Greg B said:
I think the point you are missing here is that both MDF and plywood will resonate.
Don't drink the marketing koolaid that MDF is inert and non-resonant. It isn't, not by a long shot. And the resonance it makes is decidedly unpleasant.
You're right that both MDF and Plywood resonate. MDF has 3x the damping of BB, so wouldn't MDF make better full range enclosure material?
R Hosch seems to agree with my subjective impression that the Q of MDF's primary resonant mode is higher than that of plywood's which implies that more careful attention may need to be paid to controlling that specific behavior in MDF to approximate subjective neutrality rather than going with a slight tweak in xover values you might get away with using BB.
There also appears to be no serious question that the sonic characteristics of the adhesive that holds the product together is much more predominant in MDF than in plywood.
There also appears to be no serious question that the sonic characteristics of the adhesive that holds the product together is much more predominant in MDF than in plywood.
I would expect self damping to be similar between the two. It is difficult to say exactly without knowing more details about the resin system used. In typical composite behavior, the fiber/resin ratio doesn't tend to have a large impact on the damping until extremes (in either direction) are reached. But a change in resin systems might have a noticeabe.tiroth said:RHosch,
You are obviously an expert in these matters. I was trying to compare HDF and MDF. From the figures I saw, HDF is about 30-50% denser and nearly 70% stiffer than MDF. How are the damping properties affected by this?
Should be pretty damned dead. There might be simpler construction techniques that achieve the same goal (perhaps with thicker walls or more weight as the penalty), but what you outline will certainly work well.I was planning to build a constrained layer box as below. Is this a good idea? Should a choose a different material than HDF?
I wouldn't recommend a flexible adhesive on purpose, since the torsional stiffness lost in the joint would significantly reduce the panel mode (resonant frequency). Any gain in damping in the joint would be wasted on that lower mode. Instead, concentrate on keeping stiffness where it is important (joints are critical) and add damping designed specifically to tackle that issue.My final question is, in this type of construction, does choosing a flexible glue line adhesive such as liquid nails or even contact cement make a difference compared to a hard-setting glue like polyurethane?
Okay, some very interesting info re MDF v Plywood and even that the adhesive may impact sound?
Now, if you were building the following (any one or all!) four types of cabinet what would be your preferred material/construction/thickness/bracing?
1) Small bookshelf – say two way
2) Small – medium Floor standing – 2 or 3 way
3) Large floor standing – 3 way or more
4) Subwoofer
Now do not ask be to define these sizes or enclosures (sealed, BR’s or TL’s etc) I’ll leave that for you to clarify for your construction???
Now, if you were building the following (any one or all!) four types of cabinet what would be your preferred material/construction/thickness/bracing?
1) Small bookshelf – say two way
2) Small – medium Floor standing – 2 or 3 way
3) Large floor standing – 3 way or more
4) Subwoofer
Now do not ask be to define these sizes or enclosures (sealed, BR’s or TL’s etc) I’ll leave that for you to clarify for your construction???
thoriated said:R Hosch seems to agree with my subjective impression that the Q of MDF's primary resonant mode is higher than that of plywood's...
While the Q may be a bit higher, the amplitude of resonance is significantly lower. Which is worse... a narrow 1dB peak, or a broad 3dB peak? Seems obvious to me!
There also appears to be no serious question that the sonic characteristics of the adhesive that holds the product together is much more predominant in MDF than in plywood.
Of course... there is a lot more adhesive in MDF than in plywood. You speak as if that is a bad thing sonically though.
Audiophiles are a strange lot. For some reason popular opinion is that high resin composites like MDF don't sound as good as low resin composites like plywood. Yet, I've seen numerous posts on this forum suggesting that a solid poured resin enclosure would be the ultimate! Interesting logic.
RHosch said:
Audiophiles are a strange lot. For some reason popular opinion is that high resin composites like MDF don't sound as good as low resin composites like plywood. Yet, I've seen numerous posts on this forum suggesting that a solid poured resin enclosure would be the ultimate! Interesting logic.
A few similar discussions have been going on at the Madisound forum lately, and people over there primarily favor MDF over BB. I wonder why this forum favors BB and that forum favors MDF?
