In Dec 2005 GM wrote the following regarding aperiodic (AP) designs: "When used as originally intended, among other things it allows smaller/shorter vents in alignments that will benefit from it. F3/Qtb goes down and you wind with what is essentially a ~low Q sealed alignment above Fb with gain. Indeed, if a LspCad Pro sim is to be believed, then my 'critically damped' 20 ft^3/~16 Hz Fb EBS 'subs' have a slightly lower GD in the audible passband than an IB. For sure, they have much more gain down low where it counts and even corner loaded only elicits other's observations of 'life-like'/tight/etc. performance."
This seemed exactly what I was after so I started studying all I could find on AP, which unfortunately was not much. Recently I heard a pair of Shindo Latour speakers that claimed to be AP designs from Japan. They were very revealing and among the most natural and dynamic non-horn speakers I have ever heard.
I own a pair of TAD 1603s currently in BR enclosures designed by George Augspurger as recording studio monitors. They are very good, but lack the dynamics and scale I would like in my 16'x26'x10' listening space.
The TS figures for the TAD 1603s are as follows: Vas: 10.73 cu.ft.; Fs: 28 Hz; Qts: .342; Qms: 6.8; Qes: .36; Re: 6.6 ohm; Sd: 881 sq. cm; 85 gram cone wt.
I have come up with a box design with internal dimensions of 30"x20.5"x42.640625" equaling 15.17591689 cu.ft. volume. The ducts, 6 of them, would each be 2.5"x9"x 2" in length exiting the bottom of the box which would be standing on 2.5" legs. Tuning would be slightly less than Fs, or should it be calculated at .7072xFs and the ducts made longer?
I am interested in any and all reactions to this design and changes which you might think appropriate.
This seemed exactly what I was after so I started studying all I could find on AP, which unfortunately was not much. Recently I heard a pair of Shindo Latour speakers that claimed to be AP designs from Japan. They were very revealing and among the most natural and dynamic non-horn speakers I have ever heard.
I own a pair of TAD 1603s currently in BR enclosures designed by George Augspurger as recording studio monitors. They are very good, but lack the dynamics and scale I would like in my 16'x26'x10' listening space.
The TS figures for the TAD 1603s are as follows: Vas: 10.73 cu.ft.; Fs: 28 Hz; Qts: .342; Qms: 6.8; Qes: .36; Re: 6.6 ohm; Sd: 881 sq. cm; 85 gram cone wt.
I have come up with a box design with internal dimensions of 30"x20.5"x42.640625" equaling 15.17591689 cu.ft. volume. The ducts, 6 of them, would each be 2.5"x9"x 2" in length exiting the bottom of the box which would be standing on 2.5" legs. Tuning would be slightly less than Fs, or should it be calculated at .7072xFs and the ducts made longer?
I am interested in any and all reactions to this design and changes which you might think appropriate.
TAD 1603's
Greetings,
You bring up a lot of points worth discussing. First, you listed every spec except the BL factor. But from what I can see, this is a fine woofer capable of dynamics. Ultra dynamics? IMO not in a "box". I think your 16 cubic foot box is just too big. The TAD's are professional woofers; they are not subs.
If I were limited in design to just one (per side) 15", I'd put it in a built-like-a tank classic bass-reflex. In this case, 6.75 cubic feet tuned to 33Hz. And I mean heavy-duty, double walled, and spiked to the floor.
From my experience, and what I have heard, the only way to get "ultra dynamics" from a single/per side 15" is to horn load it. If you're going to do that, then a real true dedicated subwoofer system would be absolutely required.
Get yourself a copy of the Sheffield release "James Newton Howard and Friends" on CD and if your are not totally thrilled and left exhausted emotionally after listening, then you don't have enough dynamic expression in your loudspeaker system.
Greetings,
You bring up a lot of points worth discussing. First, you listed every spec except the BL factor. But from what I can see, this is a fine woofer capable of dynamics. Ultra dynamics? IMO not in a "box". I think your 16 cubic foot box is just too big. The TAD's are professional woofers; they are not subs.
