planet10 said:
Reflection from the divider back through the cone? The carpet improved the situation immensely, but not entirely. What I'm saying is are the panel resonances in the trouble range, and are they enough to contribute? I found that giving the woofer a more rapid cut on the crossover cleaned up the rest. Perhaps by doing so, the frequency that excited the panels in the first place, was missed. Just a thought.
By the way, for what it's worth, I didn't get the reflection issue with the original version, which had the divider at the same depth. A 1st order series filter gave "good" results.
Also, I haven't abandoned the resonance test preposed earlier in this thread, I just don't have the 1/2 inch MDF yet (thought I did, but iy was 5/8). And now that I have been converted to BB, I need an excuse to buy it.
fwiw some math on the relationship between size, mass, thickness and MOE, and panel resonant frequency
www.snippets.org/pipermail/diyspeakers.not/2002-September/003539.html
www.snippets.org/pipermail/diyspeakers.not/2002-September/003539.html
rick57 said:
Damping: Is any effect, either deliberately engendered or inherent to a system, that tends to reduce the amplitude of oscillations of an oscillatory system.
Take a tuning fork as an example of an oscillatory system. What is the more effective means of damping it's vibration? Allow it be damped by the air that surrounds it, or wrap it with a cloth?
MJL21193 said:
In a sealed speaker box, the energy released from the driver into the box has to be dealt with. This energy takes the form of mechanical vibrations, as the sound wave act on the interior of the box. If a panel is free to vibrate, it will do so, dissipating the energy through movement. If it moves freely, energy is dissipated quickly. If it is constrained, energy dissipates slowly (so called energy storage). To release this energy (remember the tuning fork?) damping is needed.
rick57 said:
The Q factor or quality factor compares the time constant for decay of an oscillating physical system's amplitude to its oscillation period. In other words: does amplitude decay quickly or slowly during the oscillation period. Compare to tuning fork again, air damping - high Q, cloth damping - low Q.
Since it's the target to reduce vibrations quickly, low Q damping works the best.
MJL21193 said:
No. You haven't grooked what i'm trying to do.
If we assume that a panel will resonant no matter what we do (true), then what i am trying to do is move the resoances to a place where they will never be excited. If excited by an external source (ie a whack with a steel rod) it is not silent ... but that is not something that happens when you are listening to music.
Think of it as magic, sleight of hand.
dave
MJL21193 said:
The actual air moving inside the box actually has much, much less energy to drive a panel resonace than the mecanhcal mechanism that attahes the driver to the box... this is why push-push is so effective, The majority of that energy is actively cancelled. And airspace energy to drive a panel resonance declines vwey rapidly with frequency.
And further, a lot of the stuff used in practice to damp panels, stores energy.
dave
MJL21193 said:
True, but it is not the Q i am talking about. By high Q i am talking about the energy required to stimulate a panel resonance. This implies that a large amount of energy in a very narrow bandwidth needs to be injected into the panel to make it vibrate.
The energy is quickly released/dissipated because with music we will not have a continuos input of energy, it will be bery transient. As soon as the energy in the panel falls below a certain amount, the resistance of the panel to vibrate at any but that very narrow bandwidth will quickly damp the resonance.
dave
I should like to add my tuppence worth to this.
I have read with interest that people think that MDF "sucks the life" out of a loudspeaker. I think that's a bit extreme but it certainly can add a cold/clinical edge to the proceedings. Hardwoods, Oak and Beech in particular seem to give a more natural sound. The only problem is that a hardwood will move according to humidity.
At Arcaydis, I designed a cabinet in solid American White Oak and sold 20 pairs to a Canadian distributor. Every single pair came back. Why? Not because there was anything wrong with the speakers. The average humidity in Canada is 6%. The Oak literally "dried out". I found a remedy by laminating 8mm Oak to 12mm MDF in a flat press. Then I got the cold clinical edge.
So here's my tuppence worth. Decide what thickness cabinet you want. Let's say 30mm. That's 12mm of hardwood flat pressed to, wait for it, 18mm double density chipboard. One tip. Put a layer of veneer on one side of the chipboard. This will remove the "bendy spoon" element.
Right. That's me done. Wish you all well.
Ricky.
I have read with interest that people think that MDF "sucks the life" out of a loudspeaker. I think that's a bit extreme but it certainly can add a cold/clinical edge to the proceedings. Hardwoods, Oak and Beech in particular seem to give a more natural sound. The only problem is that a hardwood will move according to humidity.
At Arcaydis, I designed a cabinet in solid American White Oak and sold 20 pairs to a Canadian distributor. Every single pair came back. Why? Not because there was anything wrong with the speakers. The average humidity in Canada is 6%. The Oak literally "dried out". I found a remedy by laminating 8mm Oak to 12mm MDF in a flat press. Then I got the cold clinical edge.
So here's my tuppence worth. Decide what thickness cabinet you want. Let's say 30mm. That's 12mm of hardwood flat pressed to, wait for it, 18mm double density chipboard. One tip. Put a layer of veneer on one side of the chipboard. This will remove the "bendy spoon" element.
Right. That's me done. Wish you all well.
Ricky.
rick57 said:
planet10 said:
The part of excitation that originates from the sound pressure inside the box is largely proportional to 1/f² if the radiated sound is flat. Besides the direct mechanical excitation from the driver, the sound pressure inside the box is the dominating source of excitation. The pressure outside the box is typically neglectible.
So pushing the a resonance upwards by two octaves, using the numbers above should decrease excitation by 24+4 =28 dB.
In the simulation below it can be seen that the pressure inside the box (dashed) is tilted by approximately -12 dB/oct (=1/f²) relative to sound pressure at 1m (solid).
An externally hosted image should be here but it was not working when we last tested it.
Hi!
Agree, but had some good result with with low density also!
Easier will be a layer of lacquer (and cheaper) instead of veneer on the back. The furniture industry use glued paper.
There exists a ratio of how thick layer of glue and paper it has to be to avoid warping in long term. Have to ask one of our customers about that.
John,
did you finish the two boxes, mdf/bb and measured them?
Cheers
Peter
Svante said:
Thanx for that, it puts a real "face" on things... as well one can expect a bit more attenuation if the panel resonances get up into the region where the air-space damping is effective.
Another thing it twigged is that one wants to avoid putting the panel resoance into the same place/frequency as a standing wave.
dave
PS: when John Atkinsom measures panel rsonances with his little device... what kind of device is that and where would one get one (or get something to repurpose)?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.