glue, and the way its used, is important
Instead of plywood, I would make my own
19mm particle board in the middle
6mm MDF on one side(inner)
6mm plywood on other(outer)
And thats just one option
And I would suggest to "paint" all lthe inside with rubber sealing
On the outer side(finish), most will choose by the looks
And its already compromised from the very first choise, looks/design
What would be better for the sound wouldnt look good enough
All in all, noone will go all the way anyway
Hi Scottmoose, what you say is fine if the enclosure is dealing with limited bandwidth, ie just a bass box. But most boxes have to cope with full range, and the midrange is the area where our hearing is most acute. You'd have to be careful about pushing the resonance into that range.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.