If I were limited in design to just one (per side) 15", I'd put it in a built-like-a tank classic bass-reflex. In this case, 6.75 cubic feet tuned to 33Hz. And I mean heavy-duty, double walled, and spiked to the floor.
From my experience, and what I have heard, the only way to get "ultra dynamics" from a single/per side 15" is to horn load it. If you're going to do that, then a real true dedicated subwoofer system would be absolutely required.
Get yourself a copy of the Sheffield release "James Newton Howard and Friends" on CD and if your are not totally thrilled and left exhausted emotionally after listening, then you don't have enough dynamic expression in your loudspeaker system.
TL-1603 Low-Frequency Loudspeaker - Pioneer TAD
An ideal BR enclosure for this driver (not counting with others/George Augspurger alignments) 178.0 L tunned to 25 Hz.
(178L = 6.2860ft³)
This alignment can give you 20Hz@F-12dB with guaranteed best LF achievable for this driver. Different alignments are possible, this one above is (int. volume) for each driver, and less than the 500W (~350W) are enough to bring them to Max Linear Peak Excursion. Some engineers use smaller enclosures than ideal for different objectives like pro-sound.
I will check later in a couple of days, for validation of your data and more simulations for confirmation. I can not confirm the other subject, on the aperiodic (AP) design, with my software but others can. Regards.
An ideal BR enclosure for this driver (not counting with others/George Augspurger alignments) 178.0 L tunned to 25 Hz.
(178L = 6.2860ft³)
This alignment can give you 20Hz@F-12dB with guaranteed best LF achievable for this driver. Different alignments are possible, this one above is (int. volume) for each driver, and less than the 500W (~350W) are enough to bring them to Max Linear Peak Excursion. Some engineers use smaller enclosures than ideal for different objectives like pro-sound.
I will check later in a couple of days, for validation of your data and more simulations for confirmation. I can not confirm the other subject, on the aperiodic (AP) design, with my software but others can. Regards.
Accuracy
Hello Mr. Inductor:
I guess you are directing at me ? In a QB3 alignment, box tuning is always higher then the driver's Fs. I use the D.B. Keele re-writes as a reference. I don't use sims, I do long math. The ratio set forth by the drivers' Qts is
1.15fs= fb so 32.2 is exact.
From your link I see the Bl spec is 19.5 This is one fine driver.
Hello Mr. Inductor:
I guess you are directing at me ? In a QB3 alignment, box tuning is always higher then the driver's Fs. I use the D.B. Keele re-writes as a reference. I don't use sims, I do long math. The ratio set forth by the drivers' Qts is
1.15fs= fb so 32.2 is exact.
From your link I see the Bl spec is 19.5 This is one fine driver.
Thanks to both Scott L and Inductor for their input. In Thuras original BR patent, it is suggested that Vb should equal Vs and likewise Fb = Fs (or be slightly below FS according to some designers). This would make for rather long Lv; in the case of the TAD 1603 around 9.5" with Auspurger's correction for vent length. That is the reason I had hoped to use the aperiodic approach to shorten the vents and flatten the impedence hump. What do you fellows think?
no AP
I'm pretty sure both Inductor and myself are in agreement that aperiodic
is something we are not comfortable with. In "theory" the aperiodic is simply a sealed box with numerous small ports cut into the back panel and covered
with some sort of foam for damping. It's supposed to lower the impedance peak of a sealed box. It won't give you the deep bass of a BR.
In my above example, a nice quasi-butterworth 3rd order vented alignment requires 2 vented ducts, 4 inches in diameter, 6.29 inches in length.
I have no idea about Mr. Thuras. I learned it from Thiel/Small/Keele/Weems.
I'm pretty sure both Inductor and myself are in agreement that aperiodic
is something we are not comfortable with. In "theory" the aperiodic is simply a sealed box with numerous small ports cut into the back panel and covered
with some sort of foam for damping. It's supposed to lower the impedance peak of a sealed box. It won't give you the deep bass of a BR.
