He has said thicker is better but this affects cure times and can take ages to cure. Apparently it stays soft. I am gonna try some experiments but he didn't suggest spacers or screws, so I guess a think layer, a little pressure to get it to 1mm, then a few weeks to dry.
I think he keeps it simple, thick is better to a point, soft tacky flexible layer between two hard layers whether MDF or ply.
I did have a thought about my baffle though. Sawing it in half would allow the woofer section to be epoxied in place then the top half could use green glue in cld and in the sawn gap to hold it up. Should prevent any sag.
The melamine glue is almost as expensive as green glue for me, so I am still undecided which to use.
I think he keeps it simple, thick is better to a point, soft tacky flexible layer between two hard layers whether MDF or ply.
I did have a thought about my baffle though. Sawing it in half would allow the woofer section to be epoxied in place then the top half could use green glue in cld and in the sawn gap to hold it up. Should prevent any sag.
The melamine glue is almost as expensive as green glue for me, so I am still undecided which to use.
Here's an informative link for sorbothane, which has been applied and tested by Bushmeister in this thread. Sorbothane is a thermoset, polyether-based polyurethane.
damping factor = 0.5 at 50 Hz
damping factor = tan delta = 1/Q = 2 x damping ratio
Sorbothane Technical Data
Shock and vibration solutions | Sorbothane
damping factor = 0.5 at 50 Hz
damping factor = tan delta = 1/Q = 2 x damping ratio
Sorbothane Technical Data
Shock and vibration solutions | Sorbothane
Ouch! That's expensive. Bushmeister used the sorbothane for decouple mounting the tweeter and midrange drivers. So he only needed 1 to 2 ft^2. Price is still good information.
OK guys - I have been experimenting with PVA, standard 2 part epoxy and titebond melamine glues for CLD in my new baffle.
The epoxy actually seems to give as good a result as the titebond (knock test) - I suspect this may be due to the properties and thickness of the constraining layers I am using - 18mm MDF to 30mm solid oak with 1mm layer of glue.
Either way, the epoxy seems to provide reasonable damping and I would guess it would make the baffle stiffer than the titebond/green glue. I will be decoupling the mid and tweeter with sorbothane gaskets, but mounting the woofer rigidly to the new baffle. The baffle will be almost 5cm thick and I want to keep it as stiff as possible to keep the resonances out of the passband of the woofer.
So......I am thinking I will probably use the epoxy in a 1mm layer. Nearly finished routing the new baffles, so I will report back later.
The epoxy actually seems to give as good a result as the titebond (knock test) - I suspect this may be due to the properties and thickness of the constraining layers I am using - 18mm MDF to 30mm solid oak with 1mm layer of glue.
Either way, the epoxy seems to provide reasonable damping and I would guess it would make the baffle stiffer than the titebond/green glue. I will be decoupling the mid and tweeter with sorbothane gaskets, but mounting the woofer rigidly to the new baffle. The baffle will be almost 5cm thick and I want to keep it as stiff as possible to keep the resonances out of the passband of the woofer.
So......I am thinking I will probably use the epoxy in a 1mm layer. Nearly finished routing the new baffles, so I will report back later.
http://www.google.co.uk/url?sa=t&rc...HvaKexogF_zoCXA&bvm=bv.85970519,d.ZWU&cad=rja
This is an interesting paper were they use CLD with epoxy in a steel gun barrel to damp the barrel resonances caused by hitting it with a hammer.
The bit I find interesting is the the barrel's main modes of vibration after being hit are '500 Hz, 1000 Hz, 2000 Hz and 3000 Hz' - which are much more in keeping with the vibration modes of a stiff speaker cabinet than the modes you would see in a plasterboard wall that uses green glue as a CLD layer.
So perhaps, given the much higher modes we are trying to optimally dampen in speaker boxes rather than large dividing walls of buildings, epoxy isn't that crazy an idea.
This is an interesting paper were they use CLD with epoxy in a steel gun barrel to damp the barrel resonances caused by hitting it with a hammer.
