There are a few people that follow a doctrine of not worrying about "deadening" the speaker box, and insted they make it "breathe" with the music, much like the wood in a nice violine or chello. Others believe in making the box as innert as humanly possible (while others do not use boxes at all
). I want to build the innert veriety, and in true DIY spirit, I want to over-build the heck out of it. The question is, how much is too much? MDF resonates at one frequency and ply at another, so sandwitch construction with braing should be good... But what if your sandwitch has 6 layers? I am not talking about the front baffle, as it has its own considirations. But all the sides, i think, can benefit from the extra mass. Also, it seems joining such thick panels together should be very easy, since the surface area of the joints is so large... Perhaps another benefit is that with joints being as large as the cavity itself, minimal bracing will be needed. Am I wrong? Will I just end up with a wooden mess? Thanks in advance.
). I want to build the innert veriety, and in true DIY spirit, I want to over-build the heck out of it. The question is, how much is too much? MDF resonates at one frequency and ply at another, so sandwitch construction with braing should be good... But what if your sandwitch has 6 layers? I am not talking about the front baffle, as it has its own considirations. But all the sides, i think, can benefit from the extra mass. Also, it seems joining such thick panels together should be very easy, since the surface area of the joints is so large... Perhaps another benefit is that with joints being as large as the cavity itself, minimal bracing will be needed. Am I wrong? Will I just end up with a wooden mess? Thanks in advance.
SAND SANDWICH
Have you considered a sand filled sandwich box? Thick well braced inner cabinet, thinner skin outer cabinet with 1"-2" of sand in between. The heat generated by the vibrating sand absorbs the bass speaker's energy. Plus you can vacuum out the sand if you need to move. Even a dipole baffle can benefit from sand dampening.
The baffle around the speaker cannot be so thick that it creates reflections or limits air flow. Aluminum plate or steel reinforcements might be worth considering.
Midrange and tweeters generate little vibrational energy and packaging is not as difficult as long as they are isolated from the bass cabinet's vibration.
Have you considered a sand filled sandwich box? Thick well braced inner cabinet, thinner skin outer cabinet with 1"-2" of sand in between. The heat generated by the vibrating sand absorbs the bass speaker's energy. Plus you can vacuum out the sand if you need to move. Even a dipole baffle can benefit from sand dampening.
The baffle around the speaker cannot be so thick that it creates reflections or limits air flow. Aluminum plate or steel reinforcements might be worth considering.
Midrange and tweeters generate little vibrational energy and packaging is not as difficult as long as they are isolated from the bass cabinet's vibration.
sand filled is a very nice concpet but i found it diffcult to execte because it was diffult to keep the 2 wood layers independant. somewhere they had to meet.
so i decided to use curved sides to pre stress the wood and to form non parallel sides.
i also used ply - mdf combo.
see pic.
so i decided to use curved sides to pre stress the wood and to form non parallel sides.
i also used ply - mdf combo.
see pic.
Attachments
Amo,
I have a few suggestions. Don't just focus on one parameter - ie thickness, but consider the speaker box as a system. There are a number of ways to make an inert box, and if used cleverly together, they create a synergy. A 50 kg box is more impressive if it can match a 100 kg box.
A few ways to get an inert box:
- wall thickness
- bracing
- mass loaded damping
- constrained layer damping
- curved walls
1. wall thickness - I doubt anything thicker than two sheets of 18mm MDF is of any benefit if you use the other methods as well
2. bracing - construct a matrix so that all walls are solidly connected to the parallel wall and each bracing sheet is braced by the connection to other bracing sheets
3. Mass loaded damping - one member of the bass list recommends using a layer of non-curing oil based modelling clay applied to the internal walls of an enclosure. This method can be quite heavy.
4. Constrained layer damping - a box within a box - two separate layers which are joined by a flexible adhesive such as liquid nails. This is considerably more effective than doubling the thickness of the walls. This is a technique which has long been used in applications such as industrial noise control.
5. Curved walls - this makes it possible to make the walls thinner.
I think the ideal overkill design would include all of the above, perhaps with the omission of mass loading. Say with 36mm walls. I like the idea of overkill, but I dont' like the idea of excessively heavy. It means I need to get someone to help me lift thing, and so my DIY projects become less autonomous.
