Triangles are indeed wonderful. In this case, they help keep the square framework square, which if you had concern about the square shifting into a parallelogram, the triangle is your friend.
Since you are looking for anti-flex of the side walls, perpendicular bracing would be more effective.
Since you are looking for anti-flex of the side walls, perpendicular bracing would be more effective.
ScottG , Do you mean somthing like this ?
I never seen anyone build a cabinet like that before ... But it appears to be solid with 30mm wall thickness and all the gluing that needs to be done .. but , well it's weird
Anyone approves ? The problem is I dont see a way to chamfer the inside circles , Only by sanding it by hand a bit .. also incorporating braces that interlock with the design is a pain.
Also building it like that would require 192 "flanks" for 2 cabinets ... can you say labor cost hehe
An externally hosted image should be here but it was not working when we last tested it.
Yes, that's it - but it should be 35mm (good ply).
Lay-up the baffle first on your bench (without glue) using bracing, and do your routing.
And yes, it would be a lot of labor. It would also be expensive if you use a router for the cuts, (use a table saw), OR you have someone else do it. (..if it were done on a production run basis a high power laser cutter would probably be in order.)
Again, forget the bracing - done this way it doesn't need it (..but it should be 35mm).
here is some inspiration :
http://www.diyaudio.com/forums/multi-way/155739-troels-pms-multilayer-assembly-build-thread.html
http://www.diyaudio.com/forums/multi-way/155739-troels-pms-multilayer-assembly-build-thread.html
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.
Last edited:
Right!
Furthermore DIFORCE remember the brace must be more rigid than the walls it is attached to to correctly make its job.
Well, I can't even seem to lead the horse to water....
IMHO, if a panel can move enough to resonate, then it can move enough to transmit sound. I think way too much emphasis in this thread has been placed on resonances. While sure, at resonance a panel is most efficient, it will still be transparent across a broad range of frequencies. If you brace a panel to remove the largest resonance, then you just have to deal with the next largest. While putting a brace in the center of a panel may not be the most effective thing in combating resonances, it will be most effective approach to reducing the total motion of the panel, and thus in reducing the total leakage of the cabinet.
IMHO, if a panel can move enough to resonate, then it can move enough to transmit sound. I think way too much emphasis in this thread has been placed on resonances. While sure, at resonance a panel is most efficient, it will still be transparent across a broad range of frequencies. If you brace a panel to remove the largest resonance, then you just have to deal with the next largest. While putting a brace in the center of a panel may not be the most effective thing in combating resonances, it will be most effective approach to reducing the total motion of the panel, and thus in reducing the total leakage of the cabinet.
There are two different panel movements that we wish to stop.
1. If you imagine the woofer cone moving inwards then the pressure inside the enclosure increases and this tries to push the panels outward. When the cone moves outward pressure decreases and the panels are forced inward. This happens at the rate determined by the frequency. Think in terms of cycles per second (Hz.) Cross bracing joins opposing panels preventing this movement. There is a point reached where the panels will not move and thus more bracing is a waste. The size of the panel and the panel thickness determine when this point is reached.
2. Vibrations also cause panels to move. Vibrations run around the enclosure walls. They start from different sources including (1) above but also from the driver frame. Cross bracing does not stop this form of panel vibration and in fact can transmit energy from one panel to the opposing panel. To reduce this energy panels need to be braced from the panel to the two opposing corners. this is triangular bracing and will stop all forms of movement.
3. Just for completness remember that the fixing methods (eg Glue) needs to be 'hard' as opposed to flexible. Thankfully PVA is good. At the other end of the spectrum silicone is useless.
To conclude, triangular braces will stop the panel from all vibrational energy, cross braces will not.
Terry
1. If you imagine the woofer cone moving inwards then the pressure inside the enclosure increases and this tries to push the panels outward. When the cone moves outward pressure decreases and the panels are forced inward. This happens at the rate determined by the frequency. Think in terms of cycles per second (Hz.) Cross bracing joins opposing panels preventing this movement. There is a point reached where the panels will not move and thus more bracing is a waste. The size of the panel and the panel thickness determine when this point is reached.
2. Vibrations also cause panels to move. Vibrations run around the enclosure walls. They start from different sources including (1) above but also from the driver frame. Cross bracing does not stop this form of panel vibration and in fact can transmit energy from one panel to the opposing panel. To reduce this energy panels need to be braced from the panel to the two opposing corners. this is triangular bracing and will stop all forms of movement.
3. Just for completness remember that the fixing methods (eg Glue) needs to be 'hard' as opposed to flexible. Thankfully PVA is good. At the other end of the spectrum silicone is useless.
To conclude, triangular braces will stop the panel from all vibrational energy, cross braces will not.
Terry
The amount of bracing needed is dependent on the expected range and power of the system. Unless I missed it, I don't think that's been mentioned.
As the resonant frequency increases, so does ear sensitivity. That's why listening tests are important too.
