• Welcome to ASR. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Speaker design based on large woofer

ryanosaur

Major Contributor
Joined
Mar 17, 2022
Messages
2,492
Likes
4,527
Location
Cali
Background:

I've been contemplating ways to build a better Speaker... Not that I expect to, but I want to. Better than what, though? My current mains are Phil 3s, among the last pairs built during 2019. I took delivery in May that year. I have no regrets buying these: they are wonderful Speakers. My short critique is that I have never really felt a punch from them. Not that they are "polite," and not that they cannot make me uncomfortable when a full orchestra hits, but I've always felt I'm missing something outside of their amazing SQ.

Of recent trends, the Big Woofer has been making a comeback. Neo-Vintage designs with a 12 or 15" Woofer (or 2 or 3) have been cropping up as recently as Axpona this weekend. However, it's not just Neo-vintage boasting larger Woofers. The other growing trend is Active Speakers. Both of these are the major part of my inspiration.

Beyond that, what really opened my eyes to some of this possibility was watching the Sigberg Audio Manta development. While I haven't been able to hear Thorbjørn's gear, seeing the reviews and how people are talking about them have given me the spark to look at how I can couple some of that with the main aspect of what I really love about Philharmonic Audio and Dennis' designs with his very wide dispersion. Both play toward accurate and neutral SQ.

In a way, what I envision is almost as simple as taking a Philharmonic HT Tower and setting that on top of a 12" bass box, but dialing the HT Tower up a notch while I'm at it.
 
Goals (ideas guiding my design... In the long term everything is open to discussion):

4 way design featuring a 12" Woofer (higher sensitivity Pro Driver: FaitalPro, Beyma, or 18 Sound) or 2 10" Woofers (Satori Norex or ScanSpeak Discovery), 2 Purifi 6.5" Mid-woofers, 2 midrange drivers (undecided here, but possibly 2" domes... Morel and Visaton have some that look good). Mundorf ATM Tweeter.

I am a fan of Subwoofers and I want these Speakers to integrate with Subs (effectively creating a 5-way system).

12" Driver would cover 80~320 Hz. Purifi 6.5" from 320~1280. Mids from 1280-4000. Tweet from 4000 and up.

Target sensitivity is 92-94dB. Not about Loud, but ease of accessing dynamic headroom.

As stated above, I want to maintain a very wide dispersion throughout the full range of the Speaker. The ATM will start to narrow above 10kHz and there isn't much I can do about that. ;)

System is Music first, but also multipurpose HT and Gaming.

Looking strongly at fully active design, or if that isn't feasible, using active for bass and Mid-Woof with passive Mid and Tweeter

But now for the main event...
 
Lets talk Big Woofers.

Admittedly, I am still window shopping. I am actively comparing Drivers and narrowing down what I am considering, but as this is a new world looking at Pro Drivers there is a lot to see and process. Whereas I previously fantasized about using a bigger ScanSpeak Revelator or Seas Extreme to anchor the bottom end, I am now looking for more outright punchiness throughout the whole system.

Current Woofers on the leaderboard include dual SB Acoustics 10SB29NRX75-8 per Speaker, or Scan Discovery 26W/8534, or single 12" Pro Drivers like FaitalPro 12FX600. A few from 18 Sound and Beyma have caught my eye. I've looked at Sica, Lavoce, and B&C as well. While I'm not ruling anything completely out yet, my focus is turning toward Drivers with better Motor control (lower Qes, higher Qms), lower Le and Demodulation Rings.

And so its time to start the questions...

For Pro Woofers like this, am I on the right track with what I'm looking for to blend with HiFi goals?

What Le is low enough to minimize some of the nonlinearity concerns I usually see voiced about big Woofers?

Does limiting the band these cover ameliorate any of the inductance concerns if Le is as high as 1.34mH (the FaitalPro 12FX600)?

Are there known concerns with any of the brands I've mentioned which I should avoid as I'm targeting low distortion and high dynamics with neutral and accurate SQ?

Thanks in advance for any help!!!
 
Active as in "amps & DSP in the speakers" or active as in "active XO controlled by computer or miniDSP"?

My system and room is built to embrace wide dispersion. I've been modifying my DIY boxes for over 30 years updating drivers, cabinet volume, bracing and damping material and have been doing it with active XO for 15 years - my woofers are a little bigger than your plan but I've used Pro drivers from PRV, Eminence and Pyle (30 years ago when they were good) all the while keeping all drivers sensitivity at around 90-95dB. If you're crossing your woofers at 80hz to a sub(s) then excursion isn't really going to be a big problem.

