Acoustic Suspension vs. Bass Reflex
Rich Peachavanish
Can anyone explain the difference/tradeoff between these two type of
speaker design?
Also, what do you think about Boston Acoustic speaker? (sound,
reputation, etc...)
Thank you.
busenitz
All else being the same, Acoustic suspension speakers usually
are less efficient, but often have better transients, and a smoother,
more gradual rolloff on the low end. Bass reflex are more efficient,
and have a peak at the resonance where the box is tuned at, and have
a steeper rolloff below that point.
Acoustic suspension speakers usually (again, drivers being the
same) have better performance below 30 or 40 Hz, but not as good from
40 to 80 Hz, since below the tuning frequency of a bass reflex cabinet
the driver begins to act as if it is not enclosed at all.
I am not sure about this, but I think that the reason most
bass reflex boxes are not ported at 20 to 30 Hz is because this would
make the sound bland, or lack of warmth in the midbass region.
Remember, this is all a generalization.
Richard D Pierce
>In article <2dh401$a...@news.u.washington.edu> ri...@stein1.u.washington.edu (Rich Peachavanish) writes:
>>
>>Can anyone explain the difference/tradeoff between these two type of
>>speaker design?
> Remember, this is all a generalization.
Yes, and as such, much of it is simply wrong. Take for example, your
generalization:
> All else being the same, Acoustic suspension speakers usually
>are less efficient, but often have better transients, and a smoother,
>more gradual rolloff on the low end. Bass reflex are more efficient,
>and have a peak at the resonance where the box is tuned at, and have
>a steeper rolloff below that point.
Several incorrect points:
1. If, as you say below, the drivers in each are the same, then
the efficiency of a bass reflex and acoustic suspension system
will be indentical. The efficiency is determined by the driver,
not by the box you put it in.
2. Secondly, you assertion that bass reflex system have a peak at
resonance is similarily wrong. Reflex systems can and have been
designed that have had flat bandpass characteristics with no
peak whatsoever at resonance. And, interestingly enough, this is
using a design where the enclosure resonance is at maximum Q
(so-called lossless B4 alignments).
3. An acoustic suspension system with a Q of 1.5 or so (just like
the older AR models, such as the AR-3a) has provably worse transient
response and faster rolloff rates at cutoff than a bass reflex
system using lowQ QB3 alignments for a similar cutoff frequency.
Thus, as an apriori statement, acoustic suspension systems having
better transient performance and lower roloff rates is simply not
supported by the facts.
> Acoustic suspension speakers usually (again, drivers being the
>same) have better performance below 30 or 40 Hz, but not as good from
>40 to 80 Hz, since below the tuning frequency of a bass reflex cabinet
>the driver begins to act as if it is not enclosed at all.
But a properly designed reflex system and a properly designed acoustic
suspension system WILL NOT use the same driver in most cases, so your
comparison is, for the most part, invalid.
Your staements about "better" performance in the region of 30 to 40 Hz,
or 40 to 80 Hz or whatever, is completely unqualified by what you mean by
"better".
Your assertion about the action of the driver below the enclosure
resonance, while absolutely true, fails to address one of the great
advantages of reflex system in that around the region of enclosure
resonance and for a substantial region abovbe and below, cone excursion
is significantly reduced increasing power handling capability and
reducing distortion dramatically. Below that point, and below system
resonance in an acoustic suspension system, the driver is working
essentially below it's mechanical resonance, and for drivers of similar
mechanical characteristics, the actual cone excursion and thus excursion
limited capability, they're damned near identical.
> I am not sure about this, but I think that the reason most
>bass reflex boxes are not ported at 20 to 30 Hz is because this would
>make the sound bland, or lack of warmth in the midbass region.
Yes, I agree with you, you are not sure about it.
The reason that bass reflex cabinets are not tuned (if that's what you
mean by "ported") at 20 to 30 Hz has absolutely nothing whatsoever to do
with blandness of sound or lack of warmth in the midbass. The tuning
frequency of a bass reflex system is dependent upon the system alignment,
and that is determined by, essentially, the low frequency parameters, the
so called "Thiele-Small" parameters of the driver. A given driver has a
given set of parameters that are determined by things such as cone mass,
suspension compliance, mechancial and electrical losses and so on. These
parameters determine essentially the useful range of tuning for the
system.
As a generalization (one backed up with statistics from several thousand
drivers), one can say that most 6 1/2" drivers suitable for reflex use
work best with enclosure tunings between 40 and 65 Hz. Most 8" driver,
around 30 to 45 Hz, 10" around 28 to 40 Hz, 12" 25 to 35 Hz, and so on.
To be suitable for use in a design that is tuned around 20 Hz, the driver
has to have a free air resonance equal to or less than 20 Hz (systems with
Fb's less than Fs tend to be relatively poorly designed), a driver total Q
of about 0.4 or less, and so on. This essentially means large drivers with
heavy cones, high compliances and large magnets. These drivers are more
expensive and thus are sold less frequenctly than cheaper drivers. Thus
the reason most systems are not tuned so low is one of pure and simple
economic statistics driven by physical reality.
Warmth and blandness play no part.
--
| Dick Pierce |
| Loudspeaker and Software Consulting |
| 17 Sartelle Street Pepperell, MA 01463 |
| (508) 433-9183 (Voice and FAX) |
Richard D Pierce
>
>Can anyone explain the difference/tradeoff between these two type of
>speaker design?
Without getting into the theory, and avoiding a lot of the common bunk,
the difference is, conceptually, very simple (paraphrasing Richard Small)
The main difference between a sealed box and a bass reflex is the
presence of a second aperture in the cabinet which contributes to
the overall output of the system at low frequencies.
That is the entire, sole, complete and only difference between the two.
Period.
Now, that difference leads to some difference in the way it works.
THe action of this secondary aperture is twofold. First, as I said above,
it contributes to the output of the system at low frequencies. Secondly,
it, along with the air in the enclosure, forms a second resonant system
(the first is formed with the mass of the woofer cone and the
conbination of the suspension and the air in the box).
While it might seem intuitively obvious that if the woofer cone moves in,
then the air in the aperture (or port) must move out and, thus, the two
must cancel. And this is absolutely true at veryu low frequencies, like
the rate at which you can push the cone with your hand.