FWIW, I've got four sheets of 3/4" plywood saved up for my reference speaker design that I made with a center constrained damping viscoelastic layer gluing together sheets of 3/8" plywood using a generous quantity of a high solids contact cement.
I guess part of my preference for plywood results from my trying to squeeze the most performance out of the smallest package. For instance, I'm currently well along in the process of developing a pair of speakers I call (Hoffman's) 'Iron Lawbreakers', aka 'The World's Largest Bookshelf Speakers' using Altec 288G's and JBL 2220J's that give the same midband efficiency as the classic A-7 VOTTs along with flat bass response to 50hz in half space with a box having an internal volume just over 3 cu ft. I use a 32hz box tuning along with a special passive xover trick to extend useable LF response by half an octave without compromising SQ to achieve this.
I guess part of my preference for plywood results from my trying to squeeze the most performance out of the smallest package. For instance, I'm currently well along in the process of developing a pair of speakers I call (Hoffman's) 'Iron Lawbreakers', aka 'The World's Largest Bookshelf Speakers' using Altec 288G's and JBL 2220J's that give the same midband efficiency as the classic A-7 VOTTs along with flat bass response to 50hz in half space with a box having an internal volume just over 3 cu ft. I use a 32hz box tuning along with a special passive xover trick to extend useable LF response by half an octave without compromising SQ to achieve this.
greenie512 said:Okay, some very interesting info re MDF v Plywood and even that the adhesive may impact sound?
Now, if you were building the following (any one or all!) four types of cabinet what would be your preferred material/construction/thickness/bracing?
1) Small bookshelf – say two way
2) Small – medium Floor standing – 2 or 3 way
3) Large floor standing – 3 way or more
4) Subwoofer
Now do not ask be to define these sizes or enclosures (sealed, BR’s or TL’s etc) I’ll leave that for you to clarify for your construction???
I'll give this a shot.
For 1-3, if the walls are made of MDF, use BB for internal bracing (to increase rigidity). If the walls are made from BB, line the walls with good damping material. Or use a composite of BB and MDF. Be sure the rear wall is especially well damped. Double up the front and rear baffles (1.5"+ front baffle, 1.5" rear).
For 2-3, consider adding 1-2 vertical internal braces, placed at intervals along the golden mean (or a similar ratio). For 3, consider using plywood as the primary construction material (because of the large woofer) but don't forget to damp the enclosure well.
For 1-3, the key is to be sure that the enclosure is rigid AND well damped. This can be accomplished using either MDF or baltic birch.
For 4, just build an extremely rigid BB enclosure. 1.5" plywood throughout, with some plywood bracing. Damping is unnecessary here.
I've gone to using internal ribbed bracing around the perimeter of the cabinet every foot to foot and a half of height made from 3/4" BB about 2 1/2" wide. This is for a largish cabinet, of course. I've also taken to gluing and screwing 1x3" oak boards on edge for top and bottom panels. That approach could even be taken further by rigidly attaching 1" x 1/8" steel stock with screws every couple inches to the inside edges of these braces, but I haven't quite motivated myself to do that yet.
I've also decided to resolve the front panel bracing problem by doing away with dead front panel area as much as possible and having on edge internal bracing between each driver cutout (see pictures of my basement blasters in the 'system pictures and descriptions' section for this idea put into practice). This also leads to a less visibly obtrusive speaker that some say can image better due to its relatively narrow (& deep) profile.
I've also decided to resolve the front panel bracing problem by doing away with dead front panel area as much as possible and having on edge internal bracing between each driver cutout (see pictures of my basement blasters in the 'system pictures and descriptions' section for this idea put into practice). This also leads to a less visibly obtrusive speaker that some say can image better due to its relatively narrow (& deep) profile.
Well, I'm curious. Any more details you're willing to provide? What flares are you using on the Altecs?thoriated said:I guess part of my preference for plywood results from my trying to squeeze the most performance out of the smallest package. For instance, I'm currently well along in the process of developing a pair of speakers I call (Hoffman's) 'Iron Lawbreakers', aka 'The World's Largest Bookshelf Speakers' using Altec 288G's and JBL 2220J's that give the same midband efficiency as the classic A-7 VOTTs along with flat bass response to 50hz in half space with a box having an internal volume just over 3 cu ft. I use a 32hz box tuning along with a special passive xover trick to extend useable LF response by half an octave without compromising SQ to achieve this.