I have a few suggestions. Don't just focus on one parameter - ie thickness, but consider the speaker box as a system. There are a number of ways to make an inert box, and if used cleverly together, they create a synergy. A 50 kg box is more impressive if it can match a 100 kg box.
A few ways to get an inert box:
- wall thickness
- bracing
- mass loaded damping
- constrained layer damping
- curved walls
1. wall thickness - I doubt anything thicker than two sheets of 18mm MDF is of any benefit if you use the other methods as well
2. bracing - construct a matrix so that all walls are solidly connected to the parallel wall and each bracing sheet is braced by the connection to other bracing sheets
3. Mass loaded damping - one member of the bass list recommends using a layer of non-curing oil based modelling clay applied to the internal walls of an enclosure. This method can be quite heavy.
4. Constrained layer damping - a box within a box - two separate layers which are joined by a flexible adhesive such as liquid nails. This is considerably more effective than doubling the thickness of the walls. This is a technique which has long been used in applications such as industrial noise control.
5. Curved walls - this makes it possible to make the walls thinner.
I think the ideal overkill design would include all of the above, perhaps with the omission of mass loading. Say with 36mm walls. I like the idea of overkill, but I dont' like the idea of excessively heavy. It means I need to get someone to help me lift thing, and so my DIY projects become less autonomous.
It depends on the frequencies that the drivers in the box are reproducing I think. Bass boxes tend to require bracing and stiffness to keep the walls from flexing as a whole. If you can get the resonance of the panels higher than the freqs that you are reproducing, then they are less likely to resonate.
Higher frequencies require more damping of the box I believe.
Higher frequencies require more damping of the box I believe.
Thank you everyone!!!
My original intention was to machine "slices" of marine ply and press them together like TAD's model 1. I actually found a place that will do CNC routing for me. This design incorporates very thick curved walls and built in cross-bracing, and has no joints, other then horisontal... But I expect this method will set me back a good $600 USD for the unfinished cabinets (just wood and glue and CNC). So I decided to do a scaled down version of this project for now, using fewer drivers and going 3-way instead of 4-way. I also though I could use a more "standard" cab design. The reason I like the idea of such thick walls, is because they in fact eliminate the need for bracing and other more complicated tasks. I guess I will try it because it should be quick and I probably have all the wood already. I will see how it looks and sounds and if it is no good, I will go back to plan one and CNC thin layers. I also really like the idea of all that mass simply because there will be 2 ten inch drivers, and I feel like they need a sturdy base to play from...... Will see, and I will keep everone posted. Thanks again.
My original intention was to machine "slices" of marine ply and press them together like TAD's model 1. I actually found a place that will do CNC routing for me. This design incorporates very thick curved walls and built in cross-bracing, and has no joints, other then horisontal... But I expect this method will set me back a good $600 USD for the unfinished cabinets (just wood and glue and CNC). So I decided to do a scaled down version of this project for now, using fewer drivers and going 3-way instead of 4-way. I also though I could use a more "standard" cab design. The reason I like the idea of such thick walls, is because they in fact eliminate the need for bracing and other more complicated tasks. I guess I will try it because it should be quick and I probably have all the wood already. I will see how it looks and sounds and if it is no good, I will go back to plan one and CNC thin layers. I also really like the idea of all that mass simply because there will be 2 ten inch drivers, and I feel like they need a sturdy base to play from...... Will see, and I will keep everone posted. Thanks again.
Speaking to an Excess
In 1989 I designed and built a line of loudspeakers that went nowhere . They sounded great but could't be marketed. Here's why.
They consisted of a box withan extra thick baffle using you idea of constairned layer dampening. The interior of the box was made using a concrete pier tube that was cut and fitted in grooves on the baffle , the bottom and the top. We shaped it so that most interior standing waves would kill themselves in two to three reflections. Just observe a drop of water in an eliptical bowl and you will get the drift of the idea. ( just guess where I came up with that brain storm ) The voids in the interior were filled with dry sand. The whole mess was filled and tapped with rubber mallets to create a really ( THe deadest box I have ever seen bar none. Wilson audio is the only player in that field, and I believe that we beat them for a hell of alot less money )dead. It makes for a heavy cabinet. You have to calculate for an elipse in the volume end of things bit it is one of the best systems I have ever seen or heard.