But a very different thing happens as you approach the frequency at which
the cabinet is tuned (the enclosure resonance). Several things of interest
happen. First, motion of the air in the port and the woofer cone start to
move in phase. Second, and very importantly, the motion of the cone
starts to decrease, to the point that at resonance, THE WOOFER IS BARELY
MOVING, IF AT ALL! (emphasised because this is very important). AT the
enclosure resonance, it is the port alone, and not the woofer, that is
porducing the majority of the output of the total system.
Above this resonant frequency, the output of the port reduces quickly
with frequency (at about 12 dB per octave and starts to move out of pahse
with the woofer, but the woofer's direct output completely overwhelms
it). Below this frequency, the output of the port AND the woofer reduces
quickly with freqeuncy, and they are also moving out of phase with one
another.
But the big difference happens at the frequency where the enclosure is
resonating: By reducing the motion of the cone and replacing it, we gain
several advantages:
1. reduced cone motion means lower distortion. It's far easier to get
large amounts of linear air motion out of a port at low frequencies
than out t of a woofer. The mechanical susspension on the cone is
far from linear. Acoustic suspension systems tries to solve this
problem by making the air in the box the suspension. Bass reflex
solves this problem by simply reducing the demands on the woofer.
2. Reduced cone motion means significantly higher output capability
for a given set of mechanical limits in a woofer. This advantage
can be as much as 20 dB or more at the enclosure resonance, and can
be substantial over as much as an octave at very low frequencies,
just where the demands on the driver are their greatest.
3. The output of the port can be used, in effect, to extend the low-
frequency cutoff lower. For a given system efficiency and cabinet
size, this can be about 1/3 octave lower, or you could get twice the
efficiency for a given cutoff frequency or cabinet size.
The addition of the extra complexity buys you more degrees of freedom in
shoosing design tradeoffs, on the assumption you have the ability to pick
and choose parameters to suit your needs.
As alluded to above, there is an advantage in something called the
efficiency-bandwidth-enclosure size product that bass reflex system have
over sealed box system. This advantage is as much as a factor of two.
This means:
1. A bass reflex system of a given size and efficiency can go lower
in frequency than a selaed box system of similar size and frequency.
or,
2. A bass reflex system of a given size and cutoff frequency can be
up to twice as efficient as a similar sealed box system.
or,
3. A bass reflex system of a given efficiency and cutoff frequency can
as small as half the size of a simialr sealed box system
Well, this seems like a big pile of advantages, one might ask what the
disadvantages are:
1. More complex and, thus more expensive cabinets and design (but not
very much
2. As many other things being equal as possible, bass reflex drivers
tend to require larger and thus more expensive magnet systems.
3. The performance of reflex systems is more sensitive to variations in
driver parameters, this requires either better materials or better QC
on the driver source, and this can be (but need not be) more
expensive. In an acoustic suspension system, as long as the air
in the box predminates the system stiffness, than any driver will work,
and this is a significant cost advantage to sealed boxes.
4. At high frequencies, some extraneous noise might emanate from the
port and interfere with the sound.
5. WIth it's reduced degrees of design freedom, a sealed box system
requires less expertise to design and have work credibly well.. A
reflex system, being inherently more complex, can be screwed up
easily and obviously by any idiot. Thus if you were to throw a group
of inexperienced engineers into a room and have them design 10
sealed box and 10 reflex systems, the statistics would demand that
they produce more bad reflex systems than sealed box system. This,
however, is no excuse to award any prizes to those that couldn't
misdesign the sealed box systems.
It is hoped that this information van help you ut through some of the
bunk and rhetoric that will inevitably arise from people posting their
thoughts and theories on things.
Lon Stowell
>
>Your assertion about the action of the driver below the enclosure
>resonance, while absolutely true, fails to address one of the great
>advantages of reflex system in that around the region of enclosure
>resonance and for a substantial region abovbe and below, cone excursion
>is significantly reduced increasing power handling capability and
>reducing distortion dramatically. Below that point, and below system
>resonance in an acoustic suspension system, the driver is working
>essentially below it's mechanical resonance, and for drivers of similar
>mechanical characteristics, the actual cone excursion and thus excursion
>limited capability, they're damned near identical.
>
I hate to ask, but could you go into this a bit deeper?
I may have been laboring under what may be a false understanding.
If you drive a typical bass reflex below the driver/cabinet
resonance with SIGNIFICANT ( as in massive bass boost ) energy,
doesn't the driver's excursion become excessive?
The reason I ask is that I've understood a quick and dirty
way to get some of the benefits of a subwoofer is to use a
parametric EQ to add 6-8 dB boost centered at 20 Hz with a
Q of about 3. I've been told that this works best with
acoustic suspension type speakers with cones in the 8-10 inch
range...and that you may damage reflex style speakers at louder
levels.
Have Roy Allison and NAD been funning us? >:-) Or is the
state of the art in reflex design such that you can drive them well
below their useful range without risking nasty rattling noises?
Obviously I'm not a speaker designer, I just listen to the things
at fairly high levels. But perhaps I've been giving misinformed
advice??
Richard D Pierce
> I hate to ask, but could you go into this a bit deeper?
>
> I may have been laboring under what may be a false understanding.
>
> If you drive a typical bass reflex below the driver/cabinet
> resonance with SIGNIFICANT ( as in massive bass boost ) energy,
> doesn't the driver's excursion become excessive?
The function of excursion vs. frequency for constant voltage input
significantly below cutoff for sealed and vented boxes is nearly the
same, with the advantage almost always goign to vented boxes if there is
one. However, the rolloff rate for the acoustical output of reflex
systems is twice that (24 dB/oct) of sealed systems (12 dB/oct). Thus,
far below cutoff, more boost is required to get the same acoustical
output. That can mean MUCH more electrical input so, indeed, you are now
forcing the reflex beyond its mechanical limits because you are driving
it harder.
> The reason I ask is that I've understood a quick and dirty
> way to get some of the benefits of a subwoofer is to use a
> parametric EQ to add 6-8 dB boost centered at 20 Hz with a
> Q of about 3. I've been told that this works best with
> acoustic suspension type speakers with cones in the 8-10 inch
> range...and that you may damage reflex style speakers at louder
> levels.
Equalized system work best when the entire system is designed as a whole.
For example, there are some wonderful 6th order alignments based on 4th
order vented systems and 2nd order boost EQ's that exhibit flat response,
good damping and amazing power handling in relatively small packages.
