Help designing AV15-X tuned to 15Hz

[doublej9886]

MY buddy and I have been reading the forums for a while are now in the design stages. But we're having trouble with the design in WinISD.

Here's what we're trying:
1. 10cf or less (or as small as possible)
2. Tuned to 15Hz
3. AE AV15-X

Problems:
1. Keeping the air velocity below 20m/s (much less 10m/s ideally) while
2. Keeping the port length inside the "golden ratio" dimensions
3. Then keeping the 1st port resonance high enough
4. Then whenever I try to adjust the port length to fit the box, and 1st port resonance, from what I remember, WinISD automatically recalculated the tuning frequency to like 10Hz or something ridiculous. How does that work?

Any comments or help would be greatly appreciated!

Jon

 

[annunaki]

You are having problems because it should not be done. The driver does not lend well to tuning much below 22hz. The best results I could come up with was a 8.0ft^3 enclosure tuned to 22hz with a vent of 4.25"H x 18"W x 45.75"L. A 4th order infrasonic filter at 12hz would be required here. It was the lowest that I could go Iinfrasonic filter) without sacrificing too much in terms of extreme deep response.

This yields a vent velocity of 15.2m/s with 1000 watts of input power. Plenty low enough for no audible noise.

The response models about -3db from the baseline flat response at 20.61hz and reaches -6db at about 18.75hz.

SPL models flat from 24.75hz to 63.63hz +/-.5db at 118db!!!

+/-3db response is from 20.81hz - 92.65hz between 115.5db and 118.5db.

This would be a very capable sub system.

If you want to play deeper than this, I would suggest the Exodus Audio Maelstrom-X 18" driver or the Audiopulse Axis 15".

 

[lsiberian]

With the model presented you can have rock concert levels in your home.

 

[doublej9886]

Thanks for your input and modeling the sub...it would still be sweet and probably better than anything I could buy for the same price. Impressive specs and the repsonse looks awesome! Would making the box 10cf or 12cf compared to 8cf tune the box lower or not?

The Maelstrom and Audiopulse would be sweet subs but a little out of my price range at the time or for my first sub! Is there anything I can do with an AE sub a 12" or 15" without PRs to tune between 15-17Hz in an 8-12.5cf box? I'll keep playin around with it but most people out there are probably better than I at this.

 

[lsiberian]

Thanks for your input and modeling the sub...it would still be sweet and probably better than anything I could buy for the same price. Impressive specs and the repsonse looks awesome!

The Maelstrom and Audiopulse would be sweet subs but a little out of my price range at the time or for my first sub! Is there anything I can do with an AE sub a 12" or 15" without PRs to tune between 15-17Hz in an 8-12.5cf box? I'll keep playin around with it but most people out there are probably better than I at this.

There is nothing I know of that should be tuned to 15 or 17 hz.

Tuning should generally be higher than your drivers Fs parameter.

Your chasing a range that is too low to control. I suggest you get up to audible range 20hz. The extremely low stuff isn't worth the effort. Plus you haven't really experienced a sub at the level we are talking. You have no idea how amazing it is.

 

[annunaki]

The only driver I am aware of that can be tuned that low (requiring a 15ft^3 or more enclosure or so) is the Maelstrom-X.

Why is this range so important to you?? What material do you plan to play that requires this capability?

Chasing this region can get VERY expensive to do it correctly. Few, if any, drivers are designed to be capable to handle this region, other than say the new 21" Maelstrom due out soon.

I would think that once you experience the above enclosure you may realize that playing that deep is not completely necessary.

 

[doublej9886]

You are having problems because it should not be done. The driver does not lend well to tuning much below 22hz. The best results I could come up with was a 8.0ft^3 enclosure tuned to 22hz with a vent of 4.25"H x 18"W x 45.75"L. A 4th order infrasonic filter at 12hz would be required here. It was the lowest that I could go Iinfrasonic filter) without sacrificing too much in terms of extreme deep response.

This yields a vent velocity of 15.2m/s with 1000 watts of input power. Plenty low enough for no audible noise.

The response models about -3db from the baseline flat response at 20.61hz and reaches -6db at about 18.75hz.

SPL models flat from 24.75hz to 63.63hz +/-.5db at 118db!!!

+/-3db response is from 20.81hz - 92.65hz between 115.5db and 118.5db.

This would be a very capable sub system.

If you want to play deeper than this, I would suggest the Exodus Audio Maelstrom-X 18" driver or the Audiopulse Axis 15".

I tried modeling your proposed sub and just wanted to make sure that you were using WinISD and not another program and that the vent was the box shaped vent with those dimensions. It looked great! I got a rear air velocity of 12m/s but was worried about the cone excursion as it quickly rose above the Xmax below 20Hz.