Food for thought.
Mark
In 1989 I designed and built a line of loudspeakers that went nowhere . They sounded great but could't be marketed. Here's why.
They consisted of a box withan extra thick baffle using you idea of constairned layer dampening. The interior of the box was made using a concrete pier tube that was cut and fitted in grooves on the baffle , the bottom and the top. We shaped it so that most interior standing waves would kill themselves in two to three reflections. Just observe a drop of water in an eliptical bowl and you will get the drift of the idea. ( just guess where I came up with that brain storm ) The voids in the interior were filled with dry sand. The whole mess was filled and tapped with rubber mallets to create a really ( THe deadest box I have ever seen bar none. Wilson audio is the only player in that field, and I believe that we beat them for a hell of alot less money )dead. It makes for a heavy cabinet. You have to calculate for an elipse in the volume end of things bit it is one of the best systems I have ever seen or heard.
Food for thought.
Mark
why hasn't someone posted the obvious ?
5" thick walls will add add at least 8" to the external dimensions
of a loudspeaker and will be the bulk of the volume for most speakers.
e.g. a 3cuFt enclosure 1" and 5" cubic.
1" = a 19" cube. Material = 28% of total volume.
5" = a 27" ! cube. Material = 74% of total volume.
Any other shape is a lot worse.
Its an extremely dumb idea, just to avoid bracing.
sreten.
5" thick walls will add add at least 8" to the external dimensions
of a loudspeaker and will be the bulk of the volume for most speakers.
e.g. a 3cuFt enclosure 1" and 5" cubic.
1" = a 19" cube. Material = 28% of total volume.
5" = a 27" ! cube. Material = 74% of total volume.
Any other shape is a lot worse.
Its an extremely dumb idea, just to avoid bracing.
sreten.
Nothing is obvious
Screten the box idea that I mentioned adds little to the exterior dimmensions. It adds mass. A heavy mass is what some people believe makes for a clean loudspeaker. In otherwords you are hearing what your drivers are up to. Not comments your box will like to interject. I have no idea what you have done or listened to but I stand by the post I made earlier.
Engineering is a balance of tradeoffs and building concrete boxes or battleship thick walls is easy. To engineer an enclosure that encompasses most of their attributes and yet is more portable is the sign of good engineering.
Mark
Screten the box idea that I mentioned adds little to the exterior dimmensions. It adds mass. A heavy mass is what some people believe makes for a clean loudspeaker. In otherwords you are hearing what your drivers are up to. Not comments your box will like to interject. I have no idea what you have done or listened to but I stand by the post I made earlier.
Engineering is a balance of tradeoffs and building concrete boxes or battleship thick walls is easy. To engineer an enclosure that encompasses most of their attributes and yet is more portable is the sign of good engineering.
Mark
sreten said:why hasn't someone posted the obvious ?
5" thick walls will add add at least 8" to the external dimensions
of a loudspeaker and will be the bulk of the volume for most speakers.
e.g. a 3cuFt enclosure 1" and 5" cubic.
1" = a 19" cube. Material = 28% of total volume.
5" = a 27" ! cube. Material = 74% of total volume.
Any other shape is a lot worse.
Its an extremely dumb idea, just to avoid bracing.
I do not see this as a problem: For the top of the enclosure such large dimentions are welcome, as they will permit very gentle sloping backwards, for baffle difraction and aethetics. Think Thiel...
http://www.thielaudio.com/THIEL_Web/Web_Page_Images/CS7_2_page_images/7_2prnb.jpg
For the bass: Well, if I use 10 inch drivers, then my baffle will be at least 21 or so inches wide. This goes againsed current trends of thin front baffle and huge sides, but I think I know how to make it work. Further, something like Zebra wood veneer can be used to excentuate the already wide horizontal span. True, these things will be huge, but then again, I do not have to appeal to a mass market, just me... Just to clear things up, the sole intention of this is not simply to avoid bracing, but to add mass and more mass.
In all reality however, 5 inches is just a hypothetical number chosen for discussion. It may work out just as well with, say, 3 inches and some Black Hole...