No, you cannot arbitrarily add EQ to any system and expect it to work.
For example, the common system designed 4th order systems based on sealed
enclosures uses a speaker system with system Q's well below the maximally
flat Qt = 0.707 (some as low as 0.5) plus boost EQ to give a maximally
flat 4th order response that is placed just about halfway bewteen sealed
boxes and reflexes on the efficiency-bandwidth-volume curve.
> Have Roy Allison and NAD been funning us? >:-) Or is the
> state of the art in reflex design such that you can drive them well
> below their useful range without risking nasty rattling noises?
Well, maybe not funning us, maybe indavertantly basing their conclusions
on old information and understandings. Remember that the hey-day of
acoustic suspension in this sountry was in the '60's, when reflex design
was scoffed at and the works of Messrs Thiele and Small had yet to be
made available in this country.
busenitz
Thanks for the corrections, and I think I've understood
most of what you said. However, I seem to remember hearing of
a number of drivers that were appropriate for both types of enclosures,
and not optimized for either one. I think that this might compromise
some aspect of performance, but they could do reasonably well in
either box. Am I totally off on this?
Richard D Pierce
I had posted an article about 3 months ago on this specific topic. If I
can find it, I'll try to repost. It gave a couple of worked-out examples
given some real-world drivers. The poiunt was that there is a class of
drivers that can be made to work reasonably well in both.
Jim Roth
In article <2dipa3$e...@pyrnova.mis.pyramid.com>, lsto...@pyrnova.mis.pyramid.com (Lon Stowell) writes...
>>Your assertion about the action of the driver below the enclosure
>>resonance, while absolutely true, fails to address one of the great
>>advantages of reflex system in that around the region of enclosure
>>resonance and for a substantial region abovbe and below, cone excursion
>>is significantly reduced increasing power handling capability and
>>reducing distortion dramatically. Below that point, and below system
>>resonance in an acoustic suspension system, the driver is working
>>essentially below it's mechanical resonance, and for drivers of similar
>>mechanical characteristics, the actual cone excursion and thus excursion
>>limited capability, they're damned near identical.
> I hate to ask, but could you go into this a bit deeper?
> I may have been laboring under what may be a false understanding.
> If you drive a typical bass reflex below the driver/cabinet
> resonance with SIGNIFICANT ( as in massive bass boost ) energy,
> doesn't the driver's excursion become excessive?
Not if the driver compliances are the same. In other words, what
limits driver displacement at low frequencies is the same in either
a sealed or reflex enclosure.
But a reflex enclosure has a 24 db/octave acoustic rolloff below cutoff
while a sealed enclosure has a 12 db/octave acoustic roloff.
Equalizers like the Allison electronic subwoofer are
what are known as biquadratic equalizers; this circuit replaces
one pole pair of the acoustic transfer function of a sealed
enclosure with another at a lower cutoff frequency, and is somewhat
practical since the rolloff of a sealed enclosure is a more modest
12 db/octave. To obtain any significant bandwidth extension below
cutoff of a reflex enclosure you'd need "twice" the equalization
and this is impractical for any useful bass extension. it would be
better to modify the design of the system to incorporate the equalization,
such as use of a 6'th order Chebyshev design or some such.
You may be able to get sort of an equalized response with a regular
equalizer, but it is much better to use a circuit designed for
the purpose, which can be implemented with a pair of opamps per
channel, including buffers. The equalization is minimum phase, so
no extraneous ringing will result this way.
- Jim
Chris Christensen
>>I don't think this is in the FAQ but:
>>Can anyone explain the difference/tradeoff between these two type of
>>speaker design?
They are two different designs. If both are designed with the same
goal in mind the differences will be minimal. As noted in the
previous posters response the sealed box design will have a cutoff
slope that is less severe than the bass reflex. This _may_ result is
a slightly noticable deeper low end. But that's where the difference
ends.......
There is nothing better than a no compromise bass reflex design for
provding really good bass. IMHO!
>>Also, what do you think about Boston Acoustic speaker? (sound,
>>reputation, etc...)
As good as any other.......
And now for this person.......
> All else being the same, Acoustic suspension speakers usually
>are less efficient, but often have better transients, and a smoother,
>more gradual rolloff on the low end.
I can live with all of the above comments except for the "better
transients" part. This is speculation. Both designs, if done
correctly will sound very close to the same!
>Bass reflex are more efficient,
>and have a peak at the resonance where the box is tuned at, and have
>a steeper rolloff below that point.
Yes, wrong and yes. Not _all_ bass reflex designs will have a peak at
resonance. Some do and some don't. It is based on the design
chosen. Personally I perfer the maximally flat designs. I do have one
design that does have a peak in the pass band though.....
> Acoustic suspension speakers usually (again, drivers being the
>same) have better performance below 30 or 40 Hz, but not as good from
>40 to 80 Hz, since below the tuning frequency of a bass reflex cabinet
>the driver begins to act as if it is not enclosed at all.
Wrong, huh? and yes. What's your point?
If a bass reflex design is intended to perform down to 30 Hz it will
have more output that will a similarly designed sealed system.
Comment #2 implies that all BR designs are tuned somewhere between
40-80 Hz....? And the output of a bass reflex design does fall off
rapidly below the cabinets tuning point. But what if that point is at
24Hz?
> I am not sure about this, but I think that the reason most
>bass reflex boxes are not ported at 20 to 30 Hz is because this would
>make the sound bland, or lack of warmth in the midbass region.
While the LF tuning of the BR's will effect the total apparent
(audible) bandwith of the speaker, this generalization is untrue!
> Remember, this is all a generalization.
Generalization, speculation and perception. All within your personal
opinion and well within your rights to express.......
--
D.R. Chris Christensen Grass Valley Group (the day job)
chr...@gold.gvg.tek.com P.O. Box 1114 mail Stop N32B
916-478-3419 FAX 916-478-3887 Grass Valley, CA 95945
Neither I nor my employer is responsible for anything I say or do.