In my above example, a nice quasi-butterworth 3rd order vented alignment requires 2 vented ducts, 4 inches in diameter, 6.29 inches in length.
I have no idea about Mr. Thuras. I learned it from Thiel/Small/Keele/Weems.
Yes, aperiodic is often referred to as a "leaky sealed box," but such alignments have proven highly successful for Dynaudio and Seas loudspeakers. In addition the famous Onken speakers have been considered aperiodic by some commentators, though others have considered them highly damped BR and even modified 1/4 wave TML enclosures. It would seem that box design is more a continuum then individual types of enclosures...rather like the continuum of horn flare rates from exponental to conical; so from closed box to open baffle. What do you think?
Aperiodic and resistive-vent have a lot of semantic overlap; Richard Small referred to this class of enclosure as "resistive-vent" in an unpublished chapter of his doctoral thesis, and that's the term I prefer to use.
One property that distinguishes RV enclosures from canonic T/S alignments is the much broader vent tuning. When you directly measure the spectra of a T/S aligned vented-box, the vent output has a rather narrow peaked spectra, centered on the box frequency.
RV enclosures, depending on tuning, may have much broader tuning for the vent ... it's resistive, after all, and this lowers the Q substantially. For example, the vent output of the Ariel is pretty much flat from 30 Hz to 100 Hz, and doesn't look like the output of a vented enclosure at all, but more like a subwoofer.
I see two possible benefits to RV alignments: they are less sensitive to dynamic BL variations from the driver, since there is resistive damping provided by the vent as well as the power amplifier (the part that is sensitive to BL variations), and the felt, wool, or cotton damping in the vent suppresses organ-pipe modes in the vent.
The difficult-to-characterize properties of felt, wool, or cotton damping makes the modeling process more awkward, but I personally like the sound of RV and TL enclosures, even if there's more cut-n-try in the building process.
One property that distinguishes RV enclosures from canonic T/S alignments is the much broader vent tuning. When you directly measure the spectra of a T/S aligned vented-box, the vent output has a rather narrow peaked spectra, centered on the box frequency.
RV enclosures, depending on tuning, may have much broader tuning for the vent ... it's resistive, after all, and this lowers the Q substantially. For example, the vent output of the Ariel is pretty much flat from 30 Hz to 100 Hz, and doesn't look like the output of a vented enclosure at all, but more like a subwoofer.
I see two possible benefits to RV alignments: they are less sensitive to dynamic BL variations from the driver, since there is resistive damping provided by the vent as well as the power amplifier (the part that is sensitive to BL variations), and the felt, wool, or cotton damping in the vent suppresses organ-pipe modes in the vent.
The difficult-to-characterize properties of felt, wool, or cotton damping makes the modeling process more awkward, but I personally like the sound of RV and TL enclosures, even if there's more cut-n-try in the building process.
How did you arrive at this alignment? To the 1st approximation it is tuned to 60+ Hz, so damping will only begin rolling it off somewhat above this ‘knee’ point as damping is increased.
Even if it’s made as a 54” long continuous perimeter vent it will still be tuned 40+ Hz, so the vents must be made much longer like is typical for an Onken alignment to get down into the 20s.
Referring back to my post, the implication is you don’t even need a low mach T/S vent much less an Av = Sd one, especially one divided up into a bunch of small ones that only make them longer since the plan here is to stuff them to ‘taste’, not rely on their friction and sheer combined air mass resonant ‘slugs’ as the Onken’s does.
Note too that ‘buried’ in my response is the fact that my ‘EBS’ alignment [and any I’ve recommended to others] is in some form of MLTL, which shortens the vent required, so for a given vent length it can be larger if desired.
FWIW, beginning at this post, here’s one man’s take on how to achieve your goal: http://www.diyaudio.com/forums/multi-way/91372-altec-lansing-24.html#post3533487
WRT tuning choices, the 1603’s published Qts + connection wire resistance is high enough for a Fs tuning, but a passive XO and/or a high output impedance amp will shift it down to below Fs, so how much added series resistance will be in the signal chain and how accurate are the published specs?
