or Connect
AVS › AVS Forum › Audio › DIY Speakers and Subs › FoLLgoTT's build: 18 x Peerless XXLS12
New Posts  All Forums:Forum Nav:

FoLLgoTT's build: 18 x Peerless XXLS12 - Page 5

post #121 of 203
Sorry for the confusion I meant 14,000 watts bridged and thought you were running your amp bridged.
post #122 of 203
Quote:
Originally Posted by FoLLgoTT View Post

I agree that there is a strong gain, but I disagree that it is caused by a "pressurization" which can be explained by gas laws. I think the reasons for the gain are constructive interference and mirror sources on boundaries. That's my point.

OK, now I more clearly don't understand your point smile.gif

I assume you'd agree that in a perfectly sealed room with rigid boundaries that a sealed speaker driven with DC would pressurize the room; if not you're denying the gas law.

Any deviation from the ideal pressure vessel due to imperfect sealing, damping, and flexible boundaries just lessens the pressure increase.

So I'm not sure what it is you don't believe.

Perhaps that the gas law pressure doesn't equate to the P in SPL?

The conversion is here:

http://www.sengpielaudio.com/calculator-soundlevel.htm
post #123 of 203
if pressure vessel gain was based on reflections, there would be 1/4 wavelength cancellation back at the source. below the first (non zero) mode, there are no reflection cancellations because there are no reflections.
post #124 of 203
Quote:
Originally Posted by coctostan View Post

Excellent creativity Follo. That "passive shaker" setup is very interesting, especially to guys with rooms on concrete.

Here are some springs I found that might work: http://www.amazon.com/Mason-SLFH-A-125-Unhoused-Vibration-Deflection/dp/B006W0R97Y

I'm trying to wrap my head around how to better tune a setup like this so that it emphasizes the lowest frequencies and not the highest.

To do the number crunching, look up 'critically damped mass-spring system'. The problem will be one of a changing mass (depending on the amount of people on the riser), which will alter the resonant frequency, but good damping will help lessen the frequency-shifting effect, possibly why Follgott sees a near consistent resonant freq with varying weight. I think that with more weight, his springs act at an effectively higher spring constant, which could explain the increase in resonant freq with more weight, but that is all conjecture.

Ideally, you will choose a set of springs in their 'mid' position with the target weight on them, and select a spring constant which will tune the resonant freq in the middle of the freq band of choice, then damp it to avoid resonance/ringing, like a car with no shocks compare to a car with good shocks. Follgott has done a great job doing just that. It can get much more complicated with variable damping and variable rate springs, etc.

JSS
post #125 of 203
Quote:
Originally Posted by maxmercy View Post

The problem will be one of a changing mass (depending on the amount of people on the riser), which will alter the resonant frequency, but good damping will help lessen the frequency-shifting effect

It's simpler and results will be more consistent if the resonant frequency is below the operating range for all loading conditions.

Though that may not work out if that results in springs so soft that it's like getting into a boat.
post #126 of 203
Thread Starter 
Quote:
Originally Posted by noah katz View Post

OK, now I more clearly don't understand your point smile.gif

LOL tongue.gif
Quote:
I assume you'd agree that in a perfectly sealed room with rigid boundaries that a sealed speaker driven with DC would pressurize the room; if not you're denying the gas law.

I agree with that, BUT only when pressure has been balanced in the whole room.

When you feed a DC signal to a woofer, you do nothing else as creating a step response with a small surface in the room (the cone). This step respone contains all frequencies (including DC). It takes some time until the frequencies disappear and the DC component reaches a steady state in the whole room. It is just not true that all molecules move at the same time when the cone moves. This would imply that the speed of sound is infinity. There is a time of balancing where the pressure propagates and equalizes in the whole volume. And this is my argument why gas law can not be applied in our scenario. We have no steady state DC in a home theater.

I still think that the reason of the gain is constructive inteference only.

Quote:
Any deviation from the ideal pressure vessel due to imperfect sealing, damping, and flexible boundaries just lessens the pressure increase.

Flexible boundaries can be explained easily by the wave propagation model. They resonate and act as absorbers like any limp mass absorber or CBR/VBR does. And of course there is an amount of transmission loss.

With open windows it is a different thing. Many people have proven that it doesn't decrease the gain. Which is contrary to the "pressurization" theory.

Quote:
Originally Posted by LTD02 
if pressure vessel gain was based on reflections, there would be 1/4 wavelength cancellation back at the source. below the first (non zero) mode, there are no reflection cancellations because there are no reflections.