Brock Hannibal
|>
|> 3. The performance of reflex systems is more sensitive to variations in
|> driver parameters, this requires either better materials or better QC
|> on the driver source, and this can be (but need not be) more
|> expensive. In an acoustic suspension system, as long as the air
|> in the box predminates the system stiffness, than any driver will work,
|> and this is a significant cost advantage to sealed boxes.
|>
|> 4. At high frequencies, some extraneous noise might emanate from the
|> port and interfere with the sound.
|>
|> 5. WIth it's reduced degrees of design freedom, a sealed box system
|> requires less expertise to design and have work credibly well.. A
|> reflex system, being inherently more complex, can be screwed up
|> easily and obviously by any idiot. Thus if you were to throw a group
|> of inexperienced engineers into a room and have them design 10
|> sealed box and 10 reflex systems, the statistics would demand that
|> they produce more bad reflex systems than sealed box system. This,
|> however, is no excuse to award any prizes to those that couldn't
|> misdesign the sealed box systems.
|>
>It is hoped that this information van help you ut through some of the
>bunk and rhetoric that will inevitably arise from people posting their
>thoughts and theories on things.
>
Does this mean I can just replace the 10-inch bass drivers in the sealed
box bottom of my old Dahlquist DQ 10's without screwing them up too badly?
If so how can I determine the driver specs which will get me close enough
to the original performance? The reason I ask this is that I hate to dump
these old speakers but the woofers have "foam rot" and Dahlquist wants way
too much money to re-cone them (over $100 apiece + shipping). I want to
continue using these speakers as a secondary system in the party room. They're
too large and ugly( according to my wife) to be in our more formal living
room. Oh, is it possible that the bass performance might be better than it
was originally? I am an electronic engineer not a speaker designer so I need
help with this project.
Brock Hannibal
Design Engineer
Tektronix, Inc.
Richard D Pierce
>
>Does this mean I can just replace the 10-inch bass drivers in the sealed
>box bottom of my old Dahlquist DQ 10's without screwing them up too badly?
It means that you don't have as much to worry about, as far as low
frequency parameters are concerned. However, the efficiency of the driver
and it's high end characteristics may well be critical, so simply dealing
with the low end is only part of your problem.
>If so how can I determine the driver specs which will get me close enough
>to the original performance?
Well, if as you say below, the surrounds are rotted, you won't be able to
learn much. But the parameters you need to know are the basic
"Thiele/Small" parameters, resonant frequency Fs, equivalent volume of
compliance Vas, mechanical, electrical and total Q, and DC resistance.
>The reason I ask this is that I hate to dump
>these old speakers but the woofers have "foam rot" and Dahlquist wants way
>too much money to re-cone them (over $100 apiece + shipping).
Gawd, I can't believe the number of people who are getting these
ridiculous prices for reconing and driver replacement!
You might want to consider buy a surround replacement kit and doing it
yourself.
>Oh, is it possible that the bass performance might be better than it
>was originally?
The only way this would be possible (if by better you mean lower cutoff)
is by using a woofer that has greater cone mass and the same
efficiency as the one you know use. Using a driver with higher compliance
will; not have any appreciable effect because it is likely that the
cabinet is the dominating compliance in the system. A higher mass will
result in a lower system resonance, though. Other factors will need to be
taken into acount, such as sytem Q and so on.
>I am an electronic engineer not a speaker designer so I need
>help with this project.
Don't feel bad, there are plenty of speaker engineers who would need the
help also.
Henry Pasternack
>Gawd, I can't believe the number of people who are getting these
>ridiculous prices for reconing and driver replacement!
I don't know... If the manufacturer pays $10 for a driver, it takes
ten minutes per speaker to do the swap, ten minutes to pack and unpack,
ten minutes to fill out shipping forms and screw around with order
tracking, plus he may feel obliged to make sure the speaker works after
he repairs it, and then he may get sued because the customer suddenly
"discovers" a flaw after the repair. And if he sends out raw drivers,
he has to contend with customers doing all kinds of unimaginable awful
things, like shorting out the wires and blowing an amplifier... And
all this to keep some cheap slob from buying *new* speakers for another
ten years (present company excepted)... Sounds reasonable to me.
If I were a manufacturer, I'd sell an authorized dealer a replacement
driver at my cost plus 60% and tell him to handle all of the labor and
warranty issues on his own... And no direct sales to customers...
-Henry
Jay Mitchell
LS. In article <CHD7p...@world.std.com> DPi...@world.std.com (Richard D Pier
LS. >
LS. >Your assertion about the action of the driver below the enclosure
LS. >resonance, while absolutely true, fails to address one of the great
LS. >advantages of reflex system in that around the region of enclosure
LS. >resonance and for a substantial region abovbe and below, cone excursion
LS. >is significantly reduced increasing power handling capability and
LS. >reducing distortion dramatically. Below that point, and below system
LS. >resonance in an acoustic suspension system, the driver is working
LS. >essentially below it's mechanical resonance, and for drivers of similar
LS. >mechanical characteristics, the actual cone excursion and thus excursion
LS. >limited capability, they're damned near identical.
LS. >
LS.
LS. I hate to ask, but could you go into this a bit deeper?
LS.
LS. I may have been laboring under what may be a false understanding.
LS.
LS. If you drive a typical bass reflex below the driver/cabinet
LS. resonance with SIGNIFICANT ( as in massive bass boost ) energy,
LS. doesn't the driver's excursion become excessive?
LS.
LS. The reason I ask is that I've understood a quick and dirty
LS. way to get some of the benefits of a subwoofer is to use a
LS. parametric EQ to add 6-8 dB boost centered at 20 Hz with a
LS. Q of about 3. I've been told that this works best with
LS. acoustic suspension type speakers with cones in the 8-10 inch
LS. range...and that you may damage reflex style speakers at louder
LS. levels.
LS.
LS. Have Roy Allison and NAD been funning us? >:-) Or is the
LS. state of the art in reflex design such that you can drive them well
LS. below their useful range without risking nasty rattling noises?
LS.
LS. Obviously I'm not a speaker designer, I just listen to the things
LS. at fairly high levels. But perhaps I've been giving misinformed
LS. advice??
Both Mr. Pierce and Mr. Lowell make valid points. An acoustic
suspension woofer has two slight advantages over a reflex enclosure when
operated below system resonance:
1. The enclosure adds to the spring constant of the woofer, although this
addition also serves to increase the system resonant frequency.
2. Any radiation from the rear of the woofer cone is contained in the
enclosure.
