Slot-loaded Infinite Baffle - anyone care to throw sanity at this?

[willray]

Greetings diyers,

I'm not much for jumping into a community without adequate reading to be sure I'm not asking a dumb question, but, I find myself in a bit of a time crunch and so I need to ask what might be a dumb question:

Does anyone have thoughts on slot-loaded infinite baffle woofer/sub configurations? Specifically in vertically-opposed configurations?

I'm designing a new "workshop", and let's say I have a large luxury of space and design, and a much more constrained budget for experimentation/later modification.

I'm tremendously impressed with the results Nelson Pass reported with his slot-loaded open-baffle system, but it seems that while one of the advantages of the slot-loaded LF design is the increased front-radiated energy from the compression, the open-propogated rear wave still eats some of the design's lunch once one gets out to a distance of a couple baffle-widths and cancellation gets serious, plus the open-baffle design doesn't exactly look like a point-source radiator.

On the other hand, since I have quite a lot of space to work with, and can customize things pretty much however I like, it occurred to me that I could completely eliminate the back wave in a slot-loaded system, and much more closely approximate a point-source at the same time, by building slot-loaded opposing drivers, up-and-down-firing in floor-to-ceiling columns, with the speakers and a full-circumfrential slot at roughly listening height, and dumping the back wave out into the downstairs machine bay and attic.

It's not exactly a point-source radiator, more donut-ring-radiator, but it eliminates the push-pull configuration of front and back waves, provides nice infinite-baffle bass extension, and gives me the additional energy of the slot-loaded configuration - or so my understanding of the theory says.

Of course, this design would require built-in speaker columns and corresponding holes in the ceiling and floor, and I'll have to plan for that as we're building, so it's not exactly something I can wait and try just to see how it works.

So - any thoughts? Any experience (I can't imagine many people have cut holes in their floor and ceiling to try this ? Any reasons I either should, or should not, consider building this into the structure?

Many thanks for any advice!
WCR

 

[jplesset]

Your columns sound rather resonant. I, too, have columns, and decided to put small cabinets in them as simple sealed design. 4 10" inexpensive subs turned out to be plenty for my large living room. In cabinet, response down 2db at 30....

 

[Scott L]

Infinite Baffle Sub-woofer stystem

In my opinion, a true infinite baffle sub-woofer system just smokes anything else. In my set-up I use 8 Acoustic Elegance IB15's in a reaction-cancelling configuration. The slot opening up to the sound room is a tad bit greater in area than the total cone area of all eight woofers added together. The rear wave exists into a closet of about 600 cubic feet.
One picture is of the construction, the other when installed.

 

[willray]

Your columns sound rather resonant. I, too, have columns, and decided to put small cabinets in them as simple sealed design. 4 10" inexpensive subs turned out to be plenty for my large living room. In cabinet, response down 2db at 30....

You're thinking of the transmission-line-like containment of the back wave before dumping into the "infinite volume" space?

That's one of the conundrums of which I haven't been able to intuit the impact.

I'm figuring something in the 14 to 16" diameter air column, 4 to 5' long, which obviously puts some moving air mass behind the driver, but I can't figure out how to model it on paper.

 

[TBTL]

I'm tremendously impressed with the results Nelson Pass reported with his slot-loaded open-baffle system, but it seems that while one of the advantages of the slot-loaded LF design is the increased front-radiated energy from the compression [...]

Unfortunately, the slot loading does not increase the radiated energy. Sreten explains it nicely in Nelson's open baffle topic.

Hmmm .... Oh dear .... in reality practically inconsequential .....

You cannot just use pressure to indicate SPL, you need to normalise it
by multiplying it by the radiating area, otherwise you get nonsense
like small ports being more efficient than large ones, they are not.

Yes, the pressure is higher as the area is less, inevitable measured
nearfield. Farfield there is no difference, there is no efficiency gain
with the arrangement. Volume displacement is the same front and rear.

rgds, sreten.

However, slot loading could stil be benificial. The air mass in the slot is 'attached' to the woofers, increasing Qts and lowering fs. The smaller the slot area, the greater the effect.