The bit I find interesting is the the barrel's main modes of vibration after being hit are '500 Hz, 1000 Hz, 2000 Hz and 3000 Hz' - which are much more in keeping with the vibration modes of a stiff speaker cabinet than the modes you would see in a plasterboard wall that uses green glue as a CLD layer.
So perhaps, given the much higher modes we are trying to optimally dampen in speaker boxes rather than large dividing walls of buildings, epoxy isn't that crazy an idea.
Bushmeister,
Could you help out Gibbi on Post #306 with his question of what durometer of sorbothane to use for decoupling? You've already used sorbothane.
I've spend a lot of time trying to find good clear guidelines on the application of constrained layer damping. I've made some progress, but maybe I just have not found the best source yet. I' m still digging. Here goes so far.
1) CLD is a tuned system. The damping versus frequency curve is basically a single hump over a span of frequencies, it is not flat. The effective part of the damping curve could be centered over the band of frequencies desire, or it could be centered over too low or too high frequency ranges. The shear modulus of the damping material needs to match the stiffness of the base and constraining layers. Stiffer (top) constraining layer is better.
2) The effective part of the damping curve can be shifted to higher frequencies by:
a) Changing damping material to one have greater shear modulus (shear stiffness)
b) Thinner damping layer thickness
c) Cooler temperature
3) Damping materials for sheetrock walls have to deal with natural resonances in the 5 to 20 Hz range, well below the natural frequencies of loudspeaker enclosure walls, of 160 Hz or greater. So, are the shear modulus properties of constraining damping adhesives/compounds such as Green Glue optimum for loudspeaker enclosure CLD applications?
Could you help out Gibbi on Post #306 with his question of what durometer of sorbothane to use for decoupling? You've already used sorbothane.
I've spend a lot of time trying to find good clear guidelines on the application of constrained layer damping. I've made some progress, but maybe I just have not found the best source yet. I' m still digging. Here goes so far.
1) CLD is a tuned system. The damping versus frequency curve is basically a single hump over a span of frequencies, it is not flat. The effective part of the damping curve could be centered over the band of frequencies desire, or it could be centered over too low or too high frequency ranges. The shear modulus of the damping material needs to match the stiffness of the base and constraining layers. Stiffer (top) constraining layer is better.
2) The effective part of the damping curve can be shifted to higher frequencies by:
a) Changing damping material to one have greater shear modulus (shear stiffness)
b) Thinner damping layer thickness
c) Cooler temperature
3) Damping materials for sheetrock walls have to deal with natural resonances in the 5 to 20 Hz range, well below the natural frequencies of loudspeaker enclosure walls, of 160 Hz or greater. So, are the shear modulus properties of constraining damping adhesives/compounds such as Green Glue optimum for loudspeaker enclosure CLD applications?
Regarding which sorbothane - I bough 40 Duro in a 3mm sheet. I worked out optimum torque for the screws mainly by trial and error and measurements - see my earlier posted measurements - it made an enormous difference.
I think CLD is very complicated, and the problem is decoupling is a very effective technique too (sand for instance - is green glue the 'new sand' in speaker construction?) so if green glue or neoprene are mainly working via decoupling, rather than true CLD it will still work well and demonstrate improved measurements.
I suspect that the CLD designed products such as green glue work more as decoupling agents rather than true CLD layers - as they are engineered for greater wall movements and lower Hz resonance, but how easy is this to prove when both mechanisms will have similarly effective results at reducing resonance?
I have found the epoxy to work well, but is this just a combination of the densities, stiffness and damping factors of two dissimilar woods tightly bonded together producing better damping and stiffness in synergy (AKA aluminium/MDF/aluminium laminates you cited earlier providing high damping)?
I suppose a lot of this comes down to trial and error with the materials you are specifically planning on building with.
For my needs I want a rigid, high mass, well damped baffle which is mainly 'engineered' for the woofer as the tweeter and mid are decoupled, so epoxy makes more sense. Other designs may change this.