I tend to agree with Sreten in that it isn't a good idea to make excessively thick walls to avoid the difficulty of bracing. As was mentioned by Mark, it is about balance of tradeoffs.
Comercial speakers often don't include fancy construction techniques due to the extra cost or manufacture, the ones that do are usually quite expensive. Some may disagree, but I think B&w tend to make boxes that are fairly complex in their construction relative to their cost. However, when it comes to DIY it makes sense to "do more with less." We have the luxury to build boxes that are more elaborate, to make TL's and horns, or very inert boxes with elaborate internal structures to get rigidity. But to reinforce an earlier point, if you make your box too heavy, it becomes very difficult to work with and to move around, something that a DIYer will want to do a lot.
I have heard of people using the CNC router idea, but if you do that at high cost you start to kill the advantage of DIY. Why do this when you can get the same result in other ways.
Here are some good examples of making curved timber speakers by various means:
curved speaker construction
For making a curved box, the layered lamination method looks very promising. I had previously planned to build a curved subwoofer with this approach, combining matrix bracing, constrained layer damping. I would have used two curved skins bent around horizontal bracing pieces carefully cut by router at vertical increments of say 200mm. The skins would be connected to each other by liquid nails. Internal skin would be 18mm, external skin 12mm. The skins would be made of 3mm sheets laminated together laid one at a time. This allows them to bend and form a continuous curve forming the sides and rear of the box, joining at the front corners of the baffle.
I'm not sure if these images will work:
If that doesn't work, they can be seen here on my site, also with a brief description:
Curved subwoofer construction on my website
Please note: this design was not fully resolved, and for the time being has been shelved.
regards,
Paul Spencer
Comercial speakers often don't include fancy construction techniques due to the extra cost or manufacture, the ones that do are usually quite expensive. Some may disagree, but I think B&w tend to make boxes that are fairly complex in their construction relative to their cost. However, when it comes to DIY it makes sense to "do more with less." We have the luxury to build boxes that are more elaborate, to make TL's and horns, or very inert boxes with elaborate internal structures to get rigidity. But to reinforce an earlier point, if you make your box too heavy, it becomes very difficult to work with and to move around, something that a DIYer will want to do a lot.
I have heard of people using the CNC router idea, but if you do that at high cost you start to kill the advantage of DIY. Why do this when you can get the same result in other ways.
Here are some good examples of making curved timber speakers by various means:
curved speaker construction
For making a curved box, the layered lamination method looks very promising. I had previously planned to build a curved subwoofer with this approach, combining matrix bracing, constrained layer damping. I would have used two curved skins bent around horizontal bracing pieces carefully cut by router at vertical increments of say 200mm. The skins would be connected to each other by liquid nails. Internal skin would be 18mm, external skin 12mm. The skins would be made of 3mm sheets laminated together laid one at a time. This allows them to bend and form a continuous curve forming the sides and rear of the box, joining at the front corners of the baffle.
I'm not sure if these images will work:
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
If that doesn't work, they can be seen here on my site, also with a brief description:
Curved subwoofer construction on my website
Please note: this design was not fully resolved, and for the time being has been shelved.
regards,
Paul Spencer
Amo,
In my experience be wary of using MDF for it gives a particular sonic signature regardless of how thick or 'dead' the cabinet is.
This also applies to chipboard or plywood which give another two sonic signatures regardless of the thickness or 'deadness'.
Thick/multi-layered chipboard is good, and better to my ear is thick/multi-layered plywood every time.
A 'dead' cabinet is still not so dead as you may be thinking.
Eric.
In my experience be wary of using MDF for it gives a particular sonic signature regardless of how thick or 'dead' the cabinet is.
This also applies to chipboard or plywood which give another two sonic signatures regardless of the thickness or 'deadness'.
Thick/multi-layered chipboard is good, and better to my ear is thick/multi-layered plywood every time.
A 'dead' cabinet is still not so dead as you may be thinking.
Eric.
Hang on, just what are we trying to achieve here? - A clean sounding speaker, that's made even cleaner by not radiating any sound from the external panels?
