Dual 10" Sub Design For Your Review

[lsiberian]

MBM stands for Mid Bass Mid?

I do still have the 8w7.

I looked at a LMS 5400 at one point. Everything worked out great except the cost of proper amplification.

Likewise for an Axis driver.

With building supplies and EQ, that leaves about $700 for driver and amp.

Build an 8w7 for crying out loud.

That sub is among the best drivers ever engineered.

Drive it with a Behringer ep2500 and you're set.

Sell off your lot of SV10 drivers if you can or save them for a more advanced build down the road. I'm interested in hearing your thoughts on MBMs TLS Guy. My use of them is as a 3rd way to a 2 way upper register speaker. The idea is divide the sound spectrum in 4 ways. They will be crossed over with a 4th order L-R filter(or that's the plan anyway.)

The TL potential may change my plans anyway, but I still plan to build them for fun and practice

I didn't see if TLS had modeled a box, but I have one around for it.

It can fit your requirements just fine.

 

[theater_lover]

I thought the 8w7 was perfect for what I was looking for, but when I compare the cost of two 8w7s (~$500) to four SV10s (~$250), and then model them like crazy, the 8w7 comes really close to being equal, if not less. So why would I spend twice as much? Now, the 12w7 or 13w7 are a different story, but those are far more money.

I have essentially decided to either go sealed, or do a passive radiator design. I get the best results using CSS Audio's 15" PR with two SV10s. At $80 each, they are almost as much as the drivers, but I think being able to tune it after the build, and the extended bass without port troubles is worth it.

 

[TLS Guy]

I thought the 8w7 was perfect for what I was looking for, but when I compare the cost of two 8w7s (~$500) to four SV10s (~$250), and then model them like crazy, the 8w7 comes really close to being equal, if not less. So why would I spend twice as much? Now, the 12w7 or 13w7 are a different story, but those are far more money.

I have essentially decided to either go sealed, or do a passive radiator design. I get the best results using CSS Audio's 15" PR with two SV10s. At $80 each, they are almost as much as the drivers, but I think being able to tune it after the build, and the extended bass without port troubles is worth it.

If you think ports are bad, then passive radiators are ten times the trouble. If you make the box small it will still be misaligned whether you use a port or passive radiator.

The passive radiator alignment requires a slightly larger box and will require two passive radiators.

Here is your alignment.

Name: SV10
Type: Standard one-way driver
Company: Audio Concepts, Inc. (ACI)
No. of Drivers = 2
Mounting = Standard
Wiring = Parallel
Fs = 18.5 Hz
Qms = 4.205
Vas = 84 liters
Cms = 0.656 mm/N
Mms = 111.3 g
Xmax = 16 mm
Sd = 310 sq.cm
Qes = 0.394
Re = 2.72 ohms
Z = 4 ohms
BL = 9.48 Tm
Pe = 250 watts
Qts = 0.36
2.83-V SPL = 87.72 dB
-----------------------------------------
Box Properties
Name:
Type: Passive Radiator Box
Shape: Prism, square (optimum)
Vb = 3.183 cu.ft
QL = 6.501
F3 = 23.5 Hz
Fill = normal
No. of Passive Radiators = 2
Fs = 6.268 Hz
Qms = 10.44
Vas = 443.4 liters
Cms = 1.216 mm/N
Mms = 530.1 g
Rms = 2 kg/s
Sd = 506.7 sq.cm

It is far better to have one driver properly aligned than two misaligned.

It seems to me you are determined to build a miss aligned sub and have poor results.

If you insist on the size constraint then your best bet by far, is to use the JL unit you have. I'm pretty sure it will best using two if the others however you load it any way.

 

[theater_lover]

Here is where I am with my thinking:

Enclosure internal volume - 4 cubic feet
Bracing & Driver displacement - 0.35 cubic feet
Passive Radiator displacement - unknown (.15 or less?)
Remaining internal volume - 3.5 cubic feet +

APR15 Radiator Spec:
Vas - 16.34 cubic feet (463 liters)
Rms - 3Kg/s
Fs - 19 (min 6.4)
XMax - 1.3" (33mm)
QMS - 7
Sd - 124"^2 (800cm^2)
Tunable Mass - 0-1200 grams

Can you plug that into your software and see how it works out for you? In WinISD, I used 600 grams and a little EQ.

