THE PROBLEM:
how to mount drivers in cabinets with CLD structure?
if you put the bolt run through all three (or more) layers, then you connect them in a rigid way and the CLD effect if gone?
THE SOLLUTION?
you tell me.. I think the speaker drivers should be mounted only to one (outter) layer.
how to mount drivers in cabinets with CLD structure?
An externally hosted image should be here but it was not working when we last tested it.
if you put the bolt run through all three (or more) layers, then you connect them in a rigid way and the CLD effect if gone?
THE SOLLUTION?
you tell me.. I think the speaker drivers should be mounted only to one (outter) layer.
I too have been thinking about this recently and came to the conclusion that the driver should be attached to the INSIDE layer only. Basically as a result of reading this: http://www.silcom.com/~aludwig/Loudspeaker_construction.html
Which I sumarize like this:
1. The main purpose of the cabinet is to dispose of all backwards sound from the speaker leaving only the clean front firing waves.
2. The walls of a normal box will sing and it is hard to stop it with stiff panels, bracing etc.
3. Building a box within a box gets you along way toward eliminating sound radiation from the sides and back of the box.
In this case CLD is more or less a "box within a box". The driver should therefore be fixed to the inner surface. The CLD construction then reduces vibrations reaching the outside and hence unwanted sound radiation from the front panel.
That article prompted me to a simple experiment: Twang a 440Hz tuning fork, the only one I happen to have, and touch it to the panels of a speaker box. Sure enough all speakers I have tried ring like bells!! Seems possible that more sound could come from the sides of the box than the cone itself.
I would be interested in the results of doing this with your CLD boxes. Perhaps using a little speaker fixed inside the box as a stimulous and see how much ringing gets out.
Which I sumarize like this:
1. The main purpose of the cabinet is to dispose of all backwards sound from the speaker leaving only the clean front firing waves.
2. The walls of a normal box will sing and it is hard to stop it with stiff panels, bracing etc.
3. Building a box within a box gets you along way toward eliminating sound radiation from the sides and back of the box.
In this case CLD is more or less a "box within a box". The driver should therefore be fixed to the inner surface. The CLD construction then reduces vibrations reaching the outside and hence unwanted sound radiation from the front panel.
That article prompted me to a simple experiment: Twang a 440Hz tuning fork, the only one I happen to have, and touch it to the panels of a speaker box. Sure enough all speakers I have tried ring like bells!! Seems possible that more sound could come from the sides of the box than the cone itself.
I would be interested in the results of doing this with your CLD boxes. Perhaps using a little speaker fixed inside the box as a stimulous and see how much ringing gets out.
heater,heater said:
I agree with you regarding attaching the drive to only one layer, but how about this for a reason to attach it to only the outside layer.
If you attach the driver to the inside layer, then the box which you have coupled the driver to is the box directly affected by the back wave of the speaker. Therefore, the driver itself would be affected by vibrations from the inside box - as, by construction, this box would be allowed to vibrate more than a single box construction (as it is generally fixed to the floor). The outside layer would, by reasoning of not being either attached to the driver, or being directly affected by the back-wave from the dirver, would vibrate less than for a single box construction.
BUT, if you attach the driver to the outside box, then the back-wave is now transmitted only to the inside box. Not only is the inside box isolated from the outside world, but the driver itself is not directly affected by vibrations from its own back-wave.
Just a thought.
Maybe ask Geenius (Tony Gee), as his Andromeda used this construction.
Good luck, Spilferderber, I don't have the skills or the patience for this kind of construction.
Actually alot of vibration is transmitted from the mounting flange of the driver through the baffle and to the rest of the walls of the enclosure.
If you are going to go to the trouble of CLD, and would like to isolate the outer box completely, I would suggest mounting the driver by the magnet to the inner layer. This way the only contact between inner and outer is the viscous layer and the whatever material you use for a gasket between driver flange and outer baffle.