It was mentioned that vibrations from bass frequencies are harder to control than midrange and high frequencies. I think that's not true, but the reason it may appear so is because box vibrations from bass frequencies are more noticeable by touching the box with your hand, and mid/high frequencies can't be felt. Notice that a tweeter may be blaring at full volume and you still won't feel the cone's vibrations if you gently touch it, so how do you know that there aren't similar vibrations all over the box and completely smearing the sound?
Usually with bass frequencies the box vibrates as a whole (unless it is extremely floppy for some strange reason) and there is no problem. The motion is pistonic: the cone moves one way, and the box has to move the other way (fact of life), unless there are 2 identical speakers working so that the forces cancel out. It could also be an idea to put the speaker on rollers/bearings or something like that to make sure those low frequencies aren't also transmitted to the floor. If it weighs a quarter of a tonne then don't bother.
"Deadening" a box isn't anything to do with bass frequencies, but deadening the midrange and high frequencies. Even with a box with panels that are ridiculously thick, say 10", high frequencies are still likely to vibrate throughout the mass of the box in two ways:
Transverse Vibrations: Small ripples travelling all around the panels of the box. Due to things like the box's size, mass, stiffness etc, there will be standing waves from these vibrations at various frequencies.
Longitudinal Vibrations: The box material gets compacted and expands, and likewise standing waves build up all over the place.
Especially considering that most of the box is dedicated to the woofer, deadening techniques that you should focus on are ones that reduce the size of the high frequency vibrations in the first place:
-Putting the tweeter and midrange in separate enclosures. Even if those enclosures aren't that good, at least the vibrations will be confined to a smaller area.
-Detaching the tweeter/mid from the box with gaskets, and NOT solidly screwing them down metal-to-wood.
-Using an extra stiff box material for the woofer, and a softer material for the midrange. A great material for the woofer will usually be crap for the midrange and vice versa.
-The loudspeaker equivalent for double-glazing (mentioned earlier), so that those box vibrations are isolated inside of another box.
Doing all of the above may well be overkill, but it's probably worth it.
CM
It was mentioned that vibrations from bass frequencies are harder to control than midrange and high frequencies. I think that's not true, but the reason it may appear so is because box vibrations from bass frequencies are more noticeable by touching the box with your hand, and mid/high frequencies can't be felt. Notice that a tweeter may be blaring at full volume and you still won't feel the cone's vibrations if you gently touch it, so how do you know that there aren't similar vibrations all over the box and completely smearing the sound?
Usually with bass frequencies the box vibrates as a whole (unless it is extremely floppy for some strange reason) and there is no problem. The motion is pistonic: the cone moves one way, and the box has to move the other way (fact of life), unless there are 2 identical speakers working so that the forces cancel out. It could also be an idea to put the speaker on rollers/bearings or something like that to make sure those low frequencies aren't also transmitted to the floor. If it weighs a quarter of a tonne then don't bother.
"Deadening" a box isn't anything to do with bass frequencies, but deadening the midrange and high frequencies. Even with a box with panels that are ridiculously thick, say 10", high frequencies are still likely to vibrate throughout the mass of the box in two ways:
Transverse Vibrations: Small ripples travelling all around the panels of the box. Due to things like the box's size, mass, stiffness etc, there will be standing waves from these vibrations at various frequencies.
Longitudinal Vibrations: The box material gets compacted and expands, and likewise standing waves build up all over the place.
Especially considering that most of the box is dedicated to the woofer, deadening techniques that you should focus on are ones that reduce the size of the high frequency vibrations in the first place:
-Putting the tweeter and midrange in separate enclosures. Even if those enclosures aren't that good, at least the vibrations will be confined to a smaller area.
-Detaching the tweeter/mid from the box with gaskets, and NOT solidly screwing them down metal-to-wood.
-Using an extra stiff box material for the woofer, and a softer material for the midrange. A great material for the woofer will usually be crap for the midrange and vice versa.
-The loudspeaker equivalent for double-glazing (mentioned earlier), so that those box vibrations are isolated inside of another box.
Doing all of the above may well be overkill, but it's probably worth it.
CM
Regarding box vibrations and tweeters, I think one thing that should not be ignored is that how much does a tweeter move??? Not very much is all I can say. Now how much will the other drivers shaking the box actively "move" the tweeter, probably not a huge amount but then again the tweeter isnt moving much either.
