In what respect? Did you use the exact same bracing method in both of your builds? I am interested.
I've made commercial cabinets from plywood and from MDF over the years, but in the past I usually fiberglassed the inside and outside, so the mdf or plywood was more of a form and interior damping with the real strength applied afterward. In rough road use MDF was a problem once the fiberglass became damaged, as MDF sucks up water and then explodes. It has to be completely encased and protected. That's why the exterior is all ply on the speakers I am currently making. When I used to be a roadie I was behind large poorly-braced speakers often, and could hear the 'bonk bonk' noise of sheet plywood all too often, which just shows that either material can be used well or poorly. A lot of people get a bad impression of MDF from using the worst of it: the 1/2" stuff sold for shelving at the local home store. That stuff is only bonded worth a darn at its surface, and has no strength and no redeeming dimensional thickness. Planet10 mentioned the impedance change between layered materials as part of why he likes plywood. Kind of like mumetal shielding? Deflecting outside wavefronts of low incident angle? Or more like fiber optics keeping the energy reflecting inside rather than allowig it to escape? Or is it more a matter of preventing transmission in certain directions? Anyway, I'm pretty certain the thick MDF was a better choice for my baffle boards than thinner plywood; I think that the shape is important and the thickness is pretty critical and I love the way I can cut perfect circles in MDF every time without splinters.
This RedGuard stuff turned out to be an interesting experiment. After about 7 coats its built up pretty heavy in the corners. It's really like the entire inside of the cabinet is encased in a plastic coating. It's not as rubbery as I'd thought. It's deformable but doesn't rebound. I first painted the cabinets with heavy commercial speaker cabinet paint, inside and out, then I did the inside with this heavy hard-rubbery liquid plastic stuff.
Now I'm drilling the holes for bolts and t-nuts to secure the real removable baffle boards to the old original kit's thinner glued-in baffle-board. I have to decide tonight how many 1/4 inch bolts to use.
This RedGuard stuff turned out to be an interesting experiment. After about 7 coats its built up pretty heavy in the corners. It's really like the entire inside of the cabinet is encased in a plastic coating. It's not as rubbery as I'd thought. It's deformable but doesn't rebound. I first painted the cabinets with heavy commercial speaker cabinet paint, inside and out, then I did the inside with this heavy hard-rubbery liquid plastic stuff.
Now I'm drilling the holes for bolts and t-nuts to secure the real removable baffle boards to the old original kit's thinner glued-in baffle-board. I have to decide tonight how many 1/4 inch bolts to use.
I am really hoping that my input here has not already been covered. I thoroughly read the first 40% of this thread, and basically skimmed the rest to see if anything along my lines of thought had been shared. I don't believe it has but if this is not the case please forgive.
The following is "theory..."?
1. Panel resonances are logarithmically centered events occurring where a sound wave effects the panel at large somewhat equally, offset by whatever natural suppression is provided by material characteristics (thermal conversion and/or mechanical suppression)???
2. Panel resonances can be excited by higher even ordered resonant octaves (assuming the wavelength is still as large or larger than the panel)???
3. Wavelength size matters! Once the wavelength size is smaller than the panel size, the ability for that wave to propagate a resonance on that panel is logarithmically reduced???
4. Higher frequency energy is far more effectively converted to thermal energy in both the panel "thickness" and, prominently, in dampening material if added. ?? (especially as the wavelength approaches panel thickness?)
--- conclusion: pushing panel resonances up is a good thing. For every octave higher the resonances are pushed, there is a logarithmic improvement in suppression as a combination of the factors listed above.
Again, ^ THEORY^
Discuss
The following is "theory..."?
1. Panel resonances are logarithmically centered events occurring where a sound wave effects the panel at large somewhat equally, offset by whatever natural suppression is provided by material characteristics (thermal conversion and/or mechanical suppression)???
2. Panel resonances can be excited by higher even ordered resonant octaves (assuming the wavelength is still as large or larger than the panel)???