I think CLD is very complicated, and the problem is decoupling is a very effective technique too (sand for instance - is green glue the 'new sand' in speaker construction?) so if green glue or neoprene are mainly working via decoupling, rather than true CLD it will still work well and demonstrate improved measurements.
I suspect that the CLD designed products such as green glue work more as decoupling agents rather than true CLD layers - as they are engineered for greater wall movements and lower Hz resonance, but how easy is this to prove when both mechanisms will have similarly effective results at reducing resonance?
I have found the epoxy to work well, but is this just a combination of the densities, stiffness and damping factors of two dissimilar woods tightly bonded together producing better damping and stiffness in synergy (AKA aluminium/MDF/aluminium laminates you cited earlier providing high damping)?
I suppose a lot of this comes down to trial and error with the materials you are specifically planning on building with.
For my needs I want a rigid, high mass, well damped baffle which is mainly 'engineered' for the woofer as the tweeter and mid are decoupled, so epoxy makes more sense. Other designs may change this.
What will dampen a heavy metal tube and a flat wood panel would be two different things. Another question that comes into play is hardness and thickness of epoxy. While I am sure there is a significant difference than without, the real goal is to find what is optimal. Their best result was a tapered barrel encased in a straight outer tube inwhich this expanding void was filled. All in all far thicker than 1mm.
Friend has a Barrett sniper rifle with dampeners. What we found most critical as far as accuracy goes was the ammo used. Nothing off the shelf, all done by hand, calibrated. The dampeners play a distance 3rd. Once dialed in can place 5 rounds within a 3mm circle @100yds. At 600yds can place them within a 5" circle.
Durometer of 30-50 for sorbothane appears best. The softer rubbers are in the Shore 1 lowest range and work for these higher frequencies as illustrated in the barrel dampening paper. Shore 00 hardest is less than these. Butyl rubber gasket material is lower yet, but as mentioned in the sorbothane documentation, creepage may be a problem with the lower durometer range materials.
Friend has a Barrett sniper rifle with dampeners. What we found most critical as far as accuracy goes was the ammo used. Nothing off the shelf, all done by hand, calibrated. The dampeners play a distance 3rd. Once dialed in can place 5 rounds within a 3mm circle @100yds. At 600yds can place them within a 5" circle.
Durometer of 30-50 for sorbothane appears best. The softer rubbers are in the Shore 1 lowest range and work for these higher frequencies as illustrated in the barrel dampening paper. Shore 00 hardest is less than these. Butyl rubber gasket material is lower yet, but as mentioned in the sorbothane documentation, creepage may be a problem with the lower durometer range materials.
You are right of course! Except I think the 'optimum' CLD layer is probably going to be totally dependant on the thickness and type of materials used in the construction - and therefore probably only found with experimentation.
With your gun analogy, I guess the ammunition is like the crossover - without a proper crossover, the rest goes to pot!
But then we are all chasing the last final improvements which is what this thread is all about!
With your gun analogy, I guess the ammunition is like the crossover - without a proper crossover, the rest goes to pot!
But then we are all chasing the last final improvements which is what this thread is all about!
Then an accelerometer or magnetic pickup would be in order.
These days I tend to not make rectilinear enclosures or flat baffles. By minimizing radiating panel size through isolation you can achieve better results. For example the mtm mltl project working on is quite wide @24". The center section housing the mtm is a box in a box type design with an additional external floating front baffle. This enclosue is not wide @7.5" but is dampened and isolated from the flanking mltl subs that make up the arced wings to its left and right. Leaving the high mass front baffle to only the mtm section lowers overall weight and any sound that gets through the sides/top and bottom are forced to penetrate the mltl's enclosures effectively using them to attenuate what remains. The mltls are cld sonotubes seriously reducing mass and my back says thank you. Use of force cancellation greatly improves transients and reduces sub enclosure coloration. When the mltl was first tested it suprised me that it had as much vertical movement as it did. Axially mounted there was simply not enough enclosure mass to prevent. Seeing that mass reduction was a concern opting for force cancellation was better than adding another 100lbs of mass per enclosure.