 

[Nelson Pass]

Unfortunately, the slot loading does not increase the radiated energy. Sreten explains it nicely in Nelson's open baffle topic..

I don't think he did.

 

[john k...]

Stern has it correct but should have been clearer. SPL is proportional to volume velocity (velocity times area). At very low frequency, the reduced area of the slot opening has the effect of increasing the velocity of the air but with a proportional reduction in radiating area. The result is no net increase in the volume velocity. As the frequency rises the cavity resonance is encountered which does increase the volume velocity over that of the drivers mounted on a flat baffle. However, this is generally a fairly narrow band resonance with rather high Q. Higher yet in frequency and higher order resonances are encountered along with a basic low pass filter effect.

I investigated the slot loaded woofer when it first appeared using a scale model so good full range measurements could be obtained. http://musicanddesign.com/Nelson_Pass_Slot_loaded_OB.html

 

[speaker dave]

I don't think he did.

It looks to me like Sreten's initial comments were correct. Certainly John K's excellent analysis debunked any notions of efficiency gain from the system. (And with increased mass loading there will actually be an efficiency loss.)

Now I heard about this Karlson thing that is both a horn and a bass reflex cabinet....

 

[Rojoh]

Greetings diyers,

So - any thoughts? Any experience (I can't imagine many people have cut holes in their floor and ceiling to try this ? Any reasons I either should, or should not, consider building this into the structure?

Many thanks for any advice!
WCR

Since you're going for an IB-solution, your main worry should be enough (driver) displacement for your needs, enough space for airflow, driver orientation and position of the column(s) in the room. But going IB will give nice, clean and deep bass when done right, so you're on the right track IMO - given we're talking relatively low XO here.

I have cut the necessary holes in the floor, but went for an dual-opposed PP-configuration instead of a slot, but that's me

Home | "Cult of the Infinitely Baffled" Hear The Bass, Not The Box The definitive online resource for Infinite Baffle subwoofer design Established 1999

 

[willray]

Unfortunately, the slot loading does not increase the radiated energy. Sreten explains it nicely in Nelson's open baffle topic.
However, slot loading could stil be benificial. The air mass in the slot is 'attached' to the woofers, increasing Qts and lowering fs. The smaller the slot area, the greater the effect.

I apologize for questioning luminaries, but, something doesn't add up here. Entirely possible it's my adding that doesn't add up, but if someone cares to take pity and point me in the direction of my error, that'd be grand.

My acoustic physics is pretty much nonexistent, but while it looks clear that Sound Pressure Level is directly proportional to the Volume Velocity, it also seems clear that the Acoustic Energy is directly proportional to the square of the particle velocity.

Volume Velocity is directly proportional to the radiating area, and to the particle velocity.

Slot loading - let's say halving the radiating area just to make the math easier - decreases the radiating area and proportionally increases the particle velocity by the same amount, so (impedance changes due to slot effects aside), the Volume Velocity produced from a slot-loaded configuration should be identical to that produced by the plain driver.

Ok, so this says that the SPL available from the slot-loaded configuration shouldn't improve. I'm with you so far.

However, with a half-area slot, we're going to get the same mass of moving particles, at double the particle velocity - and since the Energy goes with the square of the particle velocity, the math says that the radiated _Energy_ has got to go up. Even if we say that the radiated energy needs to be normalized by the radiating surface area, the energy goes up as the square and down proportionally to the decrease in area, so it's still 2x the radiated energy from the slot.

I can't seem to make the math stop doing that - so - where am I going wrong, or, what's the additional radiated energy doing, that's not going into SPL?

 

[daqvin_carter]

Is your goal to just increase Qts and lower FS so response is flat?
You may find it easier to handle any response problems with the infinite baffle with an oddly shaped crossover or first order filtering.

I use a pair of low Qts, though low FS woofers in infinite baffle.
The woofers are some 18 inch pro woofers. FS is 25 HZ. Qts was near .3 I think.
Modeling as an infinite baffle enclosure I get a first order drop below 80 HZ due to the low Qts.