I have had very good luck with the shelf brace, but the + is far more effective than the X; since you want both compression and tensile strength, probably good plywood is better than MDF, although a 1X1 MDF attached to the shelf brace along the sides (so that the shelf brace side pieces are ~2" thick) would add some damping.
I would strongly recommend that the shelf braces not be placed integral distances from the top, bottom, and each other. The idea is to make the unbraced sections resonate at different, non-harmonically related, frequencies.
For a crude example, braces at 12 and 20 inches above the floor, giving sub-panel widths of 12, 8, and 10 inches. Always best if you can work in the golden ratio or some other irrational number.
As the resonant frequency increases, so does ear sensitivity. That's why listening tests are important too.
I have had very good luck with the shelf brace, but the + is far more effective than the X; since you want both compression and tensile strength, probably good plywood is better than MDF, although a 1X1 MDF attached to the shelf brace along the sides (so that the shelf brace side pieces are ~2" thick) would add some damping.
I would strongly recommend that the shelf braces not be placed integral distances from the top, bottom, and each other. The idea is to make the unbraced sections resonate at different, non-harmonically related, frequencies.
For a crude example, braces at 12 and 20 inches above the floor, giving sub-panel widths of 12, 8, and 10 inches. Always best if you can work in the golden ratio or some other irrational number.
This is incorrect if the "cross" braces (three perpenticular ones) are rigidly connected together at the centriod of the enclosure. Then they work basically the same as the triangular braces that you have defined.
I understand your point and connecting cross braces is much better than not but it is not basically the same. Triangular bracing is still superior, more so if you connect triangular braces in the middle. It is the corner connections that give the triangular braces their unyielding characteristics.
Terry
What you want is to prevent the cabinet sides from ballooning; the movement will be greatest the furthest from the edges, that is the middle. The X brace is the same as a simple beam with respect to the middle; you really, really want the +. Incidentally, you don't want the cross pieces equidistant from the front and back; if the cabinet is 18" front to back, then the cross pieces should be 11" inches from the front, or some such.
I agree with the post that said the X would be great at resisting torsional forces; but since there are none...
I agree with the post that said the X would be great at resisting torsional forces; but since there are none...
Well perhaps I don't understand your point, because I don't see it. Since a triangle is three pieces (per face) and what I am taking about is three pieces total, it's not apples to apples. Sure more braces, thicker, more total material is going to be stiffer. But let's say that we are comparing the same total size and number of braces, I would put my recommendation up against any other approach.
As I said before, if you want stiff, just fill it with cement ... but that's not what we are looking at. We want the most effect for the least material, not some "sky is the limit" concepts of the "ultimate stiffness".
Make the enclosure out of steel, or here is a great idea! what about alluminum - thats real practical!
This idea of the ultimate stiffened enclosure gets ridiculous and the marketplace bears testimony to that fact!!
... says the man who makes polyurethane boxes!
A comment to the anti bracing brigade ... I believe one theory of bracing is to first aim for maximum stiffness to push resonances up. At a higher frequency, MDF is generally more effective at preventing sound transmission, which should reduce their severity. I say should because without measurements I'm guessing, but it's not a crazy guess. Then you can use mass damping of the enclosure which I'd also guess is now more effective. I have seen some measurements showing that mass damping is effective in the midrange, but does nothing at all for lower frequencies. So in a sub box you want pure stiffness, but for full range stiffness and mass work together.
Of course in a DIY project things are different. It only has to travel from the workshop to the house, overkill isn't so costly. In a commercial speaker every increase has to inflate the price to a much greater extent.
Gedlee, with your suggestions you are dividing a square structure into more sqaures. Square shapes allow movement, triangular shapes do not. That is a fundamental truth.
Bracing is effective up to a point. Build a large enclosure from 1/2" material and you will need lot's of bracing. Build it from 1" material and you need very little. The size of the panel matters a great deal. The laws of diminishing returns sets in early.
Terry
I can tell you that I've built several enclosures from 1" MDF, and they certainly will resonate without additional bracing. In fact. it's impossible to kill all enclosure vibration, certainly not a chance with plain MDF or plywood, so as a general rule, more is better.
I agree up to a point. Unfortunately we tend to use this sort of thinking based on out concept of what an enclosure is. As an example, 3" driver in a 6" cube made from 1" MDF would not need any bracing.
You need to consider
1 Driver size and the volume of air to be excited
2 Wall thickness and material flex
3 Panel size.
Then consider what bracing you need.
If you combine an x brace with a + brace then you have a very good system. The x brace divides the plane into triangles and this provides absolute stability, the + pieces stop the panels from moving at point of contact. All structures that require stability are designed on the basis of triangular bracing.
Terry
Wow! There is a gem of misinformation! A square membrane vibrates, but a rectangular one does not? Do I have that right?
Hi Paul - the switch to Poly had nothing to do with stiffness or stuctural strength. It was all about assembly and finishing (but thats another topic).
I'd agree with a lot of your position: here is mine - a minimum bracing - "cross" - is essential, but beyond that are diminishing returns and you will get much more effect from CLD than any amount of increased bracing (I actually do both). I've been way down this road ...
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.