If this isn’t known, then tuning to Fs and using either some form of adjustable vent or varying stuffing density in both the vent and cab may be the ‘safe’ choice.
GM
OK,GM, I lost my mind. Let's start again. How about an internal Vb of 18,542.78125 cu. ins. and a Vb of 27.86 Hz. Internal dimensions would be 35.796875"x28"x18.5". There would 2 vents across the front, each 1.5"x11.75" and 2 vents on each side, each 1.5"x7". Lv would be between 4.66" and 5.1515" including Auspurger correction. Total Sd of vents is 77.25 sq. ins.
My other figures were based on a box for an Altec 416-8, sorry about that. too much computer time and not enough sleep for a 76-year old. Does this new set of figures seem more appropriate for the TAD 1603?
My other figures were based on a box for an Altec 416-8, sorry about that. too much computer time and not enough sleep for a 76-year old. Does this new set of figures seem more appropriate for the TAD 1603?
Hmm, ~10.73 ft^3 with one pair of vents larger than the other two pair means that they will have different tunings for a given length, so will spread out/average the tuning [Fb] a bit.
For a 1st approximation then, I can only average them to [6] ~1.5” x 8.5833” vents = a ~21+” vent length, which includes a certain amount of end correction and heavy damping to get to 27.86 Hz, so with no other info I have to conclude that your vents will be too short.
Again, not familiar with GA’s vent design routine AFAIK, but knowing his TL work is ~spot on, hence the likelihood of his vent design being just as accurate, I’m wondering if you’re using it correctly and/or not posting all the pertinent details.
Again, for semi to ~aperiodic alignments, much less vent area is required compared to a low vent mach T/S reflex and the only way short of a long TL cab I know of to get the vent length reasonably short.
Yeah, know the feeling well, I’m ‘only’ ~67, but have suffered from chronic sleep deprivation for as long as I can remember, though only in recent years has it been obviously affecting my ability to do things.
GM
For a 1st approximation then, I can only average them to [6] ~1.5” x 8.5833” vents = a ~21+” vent length, which includes a certain amount of end correction and heavy damping to get to 27.86 Hz, so with no other info I have to conclude that your vents will be too short.
Again, not familiar with GA’s vent design routine AFAIK, but knowing his TL work is ~spot on, hence the likelihood of his vent design being just as accurate, I’m wondering if you’re using it correctly and/or not posting all the pertinent details.
Again, for semi to ~aperiodic alignments, much less vent area is required compared to a low vent mach T/S reflex and the only way short of a long TL cab I know of to get the vent length reasonably short.
Yeah, know the feeling well, I’m ‘only’ ~67, but have suffered from chronic sleep deprivation for as long as I can remember, though only in recent years has it been obviously affecting my ability to do things.
GM
Thanks, GM.
If I have understood GA's vent correction it is as follows for a rectangular vent:the 4th root of the ratio of vent dimensions (larger number divided by smaller number times times 1.13 times the smaller dimension. The answer is then adjusted for the number of vents (4th root of the number of vents times the previous correction) So if the vent is 1.5"x8.5833 the correction would be 2.6216"times the 4th root of 6 vents or a total of 4.103".]
If the Sd of a single vent is 12.875 sq. ins. then using the formula Lv=(1.463*10^7*r^2/Fb^2*Vb)- 1.463*r, I get a corrected vent length of 5.30741". What am I doing wrong?
Best regards,
Bill
If I have understood GA's vent correction it is as follows for a rectangular vent:the 4th root of the ratio of vent dimensions (larger number divided by smaller number times times 1.13 times the smaller dimension. The answer is then adjusted for the number of vents (4th root of the number of vents times the previous correction) So if the vent is 1.5"x8.5833 the correction would be 2.6216"times the 4th root of 6 vents or a total of 4.103".]