I was also wondering with a vent of 46" in length, while using a golden ratio of 39" as the greatest length, how will the bend in the vent affect the sound. I've read that one bend is usually ok? And would it be easiest to build as a slot port?

 

[doublej9886]

The only driver I am aware of that can be tuned that low (requiring a 15ft^3 or more enclosure or so) is the Maelstrom-X.

Why is this range so important to you?? What material do you plan to play that requires this capability?

Chasing this region can get VERY expensive to do it correctly. Few, if any, drivers are designed to be capable to handle this region, other than say the new 21" Maelstrom due out soon.

I would think that once you experience the above enclosure you may realize that playing that deep is not completely necessary.

I thought, hopefully like many newbs, that the lower it can play the better, but you're right I havent experienced this quality of sub yet and I'm sure anything tuned to and can actually play 20Hz is AMAZING! Its hard for me to imagine that going that deep isn't worth it but then again I don't know what's all involved. Thanks for the advice!

 

[lsiberian]

I tried modeling your proposed sub and just wanted to make sure that you were using WinISD and not another program and that the vent was the box shaped vent with those dimensions. It looked great! I got a rear air velocity of 12m/s but was worried about the cone excursion as it quickly rose above the Xmax below 20Hz.

I was also wondering with a vent of 46" in length, while using a golden ratio of 39" as the greatest length, how will the bend in the vent affect the sound. I've read that one bend is usually ok? And would it be easiest to build as a slot port?

xmax is going to go up pretty fast that low. Those low frequencies don't need that much help. If you build a maelstrom or AP 15" it'll will be sicker than some theaters as far as bass production goes.

 

[annunaki]

I tried modeling your proposed sub and just wanted to make sure that you were using WinISD and not another program and that the vent was the box shaped vent with those dimensions. It looked great! I got a rear air velocity of 12m/s but was worried about the cone excursion as it quickly rose above the Xmax below 20Hz.

I was also wondering with a vent of 46" in length, while using a golden ratio of 39" as the greatest length, how will the bend in the vent affect the sound. I've read that one bend is usually ok? And would it be easiest to build as a slot port?

I can't say I have ever heard of this "golden ratio", but I have built enclosures with vents longer than 39". I have had no issues with them at all. Nor have I had issues with bends in slot vents so long as the edges are rounded and vent velocity is kept low.

The Xmax does rise quickly below 20hz. It always rises very rapidly below tuning in vented enclosures.

I use WinISD Pro Alpha.

 

[annunaki]

I thought, hopefully like many newbs, that the lower it can play the better, but you're right I havent experienced this quality of sub yet and I'm sure anything tuned to and can actually play 20Hz is AMAZING! Its hard for me to imagine that going that deep isn't worth it but then again I don't know what's all involved. Thanks for the advice!

Well executed bass below 20hz is very impressive, but to do it at a high output level with low distortion takes a special woofer and almost always a lot of space. The Audiopulse Axis will play down to the 12hz range at high output levels in sealed/eq'ed enclosures as small as 4.0ft^3. Be prepared to drop some coin on the driver though.

The Maelstrom-X is the best deal going though for a woofer capable of playing extremely deep yet remaining linear. The trade off is a very large enclosure.

 

[lsiberian]

Well executed bass below 20hz is very impressive, but to do it at a high output level with low distortion takes a special woofer and almost always a lot of space. The Audiopulse Axis will play down to the 12hz range at high output levels in sealed/eq'ed enclosures as small as 4.0ft^3. Be prepared to drop some coin on the driver though.

The Maelstrom-X is the best deal going though for a woofer capable of playing extremely deep yet remaining linear. The trade off is a very large enclosure.

I dream of the day I can try an AXIS

 

[TLS Guy]

I thought, hopefully like many newbs, that the lower it can play the better, but you're right I havent experienced this quality of sub yet and I'm sure anything tuned to and can actually play 20Hz is AMAZING! Its hard for me to imagine that going that deep isn't worth it but then again I don't know what's all involved. Thanks for the advice!

Its not worth it, because you sacrifice sensitivity, and therefore efficiency and spl when you tune too low. It is much better to have good spl and flat response form 25 to 80 Hz then to tune too low. Beginners make this mistake again and again. Believe me it is a massive mistake, and gets you off on the wrong foot and a big stumble before you start.

 

[fredk]

Its not worth it, because you sacrifice sensitivity, and therefore efficiency and spl when you tune too low. It is much better to have good spl and flat response form 25 to 80 Hz then to tune too low. Beginners make this mistake again and again. Believe me it is a massive mistake, and gets you off on the wrong foot and a big stumble before you start.