Hense my suggestion of well nuts, which if mounted cafefully you could attach the driver to whichever layer you wanted (infact you could try both configurations, it doesn't matter as it should effectively isolate the driver from the enclosure anyway) and then use a compliant gasket around the edge of the driver basket.
I believe this is suggested in the Loudspeaker Cookbook, although not specifically referenced to CLD enclosures.
I believe this is suggested in the Loudspeaker Cookbook, although not specifically referenced to CLD enclosures.
I'm planning to make cabinets, using CLD structures in the following way:
- outter layer: MDF, 16mm
- damping layer: cork plates, 3mm
- inner layer: birch plywood, 19mm
+ bitumen damping
i have three separated cabinets, one for subwoofer (BR enclosure), one for woofer (sealed enclosure) and one MTM cabinet.. for all three I'm planning to use the described CLD structure.
the issue is how to mount drivers and BR tubes.. a friend on the local DIY page also suggested that there would be no problem even if the layers would be stiff connected with the driver screws since they're connected only on a very small surface.
- outter layer: MDF, 16mm
- damping layer: cork plates, 3mm
- inner layer: birch plywood, 19mm
+ bitumen damping
i have three separated cabinets, one for subwoofer (BR enclosure), one for woofer (sealed enclosure) and one MTM cabinet.. for all three I'm planning to use the described CLD structure.
the issue is how to mount drivers and BR tubes.. a friend on the local DIY page also suggested that there would be no problem even if the layers would be stiff connected with the driver screws since they're connected only on a very small surface.
There are two things to be considered here: firstly the backwave from the drivers cone and secondly the newtonian reaction from the drivers motor.
If you mount the driver to the inner shell the cabinet wall radiation will be reduced but the motor reaction will try and move the driver in the opposite direction so the inner shell needs to be very high mass to counter this or you'll loose definition and detail.
Couple the driver to the outside shell and the CLD wall damping will be much less effective due to the outer shell being directly exited by the driver motor reaction.
I think mounting the driver to the inner shell but skipping the CLD on the bottom face of the box is likely to be best. The driver must have a clear mechanical path to ground. That way youll get the best of out of the CLD without loosing detail and impact. Ive tried well-nuts -avoid!
Simon
If you mount the driver to the inner shell the cabinet wall radiation will be reduced but the motor reaction will try and move the driver in the opposite direction so the inner shell needs to be very high mass to counter this or you'll loose definition and detail.
Couple the driver to the outside shell and the CLD wall damping will be much less effective due to the outer shell being directly exited by the driver motor reaction.
I think mounting the driver to the inner shell but skipping the CLD on the bottom face of the box is likely to be best. The driver must have a clear mechanical path to ground. That way youll get the best of out of the CLD without loosing detail and impact. Ive tried well-nuts -avoid!
Simon
mounting the driver only to the inner layer seems to bring more problems.
the main problem: the driver must somehow lean on the outter layer, right?
it seems that we must resolve one more (fundamental) question:
what causes more vibrations, the air behind cone (hitting the walls) or the mounted driver, attached to enclosure?
I think the air is the bad guy here..
the main problem: the driver must somehow lean on the outter layer, right?
it seems that we must resolve one more (fundamental) question:
what causes more vibrations, the air behind cone (hitting the walls) or the mounted driver, attached to enclosure?
I think the air is the bad guy here..
Member
Joined 2003
Here is an interesting read, courtesy SL.
I (think) I also recall a TAD discussion with a net result of mounting drivers by the magnet structure, preferably from the back wall of the box so the energy transmitted to the box is furthest from the listener.
So, the best solution may be to mount the magnet to the rear of the inner box (or bottom if in contact with the floor) with the lip of the driver isolated from the front baffle by a rubber seal, heavy felt, etc.
Paul
Linkwitz on driver mounting
I (think) I also recall a TAD discussion with a net result of mounting drivers by the magnet structure, preferably from the back wall of the box so the energy transmitted to the box is furthest from the listener.