I'm not determined to build a poor sub, just trying to get a $1.50 worth of performance for a buck, while still having some floor space. Plus if this were simple, what fun would that be?

 

[theater_lover]

Here are my current dual SV10 SPL plots, while keeping all other measurements within tolerances.

YELLOW:
TLS Guy's vented model
Volume 3.305 ft^3
Tuning 19.32 Hz
Port HWD - 2.25"x10"x45.03" (length slightly different in WinISD)

RED:
My ported model, following TLS Guy's vented model.
Volume 3.305 ft^3
Tuning 19.75 Hz
PR: CSS APR15 (specs above) with 260 grams

CYAN:
Sealed model.
Volume 2.0 ft^3

First image is without EQ. The second with EQ.

 

[lsiberian]

I thought the 8w7 was perfect for what I was looking for, but when I compare the cost of two 8w7s (~$500) to four SV10s (~$250), and then model them like crazy, the 8w7 comes really close to being equal, if not less. So why would I spend twice as much? Now, the 12w7 or 13w7 are a different story, but those are far more money.

I have essentially decided to either go sealed, or do a passive radiator design. I get the best results using CSS Audio's 15" PR with two SV10s. At $80 each, they are almost as much as the drivers, but I think being able to tune it after the build, and the extended bass without port troubles is worth it.

Theater_Lover, here are some problems we are having models aren't everything and they are largely theoretical, the linearity(sound quality) of your sv10 drivers won't come close to the 8w7 which maintians linearity even as it hits xmax. This is something few drivers in the world are capable. No one's mentioned this to you so I thought I would.

Step back for a second and consider the experience of TLS Guy. Take a look at his system picture and you will see what I believe to be one of the best systems on planet earth. You would be wise to follow his advice because he's built lot's of speakers and has lots of real world experience in this hobby. He's a scientist and engineer when it comes to this hobby. Ignoring him would be like ignoring a software engineer's advice on a building a piece of software.

My number 1 rule in engineering is always defer to the proven experience and expertise of others unless you have something known to be groundbreaking.

The reason I have stuck around this forum is because on DIY projects the members are like a peer review group for my designs. And as we all know every good engineered project undergoes peer review.

I know where your coming from I've only recently got into this hobby. I built speakers that really sucked when not really listening to these guys. I then applied their principles to my speakers and now I love them to death.

Listen! is my best advice. Pride should always be on the sidelines in an engineering project(which this is)

Now on driver quality as I've stated before the JL Audio 8w7 is one of the best drivers on this planet. You'd be a fool not to take advantage of it in your setup. Remember it's not all about volume. It's about sound quality. Plus JL Audio is notoriously conservative in their ratings.

I've got the measured T/S parameter if you'd like to try modeling with the actual parameters of the drivers. These where measured by Wmax(another speaker genius).

There are a lot of driver that can make noise, but few can make music.

Of course that's probably being brutal to the SV10, but it's not fair to compare the 2 drivers. They aren't even in the same class.

 

[TLS Guy]

Here is where I am with my thinking:

Enclosure internal volume - 4 cubic feet
Bracing & Driver displacement - 0.35 cubic feet
Passive Radiator displacement - unknown (.15 or less?)
Remaining internal volume - 3.5 cubic feet +

APR15 Radiator Spec:
Vas - 16.34 cubic feet (463 liters)
Rms - 3Kg/s
Fs - 19 (min 6.4)
XMax - 1.3" (33mm)
QMS - 7
Sd - 124"^2 (800cm^2)
Tunable Mass - 0-1200 grams

Can you plug that into your software and see how it works out for you? In WinISD, I used 600 grams and a little EQ.

I'm not determined to build a poor sub, just trying to get a $1.50 worth of performance for a buck, while still having some floor space. Plus if this were simple, what fun would that be?

That is not a sub, it is a mess.

Name: SV10
Type: Standard one-way driver
Company: Audio Concepts, Inc. (ACI)
No. of Drivers = 2
Mounting = Standard
Wiring = Parallel
Fs = 18.5 Hz
Qms = 4.205
Vas = 84 liters
Cms = 0.656 mm/N
Mms = 111.3 g
Xmax = 16 mm
Sd = 310 sq.cm
Qes = 0.394
Re = 2.72 ohms
Z = 4 ohms
BL = 9.48 Tm
Pe = 250 watts
Qts = 0.36
2.83-V SPL = 87.72 dB
-----------------------------------------
Box Properties
Name:
Type: Passive Radiator Box
Shape: Prism, square (optimum)
Vb = 3.4 cu.ft
QL = 6.501
F3 = 38.9 Hz
Fill = normal
No. of Passive Radiators = 1
Fs = 19 Hz
Qms = 7
Vas = 16.34 liters
Cms = 1.216 mm/N
Mms = 530.1 g
Rms = 3 kg/s
Sd = 124 sq.cm

You now have an F3 of 40 Hz and rapid roll off.