Yes and this is the reason why the gain increases with decreasing frequency. The cancelling parts are getting smaller and smaller. smile.gif

Quote:
Originally Posted by maxmercy 
Ideally, you will choose a set of springs in their 'mid' position with the target weight on them, and select a spring constant which will tune the resonant freq in the middle of the freq band of choice, then damp it to avoid resonance/ringing, like a car with no shocks compare to a car with good shocks. Follgott has done a great job doing just that. It can get much more complicated with variable damping and variable rate springs, etc.

This is why I choose to have 2 risers. One for each row. In the first row I have a weight range from 1 - 3 persons. And 1 - 4 persons in the second row. If there would be only one riser weight range would be 1 - 7 persons which is way too much for the springs. With one riser per row you can stay in the nominal range. smile.gif
post #127 of 203
"Yes and this is the reason why the gain increases with decreasing frequency. The cancelling parts are getting smaller and smaller."

seriously. a sub 1 meter from a wall experiences a 1/4 wavelength cancellation at about 340/4 hz, or 85hz.

if the room is 4 meters wide and the sub is placed on one wall, the 1/4 wavelength cancellation from the opposite wall would be at 340/16 hz, or 21hz.

yet no cancellation of the second type is observed. in fact, spl goes up, so the sign isn't even right. something else is at work.
post #128 of 203
Thread Starter 
Quote:
Originally Posted by LTD02 View Post

if the room is 4 meters wide and the sub is placed on one wall, the 1/4 wavelength cancellation from the opposite wall would be at 340/16 hz, or 21hz.

yet no cancellation of the second type is observed. in fact, spl goes up, so the sign isn't even right. something else is at work.

Of course it is there. You have to eliminate the other dimensions to see it. Please take a look at this simulation.



Wave theory works like a charm. smile.gif
post #129 of 203
i see the cancellation in your simulator, but i can't say that I've ever seen that in a room, so maybe the simulator is only good to a certain frequency.

the closed form wave equation that underlies the simulator probably doesn't work below the first non zero mode which is why it doesn't show any pressure vessel gain either.

this is more like what i would expect to see in that example (resonance/modal build up and high frequency cancellations, but no cancellation on the bottom end, rather a pressure vessel gain that keeps it more or less flat across the 20hz region):


Edited by LTD02 - 1/21/14 at 1:39am
post #130 of 203
Thread Starter 
Quote:
Originally Posted by LTD02 View Post

i see the cancellation in your simulator, but i can't say that I've ever seen that in a room, so maybe the simulator is only good to a certain frequency.

the closed form wave equation that underlies the simulator probably doesn't work below the first non zero mode which is why it doesn't show any pressure vessel gain either.

Sorry, but you are completely wrong on that. The simulator in REW works flawlessly and I can prove it!

I have seen this cancellation and I have the measurements from my experiments a few months ago. smile.gif

Subwoofers at 1/4 and 3/4 of width and 1/2 of height:


Measured frequency responses (2m, 3m and 4m distance from front wall):


Simulation of 2 m distance:


Simulation of 3 m distance:


Simulation of 4 m distance:



Apparently the measurements confirm the simulations. All peaks and dips are at the predicted frequencies. And there is only wave theory involved and no steady state gas laws.

Do you still believe that I'm wrong? smile.gif
post #131 of 203
Thread Starter 
Quote:
Originally Posted by LTD02 View Post

this is more like what i would expect to see in that example (resonance/modal build up and high frequency cancellations, but no cancellation on the bottom end, rather a pressure vessel gain that keeps it more or less flat across the 20hz region):


You forget the position of the microphone! This is the variable which decides at which frequencies the cancellation occur. And it is the most neglected variable in those considerations.
post #132 of 203
Quote:
Originally Posted by FoLLgoTT View Post

Of course it is there. You have to eliminate the other dimensions to see it. Please take a look at this simulation.



Wave theory works like a charm. smile.gif

In the next version REW has a checkbox "Room is sealed".

Room is sealed:



Room is NOT sealed:



It suggests that the low end is raised if the room is NOT sealed. REW makes no assumptions below 20Hz though.
post #133 of 203
Thread Starter 
@markus767
I don't understand your posting. Was it meant ironically?
post #134 of 203
^
Just posted how REW changes its prediction when the room is no longer considered as sealed.
post #135 of 203
Thread Starter 
@markus767
Are these screenshots from a private beta?

And why should the cancellation disappear when the room is not sealed? And how is "not sealed" defined? My room is not perfectly sealed and it clearly shows the cancellation as you can see in the post above.