Four-way + sub is a lot more difficult to blend well with analog or digital XO. My hardest part was finding tweeters that would go low enough without using ridiculously steep filters and finding a midrange that goes high enough without beaming. For the tweeter, I took a hint from one of Troels Gravesen builds and am currently using Audax TWO34XO that I've been very pleased with. They're 8ohm but with an active XO matching drivers is a non issue.
 
First, 5-way is a lot to get matched right in terms of FR, directivity, and phase.
Then, dome mids are notorious wall splashers, at least when left unguided ; this will dictate a wide directivity for the whole thing, if you want it to be even on an extended range. Anyway, with such a combination of drivers you'll likely end up with a coverage that's wider than average. If this is your goal it's OK but IMHO it implies short listening distance (not what comes to mind with such a project) and/or controlled room acoustics if you want a 50/50ish direct/reflected mix at LP.
FWIW, in my latest project I went for a full range, controlled directivity, decent impact 3-way which ended up surprisingly nice, at least to my taste - exclusively based on PA drivers.
IMHO you might like to refine your design goals a bit first, then shopping for appropriate drivers should be much simpler.
 
Active as in "amps & DSP in the speakers" or active as in "active XO controlled by computer or miniDSP"?
A big part is looking at overall cost. I may well drop the Purifi part of my wishlist which could save me 2k! (Certainly would like to use them, but how much exquisite detail is purely audible and how much is in the brand?)

Regardless, in an ideal world, I think plate amps are more convenient in many ways, but I think good DSP processors and external Amps would be more versatile: I'm open to both ideas. Same but different is Powered Bass similar to what Graveson does in many designs. Hypex Fusion for the lower two (sets of) Drivers while using Passive for the upper reaches.

Frankly I'm a little less interested in the hybrid notion but I respect it exists and could be pragmatic in some ways.

I've been eyeing QSC Core 110F processors, used on Reverb or fleabay. I've talked to a few cats that really dig working with that system and there is more than enough room in it for almost anything I'm likely to do in Active, including some Subs, and without getting into Dante. If not that, I might go Marani processing. MiniDSP is about same cost as Marani at the end of the day, maybe more: convenient yes, but would need 4 Balanced Flex or 2 Flex HTx.

Further thoughts on this topic are definitely welcome.
 
First, 5-way is a lot to get matched right in terms of FR, directivity, and phase.
Then, dome mids are notorious wall splashers, at least when left unguided ; this will dictate a wide directivity for the whole thing, if you want it to be even on an extended range. Anyway, with such a combination of drivers you'll likely end up with a coverage that's wider than average. If this is your goal it's OK but IMHO it implies short listening distance (not what comes to mind with such a project) and/or controlled room acoustics if you want a 50/50ish direct/reflected mix at LP.
FWIW, in my latest project I went for a full range, controlled directivity, decent impact 3-way which ended up surprisingly nice, at least to my taste - exclusively based on PA drivers.
IMHO you might like to refine your design goals a bit first, then shopping for appropriate drivers should be much simpler.
Yes. The more you split the signal between bands, the difficulty raises significantly. I get it.

I really love the very wide directivity of the Philharmonic gear. I don't know what would happen if I switched up to anything more controlled. In my room, especially with open back dipole mids, I get a very unique and convincing quasi-surround in simple stereo. (I will not be pursuing open back or any sort of dipole action in my design.)

As I said above, this isn't about chasing pure SPL out of a misplaced love for Rock'N'Roll concerts that made my ears bleed, but purely for excessive dynamic headroom. This is something that I know my system is capable of, but at a significant challenge

Thank you for the link. I dig the work you did and the end result is pretty outstanding.

To your last point, I think I am in good control at this stage. My main query here is as it pertains to the Pro Woofers as I'm not as familiar with the brands and how they behave in a general sense. I like to build from the bottom up: what I do for the Woofer will dictate what stacks above. And if I can simplify, believe me when I say I will embrace that! ;)

Considering all things are on the table, I welcome further discussion and suggestions. This is far and away the biggest project I am embracing!
 
12" Driver would cover 80~320 Hz... Target sensitivity is 92-94dB.

Nice to see real-world-do-able targets, instead of wishing for the impossible!

For sealed box with a prosound woofer, you might consider the Faital 12RS430. 92 dB ballpark, very good motor and very well-behaved cone and suspension system:


In a vented box woofer, imo the Eminence Kappalite 3012LF performs way out of its league despite not having shorting rings in the motor. I used this woofer in numerous home audio speakers, with some response-shaping circuitry which would not be needed with your 320 Hz crossover target.