Would it not depend on your overall system design? The Mal-X seems to have enough excursion and linearity to handle the job on its own, but if you were to use a high excursion driver like the CSS 15 tuned to 17Hz mated with an MBM to handle above 50Hz, you could accomplish the same thing.

 

[TLS Guy]

Would it not depend on your overall system design? The Mal-X seems to have enough excursion and linearity to handle the job on its own, but if you were to use a high excursion driver like the CSS 15 tuned to 17Hz mated with an MBM to handle above 50Hz, you could accomplish the same thing.

I can't find the T/S parameters of that CSS 15, but if you have them I can answer your question.

xmax has no importance on how low a driver should be tuned. It has everything to do with Fs and Qt. Generally a speaker is best tuned a little above F3. If you try tuning below Fs, then all you get is high distortion and poor output. The models are often deceptive if you do this. You have to model knowing what works and is practical. That takes a little experience.

 

[fredk]

With apologies to the OP. Not quite on topic, but still relevant.

If you try tuning below Fs, then all you get is high distortion and poor output. The models are often deceptive if you do this.

I'm smart enough to know that the modeling software is theoretical performance, but I never see any discussion around where reality may or may not vary from theory or even where to look.

So from the numbers below, you are suggesting that the driver should not be tuned below 19Hz? What is it about the physical properties of the drivers that causes it to distort if tuned below fs. I'm also still not clear on what the f3 of a driver is.

The numbers below are pulled from a winisd .wdr file.

Qts=0.39
Znom=6
Fs=19.2
Pe=1000
SPL=87.3
Re=3.6
Le=1.8
fLe=0
KLe=0
BL=16.5
Xmax=30 mm
Cms=0.000249864819193739
Qms=3.65
Qes=0.44
Rms=9.0891009375091
Mms=275 g
Sd=790 cm*2
Vas=0.218.4 litres

 

[TLS Guy]

With apologies to the OP. Not quite on topic, but still relevant.


I'm smart enough to know that the modeling software is theoretical performance, but I never see any discussion around where reality may or may not vary from theory or even where to look.

So from the numbers below, you are suggesting that the driver should not be tuned below 19Hz? What is it about the physical properties of the drivers that causes it to distort if tuned below fs. I'm also still not clear on what the f3 of a driver is.

The numbers below are pulled from a winisd .wdr file.

Qts=0.39
Znom=6
Fs=19.2
Pe=1000
SPL=87.3
Re=3.6
Le=1.8
fLe=0
KLe=0
BL=16.5
Xmax=30 mm
Cms=0.000249864819193739
Qms=3.65
Qes=0.44
Rms=9.0891009375091
Mms=275 g
Sd=790 cm*2
Vas=0.218.4 litres

I have looked at that driver, and optimal box tuning Fb pretty much coincides with the Fs of the driver.

That gives you an F3 of 23 Hz.

I made a typo in my last post, it should have said Fs and not F3. F3 is the point at which the output of the system is down 3 db. A reflex rolls off 24 db/octave below that a sealed enclosure and an aperiodic TL at 12 db/octave below F3.

Now when stating a design you can't start by picking Fb. The driver parameters box you in, pun intended.

A driver is a weight on a spring. Fs is the free air resonance of the driver determined by cone weight and the stiffness of the suspension. The compliance of the suspension is the stiffness of the suspension and the moving mass the weight on the end of the spring.

Now you really can't force a system like that to vibrate at frequency that produces a pure simple harmonic wave at a frequency below that. If you try motion is non linear, in other words distorted and amplitude will fall markedly.

So the whole point of reflex enclosure design is match Fs and Fb for maximum output in the allowable pass band with minimal cone excursion.

You know when you have it right as the two peaks of impedance will lie midway between Fb when you have it right.

Now the driver you selected and published the T/S parameters of, is clearly from the specs intended for a sealed box enclosure and not a vented one.

It has a very high suspension compliance. When I model it, this is what I get.