This switch confuses me. Is there more information about that?
Edited by FoLLgoTT - 1/21/14 at 3:09am
post #136 of 203
i'm calculating the diagonal of your room as 8 meters. if pressure vessel gain sets in at around less than 1/2 lambda, that would correspond to lambda of 16 meters, and pressure vessel gain from ~21hz and down. the rew simulation and measurements don't really seem to speak to the issue.
post #137 of 203
Thread Starter 
Quote:
Originally Posted by LTD02 View Post

i'm calculating the diagonal of your room as 8 meters. if pressure vessel gain sets in at around less than 1/2 lambda, that would correspond to lambda of 16 meters, and pressure vessel gain from ~21hz and down. the rew simulation and measurements don't really seem to speak to the issue.

They disprove your statement that the cancellation is not visible in real rooms. And nothing more.

If you want to see the room gain in REW just make the room small enough to shift it in the visible range and there it is.

post #138 of 203
the 1 meter cube has a diagonal of 1.7 meters and a corresponding wavelength of 3.4 for pressure vessel gain, which suggests no reflection effects below 340/3.4 or about 100hz. i don't see any reflections below 100hz in the simulation.
post #139 of 203
Thread Starter 
Quote:
Originally Posted by LTD02 View Post

the 1 meter cube has a diagonal of 1.7 meters and a corresponding wavelength of 3.4 for pressure vessel gain, which suggests no reflection effects below 340/3.4 or about 100hz. i don't see any reflections below 100hz in the simulation.

Yes, because the wave model works down to the lowest frequencies. There is no need for explanations using the gas law. smile.gif
post #140 of 203
Concerning the argument on gas model vs boundary wave model, I think what is happening is a convergence of both models at the extreme and that they are both happening. Physics isn't always so simple that only one phenomenon is occurring at a time.
Quote:
Originally Posted by maxmercy View Post

To do the number crunching, look up 'critically damped mass-spring system'. The problem will be one of a changing mass (depending on the amount of people on the riser), which will alter the resonant frequency, but good damping will help lessen the frequency-shifting effect, possibly why Follgott sees a near consistent resonant freq with varying weight. I think that with more weight, his springs act at an effectively higher spring constant, which could explain the increase in resonant freq with more weight, but that is all conjecture.

Ideally, you will choose a set of springs in their 'mid' position with the target weight on them, and select a spring constant which will tune the resonant freq in the middle of the freq band of choice, then damp it to avoid resonance/ringing, like a car with no shocks compare to a car with good shocks. Follgott has done a great job doing just that. It can get much more complicated with variable damping and variable rate springs, etc.

JSS

Yep, it would work very similar to a car's suspension. In fact, it should be a spring and a damper to control the motion which obviously oscillates. The really advanced way to do it would be with an adaptive damper, but a properly tuned damper would suit as well. The spring simply needs to be sized to allow some compression with the minimal load while not binding upon maximum load. The damper is what keeps the platform controlled whether there is minimal or maximum load and should minimize variance between the two.

It is very similar to a car hitting a bump.
post #141 of 203
Quote:
Originally Posted by FoLLgoTT View Post

I still think that the reason of the gain is constructive inteference only.

OK, but I still don't see why you need to throw physics away because HT never reaches one of the end points (DC) of the continuum.
Quote:
Originally Posted by FoLLgoTT View Post

With open windows it is a different thing. Many people have proven that it doesn't decrease the gain. Which is contrary to the "pressurization" theory.

Their measurements had to have been flawed, because what is an open window is no different than a 100% effective absorber.
post #142 of 203
Quote:
Originally Posted by coctostan View Post

Concerning the argument on gas model vs boundary wave model, I think what is happening is a convergence of both models at the extreme and that they are both happening. Physics isn't always so simple that only one phenomenon is occurring at a time.

I don't think one denies the other; they agree at 0 Hz at which the solutions are identical, and the only solution for the gas law.
post #143 of 203
Very impressive correlation of simulation and measurements, FoLLgoTT.

I must say this recent discussion calls to mind a room full of capable and knowledgeable physicists faced with a dual-slit interferometer, arguing about whether light is photons or waves. smile.gif To me it matters not, it's about finding the model that is most useful for the situation. And if a few surprising things happen in the edge cases - well, I'd be surprised if they didn't. cool.gif
post #144 of 203
Quote:
Originally Posted by antisuck View Post

Very impressive correlation of simulation and measurements, FoLLgoTT.