If you can find the sadly-discontinued 4-ohm version, even better, as it hits even harder. Based on its actual impedance curve it's more like a "5 or 6 ohm" woofer, but that category doesn't really exist, so it ended up being a "high 4-ohms woofer" whose voltage sensitivity was therefore lower than competing "4-ohm" woofers, so sadly it did not sell well enough and was discontinued.
 
Last edited:
A big part is looking at overall cost. I may well drop the Purifi part of my wishlist which could save me 2k! (Certainly would like to use them, but how much exquisite detail is purely audible and how much is in the brand?)

Regardless, in an ideal world, I think plate amps are more convenient in many ways, but I think good DSP processors and external Amps would be more versatile: I'm open to both ideas. Same but different is Powered Bass similar to what Graveson does in many designs. Hypex Fusion for the lower two (sets of) Drivers while using Passive for the upper reaches.

Yeah, I really wanted to try out some of the 6" Purifi for my midrange but for the $$$$ difference my current driver is a B&C 6MD38. I considered plate amps but I think I prefer to keep them separate but close by. Putting another hole in a big speaker just seems like more work and if an amp fails then it's a lot more work. Especially when you can get 4-6ch mix and match Hypex or Purifi amps from Buckeye for reasonable money.

If you're going to go 4-way + sub(s) you're already up to 9ch which pretty much automatically takes you into pro gear but I'd be more inclined to use it merely as a DAC (MOTU 848 or 16A for instance) and do the XO and correction with a computer. No MiniDSP stuff can do what you want. Use Audiolense, Acourate or REW and rePhase to create filters and do room correction. Once again, not getting locked into proprietary software/firmware those pro processors CAN do - but how well can they do them....

Absolutely consider your room first. Mine (ASR post w pics) is 16x20x10' (4.7x6x3m), the speakers are 25" from the front wall and the MLP is about 13' back. It's a workshop which is why I've strived for controlled wide dispersion because I move around a bit when I'm working. I made soffit mount "dif-sorbers" for front and back walls, absorption on the ceiling and shallow diffusers for sidewall reflection. But this is only what works for my room.

That visceral feeling you're after, I get it. ;) But consider how you're going to control all the drivers and signal processing because the DSP at the end will be what makes them work in your space the way you want them to.
 
I'd be considering having the large drivers in entirely separate active subwoofers personally, so they can be optimally placed in the room rather than being compromised by where the main speakers need to be for imaging. You would need appropriate DSP upstream for integration though.
 
As you know I'm a fan of 4-way, but not sure what you think you'd gain by going even further into a 5-way?
Counting separate subwoofers as the “5” part. I’m crazy, but not so much that I would want that many bands in one structure to manage.

Even 4 is almost too many but for the 2 factors I mentioned: having big impact down low and maintaining a very wide dispersion up through the Tweeter’s natural narrowing of dispersion.

I'd be considering having the large drivers in entirely separate active subwoofers personally, so they can be optimally placed in the room rather than being compromised by where the main speakers need to be for imaging.
I’ve always liked Geddes approach to managing Subs. But here I’m proposing a compromise on that by rolling my mains off and passing full responsibility for sub-80Hz to true Subs. As it stands now, I do not have any Subs on my front wall. I may change that as I work at implementing this new system, but only because I will not have mains playing down to low-20Hz, which is a big part of Geddes approach.

The only symmetry in my system, aside from the walls, are satellite positions (Surround, rear, and soon, atmos). Subs and mains are all asymmetric and placed for best measurement, alone and summed.
 
I’ve always liked Geddes approach to managing Subs. But here I’m proposing a compromise on that by rolling my mains off and passing full responsibility for sub-80Hz to true Subs. As it stands now, I do not have any Subs on my front wall. I may change that as I work at implementing this new system, but only because I will not have mains playing down to low-20Hz, which is a big part of Geddes approach.
Apologies - I'd missed you were talking about using subs as well. What frequency range are you thinking of the 12" driver covering? I've never designed a loudspeaker in my life but using such a large driver to cover principally above 80 Hz seems like a strange choice...
 
Overall I agree with the comments so far. Seems challenging but realistic. I know you're kind of looking to save on drivers so this isn't probably a very helpful suggestion, but the Purifi tweeter goes very wide up past 10khz. IIRC the bliesma tweets are also pretty wide, but not as. If you are not married to the AMT that might be worth looking at. If you went with the Purifi dome you could avoid needing a separate mid, even.
 