Name: CSS 15
Type: Standard one-way driver
No. of Drivers = 1
Fs = 19.2 Hz
Qms = 3.65
Vas = 218.4 liters
Cms = 0.25 mm/N
Mms = 275 g
Rms = 9.089 kg/s
Xmax = 30 mm
Xmech = 45 mm
P-Dia = 317.2 mm
Sd = 790 sq.cm
P-Vd = 2.37 liters
Qes = 0.44
Re = 3.6 ohms
Le = 1.8 mH
Z = 6 ohms
BL = 16.5 Tm
Pe = 1000 watts
Qts = 0.39
no = 0.339 %
1-W SPL = 87.45 dB
2.83-V SPL = 87.3 dB
-----------------------------------------
Box Properties
Name:
Type: Vented Box
Shape: Prism, square (optimum)
Vb = 5.624 cu.ft
Fb = 18.66 Hz
QL = 5.995
F3 = 22.8 Hz
Fill = minimal
No. of Vents = 1
Vent shape = rectangle
Vent ends = one flared
Hv = 4 in
Wv = 12 in
Lv = 63.22 in

The vent has to be very long, and the enclosure fairly large. The above is the optimal tuning box and port for that driver in a reflex enclosure.

Even so, there is droop in the frequency response starting at 50 Hz.

The other issue is that blow the tuning frequency all drivers rapidly decouple from the box. Cone excursion rapidly increases and output rapidly falls away. What output there is distorted.

The other issue is that I driver like you specked might well suffer mechanical damage in a reflex box, because of the floppy suspension, which is ideal for sealed enclosures, but not ported ones.

 

[lsiberian]

I have looked at that driver, and optimal box tuning Fb pretty much coincides with the Fs of the driver.

That gives you an F3 of 23 Hz.

I made a typo in my last post, it should have said Fs and not F3. F3 is the point at which the output of the system is down 3 db. A reflex rolls off 24 db/octave below that a sealed enclosure and an aperiodic TL at 12 db/octave below F3.

Now when stating a design you can't start by picking Fb. The driver parameters box you in, pun intended.

A driver is a weight on a spring. Fs is the free air resonance of the driver determined by cone weight and the stiffness of the suspension. The compliance of the suspension is the stiffness of the suspension and the moving mass the weight on the end of the spring.

Now you really can't force a system like that to vibrate at frequency that produces a pure simple harmonic wave at a frequency below that. If you try motion is non linear, in other words distorted and amplitude will fall markedly.

So the whole point of reflex enclosure design is match Fs and Fb for maximum output in the allowable pass band with minimal cone excursion.

You know when you have it right as the two peaks of impedance will lie midway between Fb when you have it right.

Now the driver you selected and published the T/S parameters of, is clearly from the specs intended for a sealed box enclosure and not a vented one.

It has a very high suspension compliance. When I model it, this is what I get.

Name: CSS 15
Type: Standard one-way driver
No. of Drivers = 1
Fs = 19.2 Hz
Qms = 3.65
Vas = 218.4 liters
Cms = 0.25 mm/N
Mms = 275 g
Rms = 9.089 kg/s
Xmax = 30 mm
Xmech = 45 mm
P-Dia = 317.2 mm
Sd = 790 sq.cm
P-Vd = 2.37 liters
Qes = 0.44
Re = 3.6 ohms
Le = 1.8 mH
Z = 6 ohms
BL = 16.5 Tm
Pe = 1000 watts
Qts = 0.39
no = 0.339 %
1-W SPL = 87.45 dB
2.83-V SPL = 87.3 dB
-----------------------------------------
Box Properties
Name:
Type: Vented Box
Shape: Prism, square (optimum)
Vb = 5.624 cu.ft
Fb = 18.66 Hz
QL = 5.995
F3 = 22.8 Hz
Fill = minimal
No. of Vents = 1
Vent shape = rectangle
Vent ends = one flared
Hv = 4 in
Wv = 12 in
Lv = 63.22 in

The vent has to be very long, and the enclosure fairly large. The above is the optimal tuning box and port for that driver in a reflex enclosure.

Even so, there is droop in the frequency response starting at 50 Hz.

The other issue is that blow the tuning frequency all drivers rapidly decouple from the box. Cone excursion rapidly increases and output rapidly falls away. What output there is distorted.

The other issue is that I driver like you specked might well suffer mechanical damage in a reflex box, because of the floppy suspension, which is ideal for sealed enclosures, but not ported ones.

So many people don't get this and it makes me very sad. Everyone thinks the model is what really happens.

Model's are guides for designs. WinISD is a great tool once you grasp what TLS is talking about.

Chasing the low stuff is really ridiculous anyway.

 

[fredk]

Thanks for your responses TLS Guy. I've been mulling over what you wrote and some of it makes more sense to me now. I had noticed that a number of drivers commonly used in ported alignments were 'suggested' to be sealed by Winisd.

Need to play with modeling a little more and look back at your posts.

 

[ParadigmDawg]

All you have to do is drive about 4 miles to my house.

Heck, with your woodworking skills, we should build the ported box that Chris designed for me and see what that puppy will do.

I dream of the day I can try an AXIS