The above notwithstanding, it is still true, as Mr. Pierce asserts, that
two identical woofers will undergo substantially identical displacements
when used in reflex or sealed enclosures and driven with identical
electrical signals well below system resonance. The reflex system will,
however, produce significantly less acoustic output under this condition,
as radiation from the port will serve to cancel that from the front
of the woofer cone. The "quick and dirty" equalization method referred to
above is of no benefit, even when used with an acoustic suspension
enclosure: electrical boost applied at frequencies so far below system
resonance (typically 40-80 Hz in the case of the small woofers referenced
above) will increase distortion in a big way, with a relatively miniscule
increase in linear acoustic output.
---
. JABBER v1.2 . Make it as simple as possible, but no simpler.
j...@cmkrnl.com
> Does this mean I can just replace the 10-inch bass drivers in the sealed
> box bottom of my old Dahlquist DQ 10's without screwing them up too badly?
> Oh, is it possible that the bass performance might be better than it
> was originally?
Good bass performance was never a strong suit of the DQ-10's to begin with,
at least not in the early versions.
--- Jamie Hanrahan, Kernel Mode Systems, San Diego CA
Internet: j...@cmkrnl.com (JH645) Uucp: uunet!cmkrnl!jeh CIS: 74140,2055
Lon Stowell
>1. The enclosure adds to the spring constant of the woofer, although this
>addition also serves to increase the system resonant frequency.
Part of the unasked question was which type of system is more
likely to suffer damage when EQ is used to try to extend woofer
response downward. I've been told that the reflex systems
are more likely in general compared to sealed boxes.
I will exclude Allison's Electronic Subwoofer from this if it is
used with Allison's recommended models of Allison speakers. One
would hope Roy knows his products.
I do consider the Bass EQ on the NAD/Proton receivers in the
question. I am still on somebody's list for tearing the
woofer cone on a Voice of the Theater while playing with Digital
Domain.
>of the woofer cone. The "quick and dirty" equalization method referred to
>above is of no benefit, even when used with an acoustic suspension
>enclosure: electrical boost applied at frequencies so far below system
>resonance (typically 40-80 Hz in the case of the small woofers referenced
>above) will increase distortion in a big way, with a relatively miniscule
>increase in linear acoustic output.
Odd, I was under the understanding that 8-10 woofers in sealed
boxes had a lower resonance than this. Guess I'll just have to go
see if I can look it up somewhere. RTFM, Bummer.
Richard D Pierce
> Part of the unasked question was which type of system is more
> likely to suffer damage when EQ is used to try to extend woofer
> response downward. I've been told that the reflex systems
> are more likely in general compared to sealed boxes.
Reflex system ARE more likely to be damaged than a sealed box system
under these conditions under some conditions.
Take, for example, two systems, one a reflex and one a sealed box. We
will presume that both are Butterworth alignments (maximally flat, -3 dB
at system cutoff), both have the same cutoff frequency, and, thus, the
sealed box will roll of at 12 dB per octave and the reflex at 24 dB per
octave below that point. One octave below cutoff, the sealed box will be
12 dB down (approximately) and the reflex will be 24 dB down.
Now, let's say we want to extend the cutoff point of both down one
octave. Obviously the reflex will require somewhere around 24 dB of
boost one octave down and the sealed but 12 dB of boost.
Now, the excursion vs frequency in the unequalized case for both systems
below cutoff is ROUGHLY frequency independent (in fact, in the region
around cutoff, the reflex has a termendous advantage over the sealed box,
because the port takes over at enclosure resonance and causes woofer
excursion to approach zero at that point, but then it returns back to
roughly the same level below that).
And the last of our premises is that at a given frequency, excursion is
proportional to frequency.
A conclusion we can then draw from this basis is that if your intent is
to provide the same level of performance below cutoff, the reflex system
WILL suffer extreme excursions. Look at the case here, the boost at one
octave below cutoff is +12 dB for a reflex system, that's equivalent to
about a quadrupling of excursion, whereas the reflex requires a boost of
+24 dB, equivalent to 16 times as much excursion!
So, yes, in the narrowly defined scenario we've laid out, you're right in
assuming a greater chance of damage.
But the problem is that these equializers are general purpose thingies,
which usually make things mnuch worse than better, just for the reasons
cited above.
If you had more control over things, such as the ability to design the
speaker and equalizer in unison, then it's a different story. Examples
such as the system developed by EV in the mid '70's which was an
electronically equalized passive radiator system with the speaker and
equalizer sold together demonstrated that it was possible to build
higher-order systems that had excellent performance in an equalized vented
box system. That means high efficiency, low cutoff, small box volume and
high power handling.
The efficiency-bandwidth-box volume product for maximally-flat lossless
reflex system is twice that for the BEST CASE selaed box (Qt = 1.25 with
about a 3 dB peak at resonance). The same quantity for a maximally flat
equalized 6th order reflex system is a factor of two better still. And
similar relations hold for the maximum output parameters as well.
That means that for a given efficieny and cutoff frequency, a 6th order
system can be 1/4 the box volume of the BEST sealed box system.
Conversely, given the same cutoff frequency and box volume, this same 6th
order system can be 6 dB more efficient. And, finally, assuming same
volume and efficiency, a 6th order system is capable of 2/3 octave lower
cutoff frequency. This, for example, would make the difference between a
40 Hz and 25 Hz cutoff frequency. Even at 12 dB/octave vs 36 dB/octave
rolloff rate difference the 6th order system will still have output
advantage from below 20 Hz and above (reflex is -12 dB @20 Hz vs the
sealed box at -24 dB @ 20 Hz).
> I will exclude Allison's Electronic Subwoofer from this if it is
> used with Allison's recommended models of Allison speakers. One
> would hope Roy knows his products.
Indeed, I do have enough faith in Roy Allison that I am reasonably
confident he understands both his products AND the relevant physics. Let
me say that I have a great dela of respect for Roy. He is, indeed, a
competent engineer. He and I have philosophical differences that run
pretty deep, but I still respect him for his knowledge and competence. I
think he was, by far, the brightest of the original AR/KLH/ADVENT crew,
by orders of magnitude
> I do consider the Bass EQ on the NAD/Proton receivers in the
> question. I am still on somebody's list for tearing the
> woofer cone on a Voice of the Theater while playing with Digital
> Domain.
The EQ on things like NAD's and such probably works reasonably well with
most relatively well designed speaker systems. The VoT does not fall in
that category. It was designed long before reflex design was at all
understood, and is far from optimum in terms of things like its excursion
vs frequency profile.