Hi, Brett -
Well, you might laugh at this, but I'm using a pair of NOS 'Pyle Pro' aluminum flat front 40 x90 horns I picked up cheap last year. I had the 288's laying around for quite a while (they were superceded by TAD 4001's in the project I had planned for them) & forget why I got the 2220J's but I figured 'why not do something fun?' with them a few months ago.
These Pyle Pro's, a heavy aluminum casting, were superceded for sound reinforcement use by CD horns with diffraction edges like the 2380, but are roughly the same size overall (read: much more compact than a 511E) and with a smaller than 2" throat. In Audioasylum, a Professor Earl Geddes has made a strong case that the most natural sounding horn or 'waveguide' wouldn't have discontinuities anywhere along its flare, among other things, making these PPs look like a more likely choice for decent SQ than one might think. Actually, they don't do too badly with the 288-16G's I have.
Also, the HF setback with this selection works really well with the xover approach I favor where it is easy to peak group delay right where the woofer xover rolloff starts. Net result is that time alignment with the HF horn is achievable with as few as three components in the LF xover along with an improvement in group delay linearity to as low as 100hz.
I'm on record as not really 'liking' Bass Reflex sound, but if the box tuning is pushed down to 20-32hz and the port response there is down 5-9 db, then a LC network can be added that actually extracts several db of voltage sensitivity out of the upper BR impedance peak. The net result is a third to half an octave of subjective bass extension compared to standard BR tuning with an effective subsonic filter and the port transient response being far less of an issue, and extracting some improvement in LF group delay down to the standard Fb for a box that size is quite possible.
Well, you might laugh at this, but I'm using a pair of NOS 'Pyle Pro' aluminum flat front 40 x90 horns I picked up cheap last year. I had the 288's laying around for quite a while (they were superceded by TAD 4001's in the project I had planned for them) & forget why I got the 2220J's but I figured 'why not do something fun?' with them a few months ago.
These Pyle Pro's, a heavy aluminum casting, were superceded for sound reinforcement use by CD horns with diffraction edges like the 2380, but are roughly the same size overall (read: much more compact than a 511E) and with a smaller than 2" throat. In Audioasylum, a Professor Earl Geddes has made a strong case that the most natural sounding horn or 'waveguide' wouldn't have discontinuities anywhere along its flare, among other things, making these PPs look like a more likely choice for decent SQ than one might think. Actually, they don't do too badly with the 288-16G's I have.
Also, the HF setback with this selection works really well with the xover approach I favor where it is easy to peak group delay right where the woofer xover rolloff starts. Net result is that time alignment with the HF horn is achievable with as few as three components in the LF xover along with an improvement in group delay linearity to as low as 100hz.
I'm on record as not really 'liking' Bass Reflex sound, but if the box tuning is pushed down to 20-32hz and the port response there is down 5-9 db, then a LC network can be added that actually extracts several db of voltage sensitivity out of the upper BR impedance peak. The net result is a third to half an octave of subjective bass extension compared to standard BR tuning with an effective subsonic filter and the port transient response being far less of an issue, and extracting some improvement in LF group delay down to the standard Fb for a box that size is quite possible.
Your construction method concerns me a bit... with a non-rigid middle layer, your walls have the a stiffness double that of 3/8" plywood, which is only 25% that of 3/4" plywood!thoriated said:FWIW, I've got four sheets of 3/4" plywood saved up for my reference speaker design that I made with a center constrained damping viscoelastic layer gluing together sheets of 3/8" plywood using a generous quantity of a high solids contact cement.
When the plys are held together rigidly (kept from sliding against each other as you try to bend it), stiffness increases with the cube (n^3) of panel thickness. When you adhere them with a damping layer with low shear strength, you lose much of the strength benefit of the additional thickness.
2 x .375^3 = .105
1 x .75^3 = .422
By my thinking, you'd have to use 6-8 3/8" ply layers (2 1/2" thick walls!) in your design to equal the strength of one 3/4" plywood sheet, though of course your damping would be far superior.