I just use a first order low pass starting at 25 HZ combined with a second order low pass at 80HZ.
Response works out well in this case.
I recommend infinite baffle venting into a ceiling or crawl space if you can do it. You just cannot beat the bass response and space saving.

One thing to think of is fire protection. If you vent into a ceiling, you want to protect the other side of the woofer. Type X drywall is a good choice. You just wany to keep a couch or christmas tree fire from spreading through the woofer and into an attic area.

 

[sreten]

The air mass in the slot is 'attached' to the woofers, increasing Qts
and lowering fs. The smaller the slot area, the greater the effect.

Hi, The latter is simply not true, for obvious reasons, rgds, sreten.

A U or H frame adds the most additional air mass, not a slot.

JohnK can do the maths, I'm more handwaving based on principles.

However I'm pretty sure a critically damped U frame drops Fs
and doesn't increase Qts by the amount predicted by simply
adding some cone mass to the physics model of a driver.

 

[sreten]

I don't think he did.

Hi,

I wasn't trying to explain the reality, just debunk the
myth that it does something that it simply cannot.

The fact that I didn't mention the slot cavity resonance
has nothing to do with the fact that it cannot do what
was claimed, and that needs no further explanation.

rgds, sreten.

In fact the slot cavity resonance is a major feature
of such an arrangement if utilised effectively in
the design, otherwise there is not much about
about slot loading that has any practical benefit.

 

[speaker dave]

However, with a half-area slot, we're going to get the same mass of moving particles, at double the particle velocity - and since the Energy goes with the square of the particle velocity, the math says that the radiated _Energy_ has got to go up. Even if we say that the radiated energy needs to be normalized by the radiating surface area, the energy goes up as the square and down proportionally to the decrease in area, so it's still 2x the radiated energy from the slot.

I can't seem to make the math stop doing that - so - where am I going wrong....

I think the difficulty is just in looking at it in energy terms. We really care more about radiated power than particle energy. I believe the radiation resistance drop with the smaller radiating area is counterbalancing your increase in particle energy.

Another thought experiment: say we send an acoustic wave down a pipe of constant cross section. If it subsequently narrows, does it pick up energy? As it expands does it lose energy? Where does it go?

We know a horn will increase efficiency as we can expand the wavefront and couple it to a larger area. Particle velocity may be smaller but apparently the increased radiation resistance trumps that.

Good question for the physics majors.

 

[willray]

Another thought experiment: say we send an acoustic wave down a pipe of constant cross section. If it subsequently narrows, does it pick up energy? As it expands does it lose energy? Where does it go?

Ah, I don't believe this is a good analog. In the pipe, the available energy of the wave is fixed. The particle velocity and mass of moving particles vary such that the energy remains (again, ignoring differing impedance effects) the same on both sides of the transition.

With a speaker, we're not /obliged/ to distribute energy identically on both sides of the driver. My acoustic physics is sufficiently weak as to not be sure whether the following is a good analog either, but a simple-kinetics demonstration of the kind of asymmetry that /could/ happen can be found in an equal-and-opposite force operating between two particles, pushing them apart - The light particle acquires more energy than the heavy particle. Again I can't begin to say there's an appropriate analytical parallel there, only that because of the V-squared term, energy /can/ do asymmetric things under appropriate circumstances.

A bit of reading, if I'm following correctly, suggests that the "Sound Intensity" term is what is changing in the slot-loaded configuration. I can't get my head around this term yet, or more specifically how it can change without sound pressure changing, what changing it actually does, and what that all means for a speaker.

How the impedance-matching properties of a horn mix with slot-loading properties, remains far beyond what I can begin to think coherently about at this point!

 

[Nelson Pass]

John, I was under the impression we had reached some accord back in post #261
of the Slot Loaded Open Baffle thread when you wrote:

"In any event, I have no problem with the observation that the OB system may
have greater on axis SPL response at low frequency than at 180 degrees. It
is a highly complex, asymmetric, 3D system. However, the observation that the
0 degree SPL is greater than the180 degree SPL does not lead to the conclusion
that the slot loading leads to greater efficiency at low frequency. If you want
to say the axial sensitivity of the system is greater at 0 degrees than at 180, fine."