Richard D Pierce
>
> boxes had a lower resonance than this. Guess I'll just have to go
> see if I can look it up somewhere. RTFM, Bummer.
>
I may have missed the referennce you're talking about, but while it may
seem counterintuitive to many, putting more woofers in a box, in fact,
raises the system resonance. This is because the effective enclosure
stiffness seen by each driver is reduced. If you have a system of volume
v with one driver and a resonance of f, then if you want to have a system
with n drivers with the same resonance, you must have n times v volume.
Rod Dorris
> >I may have missed the referennce you're talking about, but while it may
> >seem counterintuitive to many, putting more woofers in a box, in fact,
> >raises the system resonance. This is because the effective enclosure
> >stiffness seen by each driver is reduced. If you have a system of volume
> >v with one driver and a resonance of f, then if you want to have a system
> >with n drivers with the same resonance, you must have n times v volume.
> >
>
>In article <2do4d2$a...@pyrnova.mis.pyramid.com>
>lsto...@pyrnova.mis.pyramid.com (Lon Stowell) writes:
> The "reference" was either in a write-up (i.e. review) of Allison's
> Electronic Subwoofer, review of one of the NAD/Proton units with
> Bass EQ button, or in one of those instruction manuals. It was
> enough years ago (when those first appeared) I honestly can't
> recall. It was along the lines of that sealed boxes with 8-10
> inch woofers could usually be EQ'd down to the 20 Hz range with
Methinks this is apples and oranges. If I read it correctly, Lon left out the
"inches" in his original post, giving Dick the impression that he was referring
to numbers of drivers in the box. Kinda neat, considering what I learned in the
process. Danke!
----------------------------------------------------------
Rod Dorris roc...@pets.sps.mot.com
Motorola - MMTG RISC System Software Design
6501 Wm Cannon Dr, Austin, TX 78735
----------------------------------------------------------
"The more you know, the less you need."
- Australian Aboriginal saying
Richard D Pierce
>
>Bud Fried lists enclosure types from best to worst:
>
>Horn-loaded
>Transmission line
>Aperiodic (pressure release)
>Bass reflex
>Air suspension
I am not sure what sort criteria he uses, but it sure isn't based on an
engineering knowledge of the system, for example:
>To me, "aperiodic" enclosures are bass reflex cabinets with a highly
>damped port. Bud insists that they are not.
Bud would also be quite chagrined to learn that BOTH open ended
transmission lines AND aperiodic enclosure are nothing more than reflex
systems that are tuned a bit off the beaten path. The former behaves just
like a reflex system with a small box compliance and very high port mass
and port loss, while the latter behaves like a reflex with a moderate
enclosure compliance, very low port mass and high port losses. Models
based on this do a very good job of accurately predicting the measured
response of these systems, Bud Fried's handwaving notwithstanding.
Sorry, but Bud's grasp of the technical facts involved is a little weak
here. He's just MUCH more insistant about his views than even I am.
>I do tend to agree with this hierarchy. Air suspension designs don't
>do much for me.
But, as an intersting point, aperiodic designs and sealed boxes, because
of the very low port mass making a very high Fb and the high port losses
behave just like sealed boxes. And closed ended transmission lines, or
open ones with gobs of losses also behave just like sealed boxes.
Lon Stowell
>
>Methinks this is apples and oranges. If I read it correctly, Lon left out the
>"inches" in his original post, giving Dick the impression that he was referring
>to numbers of drivers in the box. Kinda neat, considering what I learned in the
>process. Danke!
>
I wondered why he segued off into multi-driver woofers!
That's why. I think I used the quote mark for inches...maybe Dick
has bad eyes?? >:-)
Lon Stowell
>
>Bud Fried lists enclosure types from best to worst:
>
>Horn-loaded
>Transmission line
>Aperiodic (pressure release)
>Bass reflex
>Air suspension
>
>
backwards unless he meant strictly in terms of efficiency, ignoring
most other issues.
>
>I feel that it is the responsibility of the amplifier not to drive the
>speaker system beyond its bandwidth limits. If your woofer cones are
>flapping around, get a better amplifier.
>
No, get a bandwidth-limited amplifier that has no bass whatsoever
and make sure none of your source material has any bass in it
whatsoever.
Amplifiers aren't there to cure speaker problems, they just amplify
with as few artifacts as possible in the process.
And the "better" amplifiers have almost without exception, very
very wide bandwidths with full power available in the deep
bass... Feed a good amp a 7-10 Hz signal and it should output
a nice high power 7-10 Hz signal.
Richard D Pierce
>In article <24.13601.26...@pcohio.com> stuart...@pcohio.com (Stuart Krivis) writes:
>>
>>Bud Fried lists enclosure types from best to worst:
>>
>>Horn-loaded
>>Transmission line
>>Aperiodic (pressure release)
>>Bass reflex
>>Air suspension
>>
> Bud Fried may make some good speakers, but that order almost looks
> backwards unless he meant strictly in terms of efficiency, ignoring
> most other issues.
And even if he did this based on efficiency, he got most of it wrong
anyway, to wit:
Max->
Horn loaded
lossless B4 reflex
Aperiodic + transmission line
Air suspension
min->
This is based on the efficiency-bandwidth-volume product for the maximum
possible in each alignment category.
Lon Stowell
>> Odd, I was under the understanding that 8-10 woofers in sealed
>> boxes had a lower resonance than this. Guess I'll just have to go
>> see if I can look it up somewhere. RTFM, Bummer.
>>
>
>I may have missed the referennce you're talking about, but while it may
>seem counterintuitive to many, putting more woofers in a box, in fact,
>raises the system resonance. This is because the effective enclosure
>stiffness seen by each driver is reduced. If you have a system of volume
>v with one driver and a resonance of f, then if you want to have a system
>with n drivers with the same resonance, you must have n times v volume.
>
The "reference" was either in a write-up (i.e. review) of Allison's
Electronic Subwoofer, review of one of the NAD/Proton units with
Bass EQ button, or in one of those instruction manuals. It was
enough years ago (when those first appeared) I honestly can't
recall. It was along the lines of that sealed boxes with 8-10
inch woofers could usually be EQ'd down to the 20 Hz range with
frequency doubling of "acceptable" amounts on most examples of the
breed.
For purpose built sealed style passive subwoofers of about 10
inches or larger, I've never had too much problems EQ'ing them
down to 20-25 Hz with fairly low doubling distortion. Some
tolerate massive drive better than others.