I’ve always liked Geddes approach to managing Subs. But here I’m proposing a compromise on that by rolling my mains off and passing full responsibility for sub-80Hz to true Subs. As it stands now, I do not have any Subs on my front wall. I may change that as I work at implementing this new system, but only because I will not have mains playing down to low-20Hz, which is a big part of Geddes approach.
Of course it depends on the room, and of course you are going to get a whole bunch of opinions, but I have found after a lot of experimenting that I prefer true full range main speakers (or carefully co-located mains/ subs). Having the 2 LF drivers within 1/4 wavelength of each other like you do for the other driver spacing (vs across the room somewhere) has a lot of benefits and allows the use of steep linear phase crossovers. I have also found that if you build enough bass headroom into the system you can use some DSP boost along with cuts to get FR just as smooth as having the subs spread out around the room and it is much easier, and finally you preserve stereo bass. So my suggestion from the peanut gallery is forget subs and go big on the woofers (15" or 18" pro driver) which will play lower and higher and cleaner and louder than a 12". Project sounds fun, good luck and keep us posted.
 
I’ve always liked Geddes approach to managing Subs. But here I’m proposing a compromise on that by rolling my mains off and passing full responsibility for sub-80Hz to true Subs... I will not have mains playing down to low-20Hz, which is a big part of Geddes approach.

While it's true that Earl does not high-pass filter his mains, they are not contributing very much down in the 20 Hz region. His mains are very-low-Q sealed boxes; Qts about .40 and -3 dB at about 114 Hz, according to my modelling program, which predicts -21 dB at 20 Hz. (He uses the B&C 15NBX100 in about 2.5 cubic feet net, sealed and stuffed, in case you'd like to model it.)

So I don't think you'd be "compromising" very much at all by using a highpass filter.

Imo there is an argument for either using a sealed box for your 12" woofer (high pass probably optional), OR using a vented box tuned ideally an octave or so below its highpass filter frequency, to get the phase-inversion at the port tuning frequency well away from the woofer's passband.
 
Last edited:
Overall I agree with the comments so far. Seems challenging but realistic. I know you're kind of looking to save on drivers so this isn't probably a very helpful suggestion, but the Purifi tweeter goes very wide up past 10khz. IIRC the bliesma tweets are also pretty wide, but not as. If you are not married to the AMT that might be worth looking at. If you went with the Purifi dome you could avoid needing a separate mid, even.
:)

I've probably spent hundreds of thousands of other peoples fantasy money on forums. Turnabout is fair play. :p

From a budgetary standpoint, unless I realize extreme savings elsewhere, my limit is around $500 per Tweeter (x3). Which gets me in the neighborhood of the Mundorf AMT I really like, or some other nice premium Be Tweets. Purifi is selling around $720. The Bliesma I looked at was hitting over $1000 per!
o_O

I get it though. The Tweeter is extremely important. This is not lost on me.
 
While it's true that Earl does not high-pass filter his mains, they are not contributing very much down in the 20 Hz region. His mains are very-low-Q sealed boxes; Qts about .40 and -3 dB at about 114 Hz, according to my modelling program, which predicts -21 dB at 20 Hz. (He uses the B&C 15NBX100 in about 2.5 cubic feet sealed and stuffed, in case you'd like to model it.)

So I don't think you'd be "compromising" very much at all by using a highpass filter.
I did not know that about his Speakers, specifically. I agree that those would not qualify in most books as full-range.

Everything I've read on his practice for placing Subs refers to counting full range Speakers as low-frequency sources, then getting the Subs (proper) off the front wall and strategically placing them elsewhere in the room, supported by measurements.

Imo there is an argument for either using a sealed box for your 12" woofer (high pass probably optional), OR using a vented box tuned ideally an octave or so below its highpass filter frequency, to get the phase-inversion at the port tuning frequency well away from the woofer's passband.
I had never heard of doing this. I like it. Thank you.
 
In a vented box woofer, imo the Eminence Kappalite 3012LF performs way out of its league despite not having shorting rings in the motor. I used this woofer in numerous home audio speakers, with some response-shaping circuitry which would not be needed with your 320 Hz crossover target.
I use this (as an 80Hz-500Hz midbass in a 3-way + sub full active setup) based on your rec and it kicks so much ass it can't fit in a post. I use it in the Eminence "Small Sealed" design since I'm crossing to a sub anyway. Every system should have such a driver dedicated to midbass. :)
 
Back
Top Bottom