These days I tend to not make rectilinear enclosures or flat baffles. By minimizing radiating panel size through isolation you can achieve better results. For example the mtm mltl project working on is quite wide @24". The center section housing the mtm is a box in a box type design with an additional external floating front baffle. This enclosue is not wide @7.5" but is dampened and isolated from the flanking mltl subs that make up the arced wings to its left and right. Leaving the high mass front baffle to only the mtm section lowers overall weight and any sound that gets through the sides/top and bottom are forced to penetrate the mltl's enclosures effectively using them to attenuate what remains. The mltls are cld sonotubes seriously reducing mass and my back says thank you. Use of force cancellation greatly improves transients and reduces sub enclosure coloration. When the mltl was first tested it suprised me that it had as much vertical movement as it did. Axially mounted there was simply not enough enclosure mass to prevent. Seeing that mass reduction was a concern opting for force cancellation was better than adding another 100lbs of mass per enclosure.
Bushmeister, I'm just throwing the following information out there, as I'm not aware of its actual performance in our applications. The 650-8 and 655-8 G/Flex epoxies from West System have 1/3rd the stiffness of standard epoxies. G/Flex has a 150,000 psi modulus of elasticity. West System states its performance fits between and epoxy and polyurethane adhesives, which is a wide gap, but may apply to some panel vibration damping situations. From the West System website:
"A toughened, versatile, liquid epoxy for permanent waterproof bonding of fiberglass, ceramics, metals, plastics, damp and difficult-to-bond woods. With a modulus of elasticity of 150,000 PSI, it is a bit more flexible than standard epoxies and polyester, but much stiffer than adhesive sealants. This gives G/flex 650 the ability to make structural bonds that can absorb the stress of expansion, contraction, shock and vibration...."
WEST SYSTEM | Specialty Epoxies - G/flex
WEST SYSTEM | Plastic Boat Repair - Understanding Flexible Properties
WEST SYSTEM | Epoxy Resins and Hardeners - Physical Properties
http://www.westsystem.com/ss/assets/Product-Data-PDFs/Gflex Technical Data Sheet 0313.pdf
Last edited:
Related link for epoxies and noise damping
http://www.noisetek.fi/en/Noisetek_ELASTE_Elastic_Epoxy.pdf
http://www.noisetek.fi/en/Noisetek_ELASTE_Elastic_Epoxy.pdf
Wow there is a lot to take in here.
I am curious about BBC style cabinets, would this work for a large bass capable full range speaker, I am thinking of a Monacor SPX 200WP 8" in a sealed and stuffed 30 litre box qtc 1.2 , F3 45 Hz or would this be too much for a thin walled well damped enclosure?
Simon
I am curious about BBC style cabinets, would this work for a large bass capable full range speaker, I am thinking of a Monacor SPX 200WP 8" in a sealed and stuffed 30 litre box qtc 1.2 , F3 45 Hz or would this be too much for a thin walled well damped enclosure?
Simon
Hi, sorry if I jump in with a maybe weird question.
I understand that the most difficult to control are the low frequencies because they have a lot of energy. And a box for a subwoofer is one of the most challenging design.
Is there a frequency above which these vibrations are not very critical because little energy is involved ? like 200Hz for instance ?
The idea would be to use two boxes mechanically decoupled.
Thanks a lot, gino
I understand that the most difficult to control are the low frequencies because they have a lot of energy. And a box for a subwoofer is one of the most challenging design.
Is there a frequency above which these vibrations are not very critical because little energy is involved ? like 200Hz for instance ?
The idea would be to use two boxes mechanically decoupled.
Thanks a lot, gino
Hi ginetto61,
a "pure" subwoofer cabinet can be made
- stiff and
- compact (small)
in this way you can push the resonances of the cabinet's walls e.g. above 220 Hz and thus out of the used bandwidth of a subwoofer ...