PS, which of the KLH engineers did the old classic Model 11
portable? I remember my owner's manual noted that it did use
modest bass EQ to make those minispeakers sound much larger...and
if you used anything else you needed to back off on the bass
control a bit. Not great fi by today's standards, but it sure
beat the old "furniture-fi" systems of the day.
Stuart Krivis
D->suspension system with a Q of 1.5 or so (just like
D-> the older AR models, such as the AR-3a) has provably worse transient
D-> response and faster rolloff rates at cutoff than a bass reflex
D-> system using lowQ QB3 alignments for a similar cutoff frequency.
D-> Thus, as an apriori statement, acoustic suspension systems having
D-> better transient performance and lower roloff rates is simply not
D-> supported by the facts.
Bud Fried lists enclosure types from best to worst:
Horn-loaded
Transmission line
Aperiodic (pressure release)
Bass reflex
Air suspension
To me, "aperiodic" enclosures are bass reflex cabinets with a highly
damped port. Bud insists that they are not.
I do tend to agree with this hierarchy. Air suspension designs don't
do much for me.
D->Your assertion about the action of the driver below the enclosure
D->resonance, while absolutely true, fails to address one of the great
D->advantages of reflex system in that around the region of enclosure
D->resonance and for a substantial region abovbe and below, cone excursion
D->is significantly reduced increasing power handling capability and
D->reducing distortion dramatically. Below that point, and below system
D->resonance in an acoustic suspension system, the driver is working
D->essentially below it's mechanical resonance, and for drivers of similar
D->mechanical characteristics, the actual cone excursion and thus excursion
D->limited capability, they're damned near identical.
I feel that it is the responsibility of the amplifier not to drive the
speaker system beyond its bandwidth limits. If your woofer cones are
flapping around, get a better amplifier.
D->The reason that bass reflex cabinets are not tuned (if that's what you
D->mean by "ported") at 20 to 30 Hz has absolutely nothing whatsoever to do
D->with blandness of sound or lack of warmth in the midbass. The tuning
D->frequency of a bass reflex system is dependent upon the system alignment,
D->and that is determined by, essentially, the low frequency parameters, the
D->so called "Thiele-Small" parameters of the driver. A given driver has a
D->given set of parameters that are determined by things such as cone mass,
D->suspension compliance, mechancial and electrical losses and so on. These
D->parameters determine essentially the useful range of tuning for the
D->system.
D->As a generalization (one backed up with statistics from several thousand
D->drivers), one can say that most 6 1/2" drivers suitable for reflex use
D->work best with enclosure tunings between 40 and 65 Hz. Most 8" driver,
D->around 30 to 45 Hz, 10" around 28 to 40 Hz, 12" 25 to 35 Hz, and so on.
D->To be suitable for use in a design that is tuned around 20 Hz, the driver
D->has to have a free air resonance equal to or less than 20 Hz (systems with
D->Fb's less than Fs tend to be relatively poorly designed), a driver total Q
D->of about 0.4 or less, and so on. This essentially means large drivers with
D->heavy cones, high compliances and large magnets. These drivers are more
D->expensive and thus are sold less frequenctly than cheaper drivers. Thus
D->the reason most systems are not tuned so low is one of pure and simple
D->economic statistics driven by physical reality.
D->Warmth and blandness play no part.
Agreed. There are very few drivers that meet the above specs. The
most common large drivers, designed for PA use, are low compliance and
their Fs is too high. Dr. R. A. Greiner (sp?) has done some work with
real subwoofers and spent a lot of money to get reasonable results.
Flat response to 20Hz is not something that you will find at your local
appliance store. (Although some dishwashers come close) :-)
... Foo Mail * <-- Registered with beads and trinkets.
stuart...@pcohio.com OR bp...@cleveland.freenet.edu
---
ş SPEED 1.4i #1424 ş I tried the rest and now use the best! SPEED READ!
busenitz
>In article <24.13601.26...@pcohio.com> stuart...@pcohio.com (Stuart Krivis) writes:
>>
>>Bud Fried lists enclosure types from best to worst:
>>
>>Horn-loaded
>>Transmission line
>>Aperiodic (pressure release)
>>Bass reflex
>>Air suspension
>
>I am not sure what sort criteria he uses, but it sure isn't based on an
>engineering knowledge of the system, for example:
>
>>To me, "aperiodic" enclosures are bass reflex cabinets with a highly
>>damped port. Bud insists that they are not.
>
used in some car enclosures, where the woofer is mounted in an infinite
baffle type environment, and an aperiodic membrane is mounted in front
of or behind the woofer, along with perhaps a processor.
If I am correct, this type of alignment creates a very smooth, flat
impedance curve, positively affecting both the mechanical (the aperiodic
membrane) and electrical components (the processor). I heard about a speaker
system a long time ago that did this in the home, but it was very large, so
it never caught on. Is this still done in home audio?
Would it be possible to mount a woofer in a wall or floor and use this
type of low Q, high damping method?
Richard D Pierce
>In article <CHHHA...@world.std.com> DPi...@world.std.com (Richard D Pierce) writes:
> ^^^^^^^^^
>>I am not sure what sort criteria he uses, but it sure isn't based on an
>>engineering knowledge of the system, for example:
>>
>>>To me, "aperiodic" enclosures are bass reflex cabinets with a highly
> ^^^^^^^^^
>>>damped port. Bud insists that they are not.
>>
> In this article, is the "aperiodic" enclosure the similar to those
>used in some car enclosures, where the woofer is mounted in an infinite
>baffle type environment, and an aperiodic membrane is mounted in front
>of or behind the woofer, along with perhaps a processor.
> If I am correct, this type of alignment creates a very smooth, flat
>impedance curve, positively affecting both the mechanical (the aperiodic
>membrane) and electrical components (the processor). I heard about a speaker
>system a long time ago that did this in the home, but it was very large, so
>it never caught on. Is this still done in home audio?
Well, regardless of the claim, the assertion about the impedance curve is
provably wrong. This has been claimed many, many times and this is a
complete myth that is so easily destroyed ("You say the impedance curve
is smooth and flat? Then how come it STILL measures with a large bump at
resonance and your emembrane seems to have not done a thing to it?").