2 x .375^3 = .105
Would that calculation assume that there is NO bonding between the sheets? I do agree that it is unlikely that two fabricated sheets of 3/8" would be a strong as the "factory team's" 3/4".
bwbass said:
Your construction method concerns me a bit... with a non-rigid middle layer, your walls have the a stiffness double that of 3/8" plywood, which is only 25% that of 3/4" plywood!
When the plys are held together rigidly (kept from sliding against each other as you try to bend it), stiffness increases with the cube (n^3) of panel thickness. When you adhere them with a damping layer with low shear strength, you lose much of the strength benefit of the additional thickness.
2 x .375^3 = .105
1 x .75^3 = .422
By my thinking, you'd have to use 6-8 3/8" ply layers (2 1/2" thick walls!) in your design to equal the strength of one 3/4" plywood sheet, though of course your damping would be far superior.
Very interesting.
But if two 3/8" pieces of plywood were glued directly together with wood glue, wouldn't they be as strong as one 3/4" piece, given the strength of wood glue?
thoriated said:every foot to foot and a half of height made from 3/4" BB about 2 1/2" wide.
At a minimum i use a similar scheme, but rarely leave that much space between braces.
dave
Yes, in my simplified numbers I'm assuming that the "viscoelastic" damping sheet and rubber cement have no shear strength compared to wood glue. Seeing as how these materials are engineered to "give" in order to absorb vibrations, I think it's not too far off for comparison.
If you laminated another 3/8" layer directly to the outer panel with wood glue, you'd have all the necessary strength and excellent isolation and damping.
It would probably be easier to do this when the existing sandwich panels are cut so size, as it would be tricky to clamp a full sheets without a vacuum setup.
It would probably be easier to do this when the existing sandwich panels are cut so size, as it would be tricky to clamp a full sheets without a vacuum setup.
A common CLD construction technique seems to be to build a box, laminate it with just such a low-shear-strength compound, then laminate that with the outer layer of rigid material. Is this really a good idea? Will the outer box benefit from bracing on the inner box in any way, or will it look like a large unconstrained panel?
If the braces were attached through the damping layer, though, then I would guess all the vibrations would transmit through to the outer box.
If the braces were attached through the damping layer, though, then I would guess all the vibrations would transmit through to the outer box.
tiroth said:A common CLD construction technique seems to be to build a box, laminate it with just such a low-shear-strength compound, then laminate that with the outer layer of rigid material. Is this really a good idea? Will the outer box benefit from bracing on the inner box in any way, or will it look like a large unconstrained panel?
If the braces were attached through the damping layer, though, then I would guess all the vibrations would transmit through to the outer box.
I think it depends on how rigid each layer is. If the outer box and the inner box are each relatively thin (say, less than 3/4"), you might have trouble with rigidity. But if you build a rigid outer layer (3/4" BB or more), then build a thin inner layer and place a damping layer between them, you'd have a good compromise of rigidity and damping.
You're right about the bracing -- if you used a 3/4" outer layer, a 1/4" inner layer, and just braced the inner layer, then you wouldn't be adding much rigidity to the box. Let's say the outer layer has a rigidity of 1 unit. The inner box would only have a rigidity of .037 (27 times smaller), so even if you braced the heck out of it, you wouldn't get it anywhere near a "helpful" rigidity. (Don't worry about the absolute values of these numbers -- the ratio between them is accurate.) You would only benefit by bracing the primary walls.
Well, my initial thought then is that it is better to brace the exterior box and suffer some additional transmission rather than allow it to flop about unbraced. Full panel braces every 4-6" add considerable rigidity, right? And since the brace connection points represent the most rigid parts of the outer box, wouldn't these in turn be the least important points to worry about transmitting energy to?
The enclosure then ends up looking sort of like an ice cube tray, with the damped pieces "floating" on their decoupling layer between the braces and main panel. Would this still reduce the excitation of the unbraced expanses of panel?
The enclosure then ends up looking sort of like an ice cube tray, with the damped pieces "floating" on their decoupling layer between the braces and main panel. Would this still reduce the excitation of the unbraced expanses of panel?
- Status
- Not open for further replies.
- Home
- Loudspeakers
- Multi-Way
- MDF v Plywood construction