I notice that elsewhere in that same post you said,

"Also regarding the baffle, the slot is like a flush mounted source but the rear is a group
of drivers mounted sideways and standing off the baffle."

This is apropos, as I later tested the response of some different slot arrangements and
saw that the response is very much dependent on the arrangement of the slot with regard
to its depth and direction.

For example, if you mount a woofer conventionally on a flat baffle and cover it with a board
in which you have cut a slot, you get very little difference in the response front/rear at low
frequencies. The air comes out faster, but the its vector is more spread out radially.

To maximize the difference, you need depth to the slot, which tends to align the momentum
of the air movement, focusing the energy forward. Too bad the extra pressure still declines
at a distance.

The mention of H frames is also appropriate, as it is another method of gathering up
the output of the woofer and encouraging it to go in a particular direction. Horns would be
the ultimate example - you can leave the rear chamber off of a good bass horn and hardly
know there's a rear wave. If you like you can think of these enclosures as very badly
designed horns.

It wouldn't be the first time anyone accused me of a building a bad horn.

In any case, these days I have settled on a design which floor-faces one large woofer, as
seen at the last two BAF's. It gives a consistent asymmetry difference from 30 to above
100 Hz and sounds very good. I think I'll build some and put the Definimax 15's and
Lowthers in them.

 

[speaker dave]

A bit of reading, if I'm following correctly, suggests that the "Sound Intensity" term is what is changing in the slot-loaded configuration. I can't get my head around this term yet, or more specifically how it can change without sound pressure changing, what changing it actually does, and what that all means for a speaker.

How the impedance-matching properties of a horn mix with slot-loading properties, remains far beyond what I can begin to think coherently about at this point!

That got me curious so I wiki-ed Sound Intensity and Sound power to see what the relationships were.

Intensity is simply acoustic Watts per unit of area. There is no unit for intensity as it is simply defined as Watts/meter squared.

This suggests a good way to look at our dilemma here. We know that if we force sound power down a path where the cross section varies (say the previous tube that gets larger and/or smaller) that the acoustic power must not change along the way. Power is neither added or subtracted.

That being the case then the Watts per meter squared can go up and down with area, as long as the Watts is constant. In other words the intensity varies with changing area (directly in proportion to area!) so that that the transmitted Watts can be constant down our pipe, as they must be.

Intensity is proportional to pressure squared (Intensity = p squared over acoustic impedance), so we get a 20 log relationship. This alone will account for the pressure change of forcing woofer power through a slot. That is, if we have 1/3rd the area then 20log(3) = 9.5dB, is the expected increase from slot loading our woofers. Power isn't changing if the area is constricted, so efficiency never really changes.

I think that all makes sense.
David S.

 

[speaker dave]

For example, if you mount a woofer conventionally on a flat baffle and cover it with a board
in which you have cut a slot, you get very little difference in the response front/rear at low
frequencies. The air comes out faster, but the its vector is more spread out radially.

To maximize the difference, you need depth to the slot, which tends to align the momentum
of the air movement, focusing the energy forward. Too bad the extra pressure still declines
at a distance.

The mention of H frames is also appropriate, as it is another method of gathering up
the output of the woofer and encouraging it to go in a particular direction.

I don't think this notion of "forward momentum" to the air flow is correct.

This is an AC wave rather than a DC stream that we are talking about. If anything, the narrow slot will always broaden high frequency directivity. Whether we consider BBC monitors with slots to widen woofer directivity or constant directivity horns with diffraction slots to widen directivity, neither case gives "forward momentum" that increases axial pressure.

David S.

 

[Nelson Pass]

You may have a point.

I guess it might depend on the result of the experiment that Feynman once
tried, seeing if when you reverse the pressure on a rotating sprinkler it would
still rotate but in the opposite direction. Unfortunately he broke a lot of lab
glassware trying.

 

[phivates]

Well as an ignorant layperson I've never gotten Newton's 3rd Law. Nothing should ever happen if an equal and opposite reaction always occurs, right? Unless time is involved, and I'm guessing that for us humans time is always involved.