What you are not considering is that when the box is resonating the major sound output from the sub is from the port, or if in the case of a passive radiator the radiators. Now one radiator will not handle that much power, its limits will be exceeded.

So even if you do a passive radiator correctly, you will end up with big holes in four sides of the box. You will have poor mechanical integrity, and doing a decent job of bracing will be next to impossible.

I don't know why you think ports are such a problem. They are simpler and cheaper than passive radiators to construct, and you have much better chance of assured performance from modeling.

The other issue is that you can not provide a little Eq to a passive radiator sub, in fact absolutely none at all, as driver decoupling is fast and catastrophic and almost immediate. In fact you need a steep high pass filter below F3.

Your best sub is your very nice JL unit. Don't be put off that it is 8". It will blow your two ACI units out of the water, especially if those are miss tuned.

Those units are rough, tough and you are unlikely to destroy it. Build that sub to spec and well braced, and you will have something to write home about.

 

[lsiberian]

Your best sub is your very nice JL unit. Don't be put off that it is 8". It will blow your two ACI units out of the water, especially if those are miss tuned.

Those units are rough, tough and you are unlikely to destroy it. Build that sub to spec and well braced, and you will have something to write home about.

Exactly you are severely underestimating that sub. I'd trade my 12" TC Sounds-1000 driver for it any day of the week. It's the best sub for the small space that I know of and it's performance rivals most of the good 12" subs.

 

[theater_lover]

Now I am really confused. TLS, when you modeled the PR, did you add the extra weight? What you are describing to me, the F3 at 40hz & steep fall off, sounds like what I saw with no weight added. My model's response is almost identical to your vented design which you described as optimal. I'm not saying your model is wrong, just trying to figure how our figures are so off again. How much weight do you need to add to make your model behave like mine? And you are saying that at that weight you are exceeding the PR's XMax?

As for the 8w7 vs dual SV10, if we compared them both when using accurate aligned enclosures, would they be that far off? I hear what you guys are saying, and I understand the quality of the 8w7, but part of this is my trying to understand it all.

 

[theater_lover]

Here is where I am guessing we are off. I'll plot out the 8w7 shortly.

YELLOW and CYAN are still your vented model and a sealed model.

RED is with the 15" PR mentioned earlier, with no added mass. TLS, I believe this is close to what you are seeing? This is an F3 at 37.75.

ORANGE is with 200 grams. F3 at 26.6

PINK is with 530 grams. F3 at 23.35

 

[TLS Guy]

Here is where I am guessing we are off. I'll plot out the 8w7 shortly.

YELLOW and CYAN are still your vented model and a sealed model.

RED is with the 15" PR mentioned earlier, with no added mass. TLS, I believe this is close to what you are seeing? This is an F3 at 37.75.

ORANGE is with 200 grams. F3 at 26.6

PINK is with 530 grams. F3 at 23.35

The problem is you still can't do this with one passive radiator. Next just calculate the inertia of adding over half a KG of weight to the radiator, if the suspension could take it. Calculate the forces involved to move that much weight to and fro at 30 Hz. That is where the models break down, so the radiator will not move the way it is modeled, and distortion is going to go through the roof.

By the way Bullock and white wrote warnings into their program to alert the designer that the data can not be trusted. With your passive radiator, every item has a red dot against it, which is thumbs down, what ever you do with it. No green ones at all, which is the thumbs up.

I can't impress on you enough to stay away from passive radiators. Unless you have an alignment that works without that huge mass, they are a dead looser.

Peerless have matched sub drivers ans passive radiator, which is the best I know of, but not as good, or even close to the performance of the subs talked about on a lot of these forums.

You have a nice JL unit, and there are very good reasons it costs more than the ACI units. I don't see your hang up ab out taking advantage of it, when you already have it in your hands. Your logic defeats me!