It is possible to do this even without extensive bracing, just by keeping the structure stiff and not too far from either a cube or a column shape.
A stiff - significantly thicker - baffle plate (where the driver tries to "excite and bend" the cabinet directly) will also help to rise "bell modes" in frequency. All those measures are simpler than bracing, and will help more than bracing a cabinet made of "thin" walls (e.g. <18mm MDF).
If larger internal volumes are needed - e.g. more then 25 liters - it is possible to have more of the small cabinets and even more drivers.
So more small enclosures "mechanically decoupled" may also be "at different positions in the room", which can be of advantage in exciting the room in a more balanced way.
As soon as a subwoofer cabinet exceeds a certain size, it will turn into a "problem" and a "battle of materials" will start if wall vibrations are to be tamed ...
The lowest modes of the subwoofers cabinets shown here are about 220Hz:
http://www.dipol-audio.de/projekt-schwingungen-an-lautsprechergehaeusen.html
"Bild 4+5" (Picture 4 and 5 show a "Chladni picture" (* )of that bell mode using fine sand (and lots of input power, voice coils had to be cool down while forming this picture in several steps...)
But such a "stiffness based" cabinet - little damping or "mass hampering" in the walls - is only useful in pure subwoofer IMHO. Used in a wideband manner (up to upper bass ?), one need's to make a "completely different" kind of cabinet, if "low sound radiation" from the cabinet is the goal.
_____________
(*)
https://de.wikipedia.org/wiki/Chladnische_Klangfigur
https://de.wikipedia.org/wiki/Ernst_Florens_Friedrich_Chladni
a "pure" subwoofer cabinet can be made
- stiff and
- compact (small)
in this way you can push the resonances of the cabinet's walls e.g. above 220 Hz and thus out of the used bandwidth of a subwoofer ...
It is possible to do this even without extensive bracing, just by keeping the structure stiff and not too far from either a cube or a column shape.
A stiff - significantly thicker - baffle plate (where the driver tries to "excite and bend" the cabinet directly) will also help to rise "bell modes" in frequency. All those measures are simpler than bracing, and will help more than bracing a cabinet made of "thin" walls (e.g. <18mm MDF).
If larger internal volumes are needed - e.g. more then 25 liters - it is possible to have more of the small cabinets and even more drivers.
So more small enclosures "mechanically decoupled" may also be "at different positions in the room", which can be of advantage in exciting the room in a more balanced way.
As soon as a subwoofer cabinet exceeds a certain size, it will turn into a "problem" and a "battle of materials" will start if wall vibrations are to be tamed ...
The lowest modes of the subwoofers cabinets shown here are about 220Hz:
http://www.dipol-audio.de/projekt-schwingungen-an-lautsprechergehaeusen.html
"Bild 4+5" (Picture 4 and 5 show a "Chladni picture" (* )of that bell mode using fine sand (and lots of input power, voice coils had to be cool down while forming this picture in several steps...)
But such a "stiffness based" cabinet - little damping or "mass hampering" in the walls - is only useful in pure subwoofer IMHO. Used in a wideband manner (up to upper bass ?), one need's to make a "completely different" kind of cabinet, if "low sound radiation" from the cabinet is the goal.
_____________
(*)
https://de.wikipedia.org/wiki/Chladnische_Klangfigur
https://de.wikipedia.org/wiki/Ernst_Florens_Friedrich_Chladni
Last edited:
Hi ! thanks a lot for the very helpful reply.
I will study it more in depth in the weekend. I am at work now.
Then it seems to me that a speaker + subwoofer solution is the way to go instead of a full range speaker with all drivers inside, when a real full range response is wanted of course.
Problem is to establish where to high cut the subwoofer.
If the sub is cut let's say at 120Hz the woofer in the speaker could still generate a lot of nasty vibes when pushed hard.
But not like a big woofer oscillating at 60Hz and high dB. That would be like an earthquake.
I am now quite sold on the idea of separating the very low bass from the rest of the range.
Thanks again, gino
P.S. an example here
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.