If the Sd of a single vent is 12.875 sq. ins. then using the formula Lv=(1.463*10^7*r^2/Fb^2*Vb)- 1.463*r, I get a corrected vent length of 5.30741". What am I doing wrong?
Best regards,
Bill
You're welcome!
Hmm, math is for the most part is a foreign language to me, so not sure what the formula looks like, but if I’m understanding you, you’re shortening the vents for multiples when the reverse is true due to increasing frictional losses with increasing number of vents. Also, the basic Lv formula has an end correction [-1.463*r], so sounds like you’re ‘double dipping’, compounding the error.
GM
Hmm, math is for the most part is a foreign language to me, so not sure what the formula looks like, but if I’m understanding you, you’re shortening the vents for multiples when the reverse is true due to increasing frictional losses with increasing number of vents. Also, the basic Lv formula has an end correction [-1.463*r], so sounds like you’re ‘double dipping’, compounding the error.
GM
GM
Actually the formula lengthens the vents. By my math, though it may be suspect, your design vent of 12.875 sq. ins. would have an Augspurger correction that adds 4.103 inches to the standard length formula. Could you explain how you arrived at the 21+ inch length? I'm wondering if you are using a different formula then the one I used per #14 or am I executing it incorrectly?
Actually the formula lengthens the vents. By my math, though it may be suspect, your design vent of 12.875 sq. ins. would have an Augspurger correction that adds 4.103 inches to the standard length formula. Could you explain how you arrived at the 21+ inch length? I'm wondering if you are using a different formula then the one I used per #14 or am I executing it incorrectly?
Oh well, shows how little I understood your explanation, so please either post the math or PM it to me.
As I noted, mine is just a 1st approximation using WinISD Pro to get the multiple vent’s basic length, end correction for a rectangular vent. The actual length will be a bit shorter due to local boundary conditions, though there’s no program available to get it really close AFAIK.
When I convert it to a single large round vent tuned to 27.86 Hz using proven math, it’s 16.25” long and adding GA’s end correction for [6] ups it to 20.35”, so for now, WinISD’s calculation could be the more accurate of the two for all I know, though I would be very surprised.
GM
As I noted, mine is just a 1st approximation using WinISD Pro to get the multiple vent’s basic length, end correction for a rectangular vent. The actual length will be a bit shorter due to local boundary conditions, though there’s no program available to get it really close AFAIK.
When I convert it to a single large round vent tuned to 27.86 Hz using proven math, it’s 16.25” long and adding GA’s end correction for [6] ups it to 20.35”, so for now, WinISD’s calculation could be the more accurate of the two for all I know, though I would be very surprised.
GM
Hi GM
I see the difference now. I have been using the diameter of one vent with an Sd of
12.875 sq. ins.= 4.04882 ins. That gives me a design Lv=1.204444 ins., plus an Augspurger correction of 2.62156 ins. times 1.5650845 (correction for 6 vents).= a total Lv = 5.30741 ins.
Augspurger specifically calls for calculating a single rectangular vent Lv correction and then multplying by a correction for multiple vents. That led me to believe that it is the volume of the individual vent that is the basis for the 27.86 Hertz resonant tuning, whereas you seem to be saying that the multiple vents should be lumped and the calculation for Lv made for a summed Sd of all vents. Am I correct in that asumption? If yes, perhaps you would not mind explaining further so that I understand the theory.
I see the difference now. I have been using the diameter of one vent with an Sd of
12.875 sq. ins.= 4.04882 ins. That gives me a design Lv=1.204444 ins., plus an Augspurger correction of 2.62156 ins. times 1.5650845 (correction for 6 vents).= a total Lv = 5.30741 ins.
Augspurger specifically calls for calculating a single rectangular vent Lv correction and then multplying by a correction for multiple vents. That led me to believe that it is the volume of the individual vent that is the basis for the 27.86 Hertz resonant tuning, whereas you seem to be saying that the multiple vents should be lumped and the calculation for Lv made for a summed Sd of all vents. Am I correct in that asumption? If yes, perhaps you would not mind explaining further so that I understand the theory.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.