My speakers are three way, all the drivers in separate boxes, with translam on the mid and tweet enclosures. Tweeter more for an aesthetic match to the mid. Either way if I touch the tweeter enclosure when the music is playing (this doesnt have to be loud) you can feel it being shaken quite a lot by the other drivers movement. I put some large cell bublewrap between the tweet and midrange cabinet and they ALL (this is no joke I was surprised at how much this worked) dissapeared. Listening was also better, vocals sounded smoother and more airy, everything sounded more relaxed or just flowed from the speaker more easily. It sounded kinda like you would expect it to, just improving the high frequencies. Which subjectively improved the rest of the sound.
I agree that separate boxes are the way to go, with different construction techniques for each box.
I like the gasket idea and not screwing things in tightly. I have done something similar with blutak on my mids cabinet but nothing on the tweets as I didnt think the tweeter would cause anyting to vibrate. Hmm I guess I was wrong, now where did all that spare rubber go.....
My speakers are three way, all the drivers in separate boxes, with translam on the mid and tweet enclosures. Tweeter more for an aesthetic match to the mid. Either way if I touch the tweeter enclosure when the music is playing (this doesnt have to be loud) you can feel it being shaken quite a lot by the other drivers movement. I put some large cell bublewrap between the tweet and midrange cabinet and they ALL (this is no joke I was surprised at how much this worked) dissapeared. Listening was also better, vocals sounded smoother and more airy, everything sounded more relaxed or just flowed from the speaker more easily. It sounded kinda like you would expect it to, just improving the high frequencies. Which subjectively improved the rest of the sound.
I agree that separate boxes are the way to go, with different construction techniques for each box.
I like the gasket idea and not screwing things in tightly. I have done something similar with blutak on my mids cabinet but nothing on the tweets as I didnt think the tweeter would cause anyting to vibrate. Hmm I guess I was wrong, now where did all that spare rubber go.....
amo, doing 5-inch walls is OVERKILL! In my opinion, anything thicker than 1.75-inch should be reconsidered, unless of course, you're making a subwoofer box with two or more 18-inchers. I think the best solution is to use a composite of materials (layers of stuff sandwiched together) to deaden the cabinet. MDF is usually the material of choice for most DIYers and commercial manufacturers, but when two relatively thin panels are combined with, say, a layer of foam, you'll achieve better results with less weight. Of course, don't expect to get away without bracing this kind of cabinet.
Good luck.
Good luck.
I saw and listened to the TAD speaker at the audio show here in S.F. It was built out of stacked layers as described. There were metal rods vertically clamping it -presumably it's glued also
All thread would wwork well for this.
It's a method of making cabinets I've considered a lot. $600 for materials and cutting might be a pretty good deal. 5" walls are going to be a problem because the may block sound from the back of the woofer a bit. The walls of the TAD were about 2.5" to 3" I think. Some of the layers had Y shaped bracing between the walls. From my inspection there were say 3 layers with braces, the a bunch without, then more braces.
After seeing and hearing the speaker, I'd say GO FOR IT!! but walls like TAD uses are probably enough. The only problem with thicker walls are that the speaks will be HUGE and material cost.
The TAD "head" with the combo 3" mid and 1" dome (both Berillium) is mounted on firm but flexible material to decouple it from the bass box exactly as mentioned above by 5th Element.
The speakers were some of the more impressive I heard- especially for large scale stuff like symphonic music
All thread would wwork well for this.
It's a method of making cabinets I've considered a lot. $600 for materials and cutting might be a pretty good deal. 5" walls are going to be a problem because the may block sound from the back of the woofer a bit. The walls of the TAD were about 2.5" to 3" I think. Some of the layers had Y shaped bracing between the walls. From my inspection there were say 3 layers with braces, the a bunch without, then more braces.
After seeing and hearing the speaker, I'd say GO FOR IT!! but walls like TAD uses are probably enough. The only problem with thicker walls are that the speaks will be HUGE and material cost.
The TAD "head" with the combo 3" mid and 1" dome (both Berillium) is mounted on firm but flexible material to decouple it from the bass box exactly as mentioned above by 5th Element.
The speakers were some of the more impressive I heard- especially for large scale stuff like symphonic music
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.