I must say this recent discussion calls to mind a room full of capable and knowledgeable physicists faced with a dual-slit interferometer, arguing about whether light is photons or waves. smile.gif To me it matters not, it's about finding the model that is most useful for the situation. And if a few surprising things happen in the edge cases - well, I'd be surprised if they didn't. cool.gif

THAT'S A BINGO!
Is that the way you say it, 'That's a Bingo?'

You hit the nail right on the head. I frequent a surfboard forum where they have endlessly debated the true cause of 'lift'. Some say it is all Newtonian deflection, some say it is Bernoulli, some say Coanda.

Just like Light is Light, Lift is Lift, and Low Freq sound is Low Freq sound. There are various theories to think of/model the mechanism truly at play, but they are all theories. Use the best one that will allow you to make accurate predictions, with the least amount of effort. No need to use quantum mechanics when wave theory will suffice, or when ray theory will do just as well...

JSS
post #145 of 203
Surfers discuss physics? I'm having a tough time wrapping my head around that, dude.

fast-times-at-ridgemont-high-spicoli-surfer-dream.jpg
post #146 of 203
Quote:
Originally Posted by coctostan View Post

Surfers discuss physics? I'm having a tough time wrapping my head around that, dude.

I remember reading some articles many years ago on surfboard hydrodynamics in Surfer magazine. Did a google search and here they are, dated 1969!

Edit: Sorry for the OT, but this conversation jogged some very old memories. smile.gif
Edited by andyc56 - 1/21/14 at 4:01pm
post #147 of 203
Some of them do a thesis on it:

http://books.google.com/books/about/Hydrodynamics_of_Surfboard_Fins.html?id=C7y1YgEACAAJ

JSS

Man, sorry to be OT again...

JSS
post #148 of 203
Quote:
Originally Posted by noah katz View Post

Their measurements had to have been flawed, because what is an open window is no different than a 100% effective absorber.

The measurements aren't flawed.

Here's my room with 2 doors and 2 windows closed vs the same openings opened:



Nils,

Too much here to catch up on but a couple of things:

First, regarding your "English not so good" comment. Your use of English is commendably good, enough to embarrass some posters whose 1st language is English, IMO.

Yes, regarding pressure vessel gain/air tight room vs progressively constructive reflections/boundary transmission losses, all of your posts are spot on and backed by your (and everyone else's) posted evidences, data and observations.

There is no "extensive" data available, as LTD likes to say to mislead readers. He can't show a single piece of real data from an actual room measured on planet earth.

The speed of sound increases by 5X (or, as you noticed, to infinity), the rooms ambient barometric pressure increases, the sound becomes a gas, or water, the gain begins at SOS* 2X the the rooms longest dimension, "my subs pressurize my room", room gain is proportional to how sealed your room is against air leakage, etc.

All proven wrong by data.

The only thing we can be sure of if the room is perfectly sealed is that the listener will soon pass away. smile.gif

Also, I would think the decay plot might be radically different if the SOS increased by 5X <20 Hz?

Great stuff, this thread. I appreciate the hard work and your sharing it.
post #149 of 203
"Yes, because the wave model works down to the lowest frequencies. There is no need for explanations using the gas law."

having slept on it, i see that there isn't any disagreement here at all.

below the first non-zero mode, the closed form wave equation in a rigid enclosure gives a solution of 12db / oct rising response. for a direct radiator (e.g., a sealed subwoofer) in a 2 pi space, the response tends to rolloff at about 12db / oct and exactly at that rate when q is equal to 0.707. together, those combine to form the measured result of "flat" spl. again, below the first non-zero mode. this result is obtained because pressure is more or less rising and falling throughout the room at the same time.

on the other hand, the room pressurization theory asserts that changing the air pressure in a room will change the sound pressure in the room in a frequency independent manner, so long as the range of frequencies is below the critical frequency. the critical frequency being the one where a half wavelength is about equal to the diagonal of the room, i.e. half a wavelength. by modulating the pressure in the room, spl in the room can be determined by the empirically derived equation: spl = 197 + 20 log delta V - 20 log V [m^3, db peak].

both theories give the same result, so both are equally valid.

cheers.
post #150 of 203
Quote:
Originally Posted by bossobass View Post

The only thing we can be sure of if the room is perfectly sealed is that the listener will soon pass away. smile.gif

Thanks for the laugh. I needed it.
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: DIY Speakers and Subs
AVS › AVS Forum › Audio › DIY Speakers and Subs › FoLLgoTT's build: 18 x Peerless XXLS12