So, the best solution may be to mount the magnet to the rear of the inner box (or bottom if in contact with the floor) with the lip of the driver isolated from the front baffle by a rubber seal, heavy felt, etc.
Paul
Linkwitz on driver mounting
And, one more thing, unless you have incredible damping material inside your inner cabinet (or design the inner cabinet to be a tapered, closed, transmission line) then the sound will try to force its way out through the driver after bouncing around inside the cabinet.Spilferderber said:
One advantage of CLD over standard cabinets (however you mount the driver) is that you have a heavier cabinet that is inherently less rigid. As long as your driver is able to behave properly (that is, it doesn't move excessively due to it's own diaphram movement) then the cabinet will provide better performance.
You could always buy some large blocks of granite and hollow them out, cover them in bitumen and then a layer of your favourite wood. Your only problem would be moving them
! Sorry, being a bit silly here - but I'm just illustrating that you will have to compromise somewhere (it's just a matter of you choosing where to draw the line).There are some other issues that you need to think about which can color the sound as much as the cabinet vibrations induced from pressure waves inside the cabinet. The speaker frame itself needs a very rigid and strong mounting place. For midrange and higher frequencies it is more important to have higher mass attached to the driver frame which resists excitation.
IF you try to go to far in isolating the driver mounting you will make the sound mushy because of too much compliance at the speaker frame attachment location.
The driver baffle needs to be the heaviest and most rigid and dense panel in the speaker cabinet. The cabinet is excited as much by the physical motion of the driver as from pressure waves inside the cabinet.
If you want fantastic sound make the front baffle out of four inch thick cement or very thick layers of MDF laminated together. Not constrained layer damped at this baffle. Also, if the center damping material is too thick it will make the panel very damped but slow to start and stop once excited. So the sound can be a bit sluggish. The constrained layer damping only works well with two panels of equal thickness. They must have the same mass to counteract each other.
The speakers that I have heard which used this thick front baffle approach sound fantastic. But often they are so heavy to ship that companies use other methods. But for the DIY projects this is not so much of a big deal.
I would favor an attachment to the outer cabinet with a very thick panel. If it is very thick it will actually work better and sound better than thinner constrained layer damping.
In fact, you could make the front baffle only a single thick panel of at minimum 2 inches thick and attach it to both the inner and outer cabinet walls. 3 inches thick would be even better. Four would be sublime. But these are trade off you have to decide based on space limitations.
However, increasing the depth of the cabinet is often the easier thing to live with.
In practice, mounting the driver solidly to the baffle with some firmly clamped resilient material in between the driver frame and the baffle seems to work very well because it holds the driver firm to the baffle but allows some energy stored in the driver frame to be absorbed in a controlled way.
IF you try to go to far in isolating the driver mounting you will make the sound mushy because of too much compliance at the speaker frame attachment location.
The driver baffle needs to be the heaviest and most rigid and dense panel in the speaker cabinet. The cabinet is excited as much by the physical motion of the driver as from pressure waves inside the cabinet.
If you want fantastic sound make the front baffle out of four inch thick cement or very thick layers of MDF laminated together. Not constrained layer damped at this baffle. Also, if the center damping material is too thick it will make the panel very damped but slow to start and stop once excited. So the sound can be a bit sluggish. The constrained layer damping only works well with two panels of equal thickness. They must have the same mass to counteract each other.
The speakers that I have heard which used this thick front baffle approach sound fantastic. But often they are so heavy to ship that companies use other methods. But for the DIY projects this is not so much of a big deal.
I would favor an attachment to the outer cabinet with a very thick panel. If it is very thick it will actually work better and sound better than thinner constrained layer damping.
In fact, you could make the front baffle only a single thick panel of at minimum 2 inches thick and attach it to both the inner and outer cabinet walls. 3 inches thick would be even better. Four would be sublime. But these are trade off you have to decide based on space limitations.
However, increasing the depth of the cabinet is often the easier thing to live with.