I think you should wander into the auto industry, where I understand design starts with the beverage holders. Whereas the first question to ask yourself is. Will it be any good? The next question to ask, is it practical?

 

[theater_lover]

The problem is you still can't do this with one passive radiator. Next just calculate the inertia of adding over half a KG of weight to the radiator, if the suspension could take it. Calculate the forces involved to move that much weight to and fro at 30 Hz. That is where the models break down, so the radiator will not move the way it is modeled, and distortion is going to go through the roof.

I can't impress on you enough to stay away from passive radiators. Unless you have an alignment that works without that huge mass, they are a dead looser.

Peerless have matched sub drivers ans passive radiator, which is the best I know of, but not as good, or even close to the performance of the subs talked about on a lot of these forums.

You have a nice JL unit, and there are very good reasons it costs more than the ACI units. I don't see your hang up ab out taking advantage of it, when you already have it in your hands. Your logic defeats me!

I think you should wander into the auto industry, where I understand design starts with the beverage holders. Whereas the first question to ask yourself is. Will it be any good? The next question to ask, is it practical?

I don't think that my trying to understand all the options is poor logic. You are right, I do own one 8w7. Cost me $250. I bought four SV10s for the same price. Are you telling me that one 8w7 is equal to four SV10s if the enclosures are designed correctly? If I added another 8w7, now spending $500, I can see a solid argument for them being equal, or the 8w7s being superior, but I have spent twice as much doing it. Is it twice as good? I don't believe so.

As for the passive radiator, I guess I am not understanding how it can model well and the PR is designed to take over twice the weight I am suggesting yet not work. What you are saying is that the PR is so heavy with this design that it won't respond the way it is modeled - that so much of the cabinets internal forces are required to change direction - that it will create distortion instead of clean output - and this is not reflected in WinISD.

Here is a plot of the dual SV10s as vented (yellow), ported (pink), and sealed (cyan). I have also included three different models of the 8w7.

Green is WmAx's T/S figures for an 8w7, with his ideal enclosure. 1.3 cubic feet, 24 Hz tune, 8" x 2.25" x 61" slot port.

Orange is TLS Guy's suggested enclosure for standard 8w7 spec. 1.463 cubic feet, 26.56 Hz tune, 2.5" x 6" x 35.8" slot port (port length is slightly less in WinISD)

Red is my enclosure for 8w7 standard spec. 2.1 cubic feet, 23.05 Hz tune, 7"x3.5"x55" slot port.

 

[lsiberian]

To me this is an obvious choice. To TLS it's an obvious choice.

You got 1 expert and 1 hobbyist.

You have all the drivers so feel free to experiment. If you got the time build 1 PR sub, and 1 sealed, and 1 ported from your lot of subs. Then do some testing for yourself. You might be surprised at the results.

FYI 100db is the equivalent of a jackhammer so your not gonna be quiet at 108db which is around twice as loud.

I'm not sure if I like the low tune on the JL Audio 8w7 as much. It's strange to have it so far below the Fs. But it's a tough enough driver to take it that's for sure.

 

[TLS Guy]

I don't think that my trying to understand all the options is poor logic. You are right, I do own one 8w7. Cost me $250. I bought four SV10s for the same price. Are you telling me that one 8w7 is equal to four SV10s if the enclosures are designed correctly? If I added another 8w7, now spending $500, I can see a solid argument for them being equal, or the 8w7s being superior, but I have spent twice as much doing it. Is it twice as good? I don't believe so.

As for the passive radiator, I guess I am not understanding how it can model well and the PR is designed to take over twice the weight I am suggesting yet not work. What you are saying is that the PR is so heavy with this design that it won't respond the way it is modeled - that so much of the cabinets internal forces are required to change direction - that it will create distortion instead of clean output - and this is not reflected in WinISD.

Here is a plot of the dual SV10s as vented (yellow), ported (pink), and sealed (cyan). I have also included three different models of the 8w7.

Green is WmAx's T/S figures for an 8w7, with his ideal enclosure. 1.3 cubic feet, 24 Hz tune, 8" x 2.25" x 61" slot port.

Orange is TLS Guy's suggested enclosure for standard 8w7 spec. 1.463 cubic feet, 26.56 Hz tune, 2.5" x 6" x 35.8" slot port (port length is slightly less in WinISD)

Red is my enclosure for 8w7 standard spec. 2.1 cubic feet, 23.05 Hz tune, 7"x3.5"x55" slot port.