The reason it doesn't work is that THE principal source of damping in all
modern drivers is electrical. The electrical losses completely overwhelm
the combined mechanical and acoustical losses, typicaly by a factor of 5
or 10 to 1. And the mechanical losses in the driver suspension typically
are much greater than the acoustical losses by a similar number. Thus even
if we were able to double or quadruple the acoustical losses (which this
membrane has little chance of doing), it would not have a great effect on
the total losses, or damping, of the system. For similar reasons, the
total system Q of a sealed box does not change much with internal damping.
Aperiodic system, as much as transmission lines, are surrounded by much
myth and ballyhoo. The may well be operating configurations that mioght
have some performance features that are desirable, but they simply do not
work in the magical and mystical ways claimed by their proponents.
Specifically, their claims make certain predictions about performance
that are measurable in pretty unambiguous fashions, and they fail to
measure according to the predictions. Thus, I submit, the claims are wrong.
>Would it be possible to mount a woofer in a wall or floor and use this
>type of low Q, high damping method?
Why not just use a low Q, highly damped driver and get it over with? Why
bother with badly developed, poorly thought out "theories" that simply
fail in their basic requirement to predict reality?
Matt Kennel
Given two "approximately equal" speakers say with similar construction and
driver budgets and intended frequency responses, one sealed-box and the
other reflex, which is more likely to drive unwanted cabinet modes?
(i.e. going beyond the linear theory)
You may define "approximately equal" in whichever ways are most
educational.
Thanks,
matt
--
-Matt Kennel m...@inls1.ucsd.edu
-Institute for Nonlinear Science, University of California, San Diego
-*** AD: Archive for nonlinear dynamics papers & programs: FTP to
-*** lyapunov.ucsd.edu, username "anonymous".
Richard D Pierce
>So how about a down-to-earth question: Suppose you have
>approx $1500-2000 to spend on a pair of speakers, would
>you go for a sealed box or a reflex? (other things being
>equal, size not a factor, etc).
Despite my active participation in the reflex vs selaed box technical
discussuions of late, if I were looking for $1500-$2000 speakers, I would
consider enclosure configuration such as this one of the lesser issues,
believe it or not. Both system can be designed to do jobs well.
If it were my money, I would focus more on whether it suited my
requirements for sound, accuracy, output capability, whether it fit in my
listening space, whether the elctronics and it were compatible,
reliability, longevity, whether the dealer and the manufacturer were going
to stand behind it, and so on.
Then, as a metter of academic interst, I might rip the grill cloth off
and look at it.
Lon Stowell
>
>So how about a down-to-earth question: Suppose you have
>approx $1500-2000 to spend on a pair of speakers, would
In that price range a sealed box or reflex is a pretty good idea.
In other words, in that price range, the difference between acoustic
suspension and reflex should be near the bottom of your selection
criteria unless you have prejudices unrelated to sound quality.
Steve Safarik
approx $1500-2000 to spend on a pair of speakers, would
busenitz
I wouldn't make the decision based upon enclosure type.
I think the best way is to take some favorite CD's and listen
for a while, preferably in your home, and see which sounds
better. I think that there are too many other factors to just
decide between sealed and ported.
Brock Hannibal
|>>So how about a down-to-earth question: Suppose you have
|>>approx $1500-2000 to spend on a pair of speakers, would
|>>you go for a sealed box or a reflex? (other things being
|>>equal, size not a factor, etc).
|>
|>If it were my money, I would focus more on whether it suited my
|>requirements for sound, accuracy, output capability, whether it fit in my
|>listening space, whether the elctronics and it were compatible,
|>reliability, longevity, whether the dealer and the manufacturer were going
|>to stand behind it, and so on.
|>
Bravo Dick! Some of the sanest advice I've seen in this newsgroup, but you
neglected the most important factor --SAF (spouse acceptance factor).
Brock
Chris Christensen
Note: There may actually be something worth reading in this posting.
If you usually pass over my postings, you might want to read this
one! If you have me in your kill file, you won't see this pitiful
plea..... :-)
Both Dick and Lon has provided nice responses to this posting. So why
am I posting? :-) I thought that I would give you the Lunatic Fringe
response.....
With the parameters you have set I would go for a combination design.
The system would surely have dual subwoofers. The subs might be part
of the "main" cabinet design just to confuse everyone...... :-).
The HF, MF and LMF would be a sealed design. The sub would be a
reflex design that would have a cutoff as close to 20 Hz as possible.
Oh and the design would be tri-amped, electronic crossover! The amps
and crossover aren't included in the $1500-2000.........
I would consider a design that used reflex tuning for the LMF and the
MF portions but (and since I don't know any better..) I would be
concerned that the "other" ported sections would interfere with the
overall performance of the system. My reasoning is that each port is
the front end of a Helmholtz resonator........ Please correct me if I
am wrong!
Stuart Krivis
-Lon StowellÄ-----------------------------------------------------
| And the "better" amplifiers have almost without exception, very|
| very wide bandwidths with full power available in the deep |
| bass... Feed a good amp a 7-10 Hz signal and it should output|
| a nice high power 7-10 Hz signal. |
------------------------------------------------------------------
Why? There isn't any music at 7-10 Hz.
... All left-handed people, please raise your right hand!
stuart...@pcohio.com OR bp...@cleveland.freenet.edu
---
ţ SPEED 1.4i #1424 ţ I tried the rest and now use the best! SPEED READ!
Lon Stowell
>
>
>
>-----------------------------------------------------
>| And the "better" amplifiers have almost without exception, very|
>| very wide bandwidths with full power available in the deep |
>| bass... Feed a good amp a 7-10 Hz signal and it should output|
>| a nice high power 7-10 Hz signal. |
> ------------------------------------------------------------------
>
>
>Why? There isn't any music at 7-10 Hz.
If you have to ask, you wouldn't understand the answer. >:-)
There are SOME CD's with frequencies that low...and some of the
multi-kilobuck subwoofers will attempt to reproduce same.
But the original issue was good power amplifiers. Some designers
like to design them with bandwidths down into that range. I like
to reward those designers with my dollars. You might guess that
I'm a fan of the H/K wideband style amplifiers.
Some amplifiers rolloff the bottom end to avoid problems with phono
inputs combined with some speakers. I'd prefer to put any such
rolloff as a switchable feature in the phono preamp.
There really isn't any "music" much below 30-40 Hz, but that
doesn't mean that there isn't signal below that on some better
digital recordings, and you certainly can feel the difference.