Removable "Active Brace"

[polkymon]

I never like to entomb my x-overs. I can either get to them through the woofer cutout or make an access panel.

My latest project made an access panel difficult to implement due to aesthetics and internal bracing would have made it impossible to use the driver cutout.

I came up with an idea using a turnbuckle. I attached large eye hooks to the inside of the side panels with a threaded rod between them and large washers and locknuts.

Turning the turnbuckle by hand gives me plenty of torque and also makes construction easier by securing the side panel during assembly.


The x-over rests on the long slanted panel just below the woofer. I also mounted the x-over with velcro. Bottom compartments are filled with kitty litter.

[sdclc126]

This is a very interesting idea! I've often wondered if something like this could be done - I just never conceived of how to implement it.

My idea was to use some kind of tension- or torque-based wood bracing, i.e., using bracing that was slightly over- or under-sized for the cabinet; in the case of undersized bracing, screws would be used in addition to glue to provide a secure attachment and to maximize the torque.

Putting some kind of tension on the cabinet walls seemed at least conceptually to me that it would minimize resonance through brute force. Perhaps this would raise resonance frequencies above audibility, and/or reduce their amplitude below audibility.

But your idea is much simpler though, and it's adjustable! Now I'm wondering if this imparts a measurable advantage, and if you've attempted to measure this in any way?

[polkymon]

Nope. No measurement tools here. It would be interesting though to see the differences at different torque levels.

(OT) I am going to be looking into meas. tools and would like to hear your recommendations.

[sdclc126]

I can't personally recommend anything as I don't have any equipment either, but I'm sure there are lots of others here who will make some suggestions.

I don't know if box resonance is/can be measured in the same way or with the same stuff used to evaluate drivers, so this would be good info for us both.

[nunayafb]

As far as rigidly bracing two opposing walls goes the higher the Modulus of Elasticity(E) the better. E refers to strain(change in lenth divided by total length) resulting from stress(Force divided by area), higher E means stiffer. Typical E for woods is on the order of .5Mpsi, typ. MDF is .4Mpsi, steel is around 30Mpsi.
The important aspect of E is that for most common construction materials it is a linear response up to near the region of the materials yield strength, where permanent deformation occurs.
Why do you care?
Well if you use metal braces you get very high rigidity due to the high E, however if you preload the walls of the box you dont get any additional benefit- remember that linear E, the deflection of the panels wont be affected by preloading due to the constant "spring" rate. This is actually a good way to look at it, if you preload a linear spring a little and them apply a force it will still deflect the same as without the preload.

Ex.
Stress = Force/area, Assume F=50lb, area for your steel,(3/8"dia.?)
As=pi (3/16")^2. = .11in^2 stress is about 450psi.
stress=strain*E -> strain = stress/E: = .000015in mult. this by brace length to obtain change in length (of brace)

How big does an MDF brace need to be to match this?
some algebra and : A = F/(E*Strain); (E=.4mpsi) (strain from above)
A= 8.33in^2, possible dimension, A/thickness gives width of brace: if thickness is .75" the width needs to be about 11" inches to give the same rigidity as the rod. Except for the deflection in the handle you used.

[sdclc126]

"...if you preload a linear spring a little and then apply a force it will still deflect the same as without the preload."

Ah - "preload" was the word I was looking for. Then a progressive spring would work better? If so could this be implemented in a loudspeaker enclosure? Perhaps this is another way of describing constrained layer construction?

The other question this raises for me is, based on E it seems that metal would be a far superior material for bracing - has this been done? I could do a search but since we've brought the subject up here might as well ask now. And how would metal braces best be attached? Screws seem obvious but perhaps there are other methods.

[nunayafb]

Then a progressive spring would work better?

Well "progressive" only works for springs, and interestingly, progressive springs still have linear E's.

In addition, I only addressed the physical bracing of the metal, due to its high density(high mass lowers Fs) and crystalline structure it also transmits vibrations very well, on the flip side have you ever heard of an MDf tuning fork?

If so could this be implemented in a loudspeaker enclosure?

If you damp the steel yeah, if you damp the mounting point of the steel to the wood then no. The steel needs to be rigidly bonded to the surface or else you might as well just glue on some grained wood, if the E of the glue is lower than the brace then it will stretch whether the brace is steel or wood. Although by a very small amount, factoring in mounting, damping and cost I would just stick with wood. Bracing a box does little to reduce panel vibration amplitude anyway, but does drastically raise the resonant point. Internal damping via C.L.D and fiber based, rubber based etc. all lower panel resonance more than bracing.

The other problem with steel is that one might be tempted to use thin rods. The brace does two things, it reduces wall flexure by tying the opposing walls together, and it bisects the walls which raises the resonant frequency of the wall- the goal being to drive it beyond audibility.
If you use shelf bracing or equivalent you truly divide the walls, if you use thin rods/dowels you only partially bisect a small portion of the panel area regardless of how rigid the brace is.

[polkymon]

I considered the transfer of vibrations and wrapped four layers of duct tape around the threaded rod then sprayed Quiet Coat over the entire assembly. Not very scientific and not sure how much difference it made. Btw the walls are 2 x 3/4" mdf.

[noah katz]

"Well if you use metal braces you get very high rigidity due to the high E, however if you preload the walls of the box you dont get any additional benefit"

Apparently you haven't heard of stress stiffening.

Google hit #4 is apropos to speaker panels: "Stress stiffening comes from the fact that the membrane is fixed on all sides which causes the membrane to change stiffness with deflection."

[nunayafb]

From your own link:

Also, you may have
material nonlinearities (plasticity) that will arise from the large stress
induced by trying to deflect a membrane that much. The stresses tend to be much higher because of the constraints involved and large axial strains are required to accomplish a small lateral displacement.

Maybe you should read closer noah, stress stiffening is accomplished by axially loading a membrane near the yield point. ie, large stresses that push the part to near failure, this requires rigid end conditions at the edges of the membrane to hold it in biaxial strain.

Preloading with a brace is simply applying a normal force to a linearly elastic panel which is not under any significant axial load(neglecting moisture expansion/contraction).

The stress(St) is St = Mc/I M is the bending moment induced by the preload force c is the distance from the midpoint to the outer/inner fiber(thick./2) and I is the inertia about the neutral axis(mid-thickness).

"Apparently you haven't heard of stress stiffening."