I am telling you that you can not trust models of passive radiators a lot, especially when the mass is heavy. The programs do not take inertia into account. Box pro has indicators built in to warn you the numbers may not mean much.

JL audio have sold drivers in huge numbers despite their cost. The attachment of cone to voice coil is unique and belt and braces, which Anunaki pointed out to me. JL audio, like the best manufacturers under rate their drivers, poorer ones guild the lily. So I will say that the JL unit is four times the ACI, in overall performance, including longevity. I would definitely build that sub. You can also build the sub with two ACI units properly aligned. It will work with the JL sub fine, is it performs as modeled.

Now you are making a beginners mistake of tuning too low. This lowers efficiency and drops the power response where you need it. Also tuning for too low an F3 stresses drivers and can lead to failure.

The discrepancy in vent length is because I specified a flare on the vent at the point where it leaves the box. This is to reduce turbulence and make it cleaner. You can flare it with a router bit or a rasp. Flaring a port results in it being a little bit longer.

It is a grave mistake to tune too low. That saddle in the response is bad news, as that is where you really need the power. The power response for music declines exponentially below 60 Hz. The largest power requirements are in the 100 Hz to 1500 Hz range. Subs are thirsty on power not by virtue of the power response generated, but by virtue of inefficiency. So if you have an F3 of 25 Hz you are just fine. You need to tune for the best power response above 25 Hz not below it. In any event you don't actually achieve much below the Fs of the driver for most drivers and distortion is high. That JL driver is unusual in that you can get a good power response up to about 12 Hz below Fs. That is probably another reason for the cost.

My advice is to build the JL sub first. I know if you build it correctly you will be pleased. Then if you want another sub, build the two ACI unit vented and with the correct box and porting. That will be interesting.

Then you will have two drivers you could make a third sub with for yourself or somebody else.

 

[lsiberian]

I am telling you that you can not trust models of passive radiators a lot, especially when the mass is heavy. The programs do not take inertia into account. Box pro has indicators built in to warn you the numbers may not mean much.

JL audio have sold drivers in huge numbers despite their cost. The attachment of cone to voice coil is unique and belt and braces, which Anunaki pointed out to me. JL audio, like the best manufacturers under rate their drivers, poorer ones guild the lily. So I will say that the JL unit is four times the ACI, in overall performance, including longevity. I would definitely build that sub. You can also build the sub with two ACI units properly aligned. It will work with the JL sub fine, is it performs as modeled.

Now you are making a beginners mistake of tuning too low. This lowers efficiency and drops the power response where you need it. Also tuning for too low an F3 stresses drivers and can lead to failure.

The discrepancy in vent length is because I specified a flare on the vent at the point where it leaves the box. This is to reduce turbulence and make it cleaner. You can flare it with a router bit or a rasp. Flaring a port results in it being a little bit longer.

It is a grave mistake to tune too low. That saddle in the response is bad news, as that is where you really need the power. The power response for music declines exponentially below 60 Hz. The largest power requirements are in the 100 Hz to 1500 Hz range. Subs are thirsty on power not by virtue of the power response generated, but by virtue of inefficiency. So if you have an F3 of 25 Hz you are just fine. You need to tune for the best power response above 25 Hz not below it. In any event you don't actually achieve much below the Fs of the driver for most drivers and distortion is high. That JL driver is unusual in that you can get a good power response up to about 12 Hz below Fs. That is probably another reason for the cost.

My advice is to build the JL sub first. I know if you build it correctly you will be pleased. Then if you want another sub, build the two ACI unit vented and with the correct box and porting. That will be interesting.

Then you will have two drivers you could make a third sub with for yourself or somebody else.

Agreed.

But I'd only build the jl sub tuned to 25 hz for home theater. For those looking at it for music application It can get 120db in the right box.

Look at the curve on the Kappa project to get a grasp for a good looking curve. If you want to make the sub smaller it's fine just as long as you raise that tune. Generally you want a curve with no relative minima. Don't understimate your rooms ability to boost the 20 to 30 hz range. I regularly see subs flat to 30hz or 25hz in models get down to 20 in application because of room gain.

 

[theater_lover]

Let's try this all out.