In practice, mounting the driver solidly to the baffle with some firmly clamped resilient material in between the driver frame and the baffle seems to work very well because it holds the driver firm to the baffle but allows some energy stored in the driver frame to be absorbed in a controlled way.
You are absolutely right, there is no rigid way to mount a driver by its frame and isolated vibes. Which is the main reason for mounting by the magnet.
Be careful here, if you add more mass than rigidity the resonance of the panel is actually lowered: res. = sqrt(stiff./mass)
A general rule of thumb higher density is usually accompanied by lower internal damping- of course as I wrote that I began to question if it is true as I learned in my materials class. I am leaving that in hoping someone who knows more about the relationship between density and damping will chime in.
This may be true in some respects, but I discussed this with a college professor who has done alot with vibration isolation and based on his statements the two panels should have nearly equivalent stiffnesses and the mass and thickness are not factors.
Mass-counteraction, is not quite how the energy is dissipated, the visocelastic layer between the substrate and the constrained layer deforms under shear easily (more than other layers)by converting the shear energy into heat.
The problem with really thick layering is that all of those layers store energy and what energy doesnt get converted to heat WILL reradiate as sound. Here is what sucks about bracing/layering more than dampening- if you reflect too much of the internal sound energy in the enclosure off the interior walls(low transmission through panels), then the only way out is through the cone! This is why OB sounds so good- almost all enclosures reflect too much sound back onto the cone
CLD converts the energy to heat instead of simply transmitting the vibe or reflecting it back inside.
nunayafb,
This is an example were your analysis is not complete. While it's true that the thicker panels can store more energy. It requires a lot higher amounts of energy to get the energy into the panel because the panel resists energy being put into it to a much larger degree. Assuming that it is also stiff and dense. As such, in everyday real terms the method works very well.
Because of this, more of the drivers motional energy is transduced into the air and this improves the dynamics and resolution.
This is an example were your analysis is not complete. While it's true that the thicker panels can store more energy. It requires a lot higher amounts of energy to get the energy into the panel because the panel resists energy being put into it to a much larger degree. Assuming that it is also stiff and dense. As such, in everyday real terms the method works very well.
Because of this, more of the drivers motional energy is transduced into the air and this improves the dynamics and resolution.
...resists energy being put into it... ok I would love to hear the technical explanation of how this is possible.
you are completely ignoring the effects of damping, stiffer/thicker panels do not damp vibrations(neglecting small internal damping) they only increase the resonant frequency- but multi layered panels also get heavier which causes a reduction of the resonant frequency.
Any baffle that gets significantly more stiff with minor increases in weight will have a higher resonant frequency which is usually accompanied by extremely small baffle deflections, this is good. But that high frequency energy will efficiently transfer to air and along the baffle to the side panels better than low frequencies due to the frequency dependent impedance of air.
If there is no damping on a thick baffle and the driver is rigidly mounted to it by the flange then the energy will transfer to the baffle. I dont see the benefit of a thick heavy baffle when the only reason the magnet assy. deflects is because the driver is flange mounted. If you mount it by the magnet the frame isnt stressed and no (significant) energy will transfer to the baffle at any frequency, damped baffle or not.
Furthermore, if the thick baffle did resist energy input then the energy in the driver frame would have to go somewhere right? It either goes into the cone or into the baffle, neither one is good right? The energy needs to get converted to heat or it will get converted to sound, CLD is designed for that purpose while thick baffles were designed for stiffness.
Fyi, I am aware of how important rigid panels are and the thicker the better here. Referencing the thick baffled speakers you heard Im curious did you ever stop and think if they might sound better with a thinner(1-2") highly damped panel? Not only thin and damped but also mounted by the magnets instead of flanges.You have to admit they might, and Im willing to say they might not although I think they would sound the same or better. I am not doubting that those speakers sounded good, even great, but from a standpoint of energy storage and transfer, weight, bulk and rear "breathing" room the CLD would be better.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.