3. Wavelength size matters! Once the wavelength size is smaller than the panel size, the ability for that wave to propagate a resonance on that panel is logarithmically reduced???
4. Higher frequency energy is far more effectively converted to thermal energy in both the panel "thickness" and, prominently, in dampening material if added. ?? (especially as the wavelength approaches panel thickness?)
--- conclusion: pushing panel resonances up is a good thing. For every octave higher the resonances are pushed, there is a logarithmic improvement in suppression as a combination of the factors listed above.
Again, ^ THEORY^
Discuss
The plywood had greater DDR -- downward dynamic range... it was as if the MDF box was oozing out a whole buch of time smeared energy/noise that was burying the low level stuff.
This was early, and the bracing methods can be seen in many of my plan documents.
If you can wade thru it the thread i linked to above covers off a lot.
dave
Everytime the energy going thru the plywood there is a loss of energy (ie damping/turned to heat) at every change of impedance. Each layer will have a different impedance from the adjacent on due to the change in grain direction as well as one at every glue layer.
dave
5. As frequency goes up the energy decreases (inversly to the square of the freuency, and it can be argued that other factors, (1 of which is #4) bring it to the 4th power)
dave
Now if I could just figure out how to translate this into reality. If I take voltage @2K and put it across a driver, then put the same voltage across the same driver @4K, and get equal dB (assuming something like a dome tweeter with very similar sensitivity at each point), then why are we not getting twice the dB at 4K with equal voltage, efficiency, current and power consumption?
I believe what you are saying is true, in fact, #5 was edited out before posted and very closely resembled what you are saying, but I couldn't convince myself to leave it in there because there is too much evidence to support the opposite. The only time i see this concept start to "show" it's ugly face is in the sub-100hz range, where there is no free lunch, only piles upon piles of lost energy and attempts to reclaim it.
I do not see what that has to do with what i said. If we have acoustic energy at 1k and an equal level at 2k the energy in the 2k signal is 1/4 as much.
This tidbit 1st cam eto my attention from a post by Svante, and since them i have dug out of one of my acoustics text the equation that it comes from.
To excite a panel resonance you have to inject energy within the bandwidth of theresonance. If the resonance is higher there is less energy to excite the resonance. If the bandwidth is narrower (higher Q) there is less energy to excite it. So if you design a box to push the resonances up higher (as long as you get above a certain threshold) then it is unlikely to get excited. If a resonance never gets excited it as if it doesn't exist.
dave
Thanks for the response. This is definitely a subjective and objective observation. I believe that the bracing for any speaker cabinet construction is a key factor in the actual end result.
I agree...
but where did the power go? a typical response chart with equal voltage through the range should look like a skyrocketing screech to heaven for any driver based on my lack of understanding?
I realize that I am missing a piece of the puzzle here. Try to fill me in if you can. If power requirements drop by 1/4 per octave, please try to fill in a blank for that correlates the ability for a driver to stay "flatish" while being driven with equal voltage.
Thank You, and Regards,
Eric
Go read up on how a dynamic driver works. I'm not the one to explain the actual details.
dave
But it is far from being the only factor.
Can you post apic of how you would brace a box?
dave
I absolutely agree with that! We all well know that driver selection, crossover design, and magic (in no particular order) are relevant to success! Actually, I lean towards the crossover design as being one of the most important determining factors. Uh oh, didn't mean to go THERE again.
Peace, Scott
Yes I can. Not only how I would, but how I did. It's late here so maybe not today, but I will post my build.
Regards!
we are going off-topic here, i was talking about cabinets, you ar enow bringing up the whole speaker
Crossovers are evil. The best XO is no XO, But if you want True 10 octave you have to compromise. One should at least keep it out of the critical band, and try to maintain 1/4 to 1/2 octave c-c driver separation
dave
Better than most and you gotta love the BB.
The baffle and drivers are missing their braces.
Would be better if the braces reached to the other side.
Drawing does not fully show brace placement.
Crossover <400 Hz?
dave
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.