I'm not 100% convinced that I should throw out my PR idea. I have received positive support for the design from "experts" on other forums. When I presented TLS Guy's thoughts about a single heavy mass PR, inertia and the resulting distortion, the response was this: "It sounds like a 1980's conversation. Current PR's have advanced in the last 20 years and are designed to handle the weight and excursion. The above conversation only would make sense when PR's had 8 mm Xmax and were easily overloaded.(15-20 years ago) That is not the case today with PR's that have excursion in excess of 35mm. There are PR builds in the Database, I've yet to hear any complaints. I personally have tested the CSS 10 inch PR's. I tortured them with test tones with lots of mass added, they never missed a beat."

One thing I have found about DIY is there is never a shortage of wide ranging opinions. However, I'll put the PR design in my back pocket for now and concentrate on vented and sealed designs.

TLS and Isiberian, I'd like to hear the following wrapped up nice and neat from each of you, guided by my goal of using these as stereo subs to support my monitors which get down to 38 Hz:

What one vented box for the 8w7 would you make?
What one vented box for dual SV10s would you make?
What one sealed box for dual SV10s would you make?

Forget about size. What is the best aligned, best tuned, best designed box for each in your opinion?

I will make a quick, ugly prototype box for each and give them a go.

 

[TLS Guy]

Let's try this all out.

I'm not 100% convinced that I should throw out my PR idea. I have received positive support for the design from "experts" on other forums. When I presented TLS Guy's thoughts about a single heavy mass PR, inertia and the resulting distortion, the response was this: "It sounds like a 1980's conversation. Current PR's have advanced in the last 20 years and are designed to handle the weight and excursion. The above conversation only would make sense when PR's had 8 mm Xmax and were easily overloaded.(15-20 years ago) That is not the case today with PR's that have excursion in excess of 35mm. There are PR builds in the Database, I've yet to hear any complaints. I personally have tested the CSS 10 inch PR's. I tortured them with test tones with lots of mass added, they never missed a beat."

One thing I have found about DIY is there is never a shortage of wide ranging opinions. However, I'll put the PR design in my back pocket for now and concentrate on vented and sealed designs.

TLS and Isiberian, I'd like to hear the following wrapped up nice and neat from each of you, guided by my goal of using these as stereo subs to support my monitors which get down to 38 Hz:

What one vented box for the 8w7 would you make?
What one vented box for dual SV10s would you make?
What one sealed box for dual SV10s would you make?

Forget about size. What is the best aligned, best tuned, best designed box for each in your opinion?

I will make a quick, ugly prototype box for each and give them a go.

JL Audio 8w7

Two ACI SV 10 ported/optimal box.

Two ACI SV 10 sealed/optimal box.

Note you will have to be a little careful with Eq in the sealed box as those ACI drivers are limited to about half specified power by 35 Hz and limited to 20 watts at 20 Hz. This data comes from independent data stored in box pro. One of the really nice things about box pro, is that it has the largest and most reliable data base of drivers in the industry. You will note that in the vented enclosure, power is limited to 350 watts for the two drivers at around 30 Hz, but amplitude power response is maintained because of port output. Note the the power than can be applied in this alignment rapidly falls below 25 Hz, in fact like a stone, so you might want to consider a high pass filter cutting in around 20 Hz, to rapidly cut power below 20 Hz.

Note that with the JL audio driver full power can be applied until cut off. That is one of the results of the price differential.

And that is the best advice than I can give you, but you are free to listen to whoever you want.

 

[lsiberian]

JL Audio 8w7

Two ACI SV 10 ported/optimal box.

Two ACI SV 10 sealed/optimal box.

Note you will have to be a little careful with Eq in the sealed box as those ACI drivers are limited to about half specified power by 35 Hz and limited to 20 watts at 20 Hz. This data comes from independent data stored in box pro. One of the really nice things about box pro, is that it has the largest and most reliable data base of drivers in the industry. You will note that in the vented enclosure, power is limited to 350 watts for the two drivers at around 30 Hz, but amplitude power response is maintained because of port output. Note the the power than can be applied in this alignment rapidly falls below 25 Hz, in fact like a stone, so you might want to consider a high pass filter cutting in around 20 Hz, to rapidly cut power below 20 Hz.

Note that with the JL audio driver full power can be applied until cut off. That is one of the results of the price differential.

And that is the best advice than I can give you, but you are free to listen to whoever you want.

Those designs are exactly what I would make. I'd make a sub for each of em. No need to waste good drivers. Then I'd test em and see what I liked.