@Yourmando and guess what... latest issue of Stereophile arrived yesterday and what should I find on page 23 but a "Heretic AD614" bass reflex with 97dB SPL and a Faital 12HX230 coax 12"woofer, same driver I used in the Faital Dipoles above, with a series-wired crossover. Article is all about how heretical the design is, raving about the huge soundstage and advantages of big coaxial drivers.
Attachments
Last edited:
For sure, any smaller, low efficiency speaker or subwoofer has always boxy, small, and less dynamic. I remember a comment that Bennett Prescott of B&C made—that speakers the size of a person sound more realistic. He believes it’s because the directivity control is similar to instruments or people which also around that size. I think it might be multiple factors, since research on specific variables is not a slam dunk. Perhaps a combination of low distortion, high dynamic range, directivity control, etc.
BTW, living in the Bay Area, I’ve also had relatively large, full range dipoles in rooms smaller than I would like. I was always toward the bottom end of Linkwitz’ recommended room size for this system.
All the units have the same number or total taps: 4096. The Flex Eight differs because it debuts the new Device Console software, so signal flow is different. I believe it will be different for the other devices as well, once they get support for the new software, which supersedes the old plugin.
From the manuals:
2x4HD
“The 2x4 HD can compute a total of 4096 taps. These taps can be distributed as you wish across the four output channels, with the limitation that each output channel must have 6 or more taps and can have no more than 2048 taps.”
Flex
“The miniDSP Flex can compute a total of 4096 taps. These taps can be distributed as you wish across the four output channels, with the limitation that each output channel must have 6 or more taps and can have no more than 2048 taps.”
Flex Eight
“Allocating taps
Taps are allocated to each block according to the number of loaded taps. The miniDSP Flex Eight can compute a total of 4096 taps. These can be distributed as you wish across the two input channels, with the limitation that each channel must have 6 or more taps and can have no more than 2048 taps.”
About the Device Console:
“We're initially rolling out miniDSP Device Console with support for the Flex Eight, but the other members of the Flex family, the SHD series, and the 2x4 HD and DDRC-24 will be following shortly afterwards. We will make announcements on our forum when they are ready, so stay tuned!
Please be aware that, unfortunately, some of our older platforms won't be able to transition to the Device Console, either because they don't have the necessary hardware interface or because they are approaching the end of their product life cycle”
The MiniDSP device console‘s new signal flow.
So, starting with the Flex 8, would would do normal IIR xover & PEQs on the outputs, the use the FIR inputs to linearize the phase. I write that like I know what I’m talking about, but believe me, I don’t. I’m a software engineer, but not with a background in audio or DSP. I hope it’s not too hard to change the flow to linearize up front vs at each output
So the since the total number of taps are the same, we should be able to linearize to the same 100-150hz or so regardless of device.
Dirac on the Flex (and other) devices is optional. If you buy a Dirac license and activate it in the MiniDSP, it changes ”modes.” Internally it runs at 48kHz instead instead of 96kHz, and you lose access to input PEQ and FIR because Dirac uses them internally. I would personally not get the Dirac license so I could have full control.
@perrymarshall: forgive me if I have posted something along these lines before (I forget) but as you are clearly such a fan of DSP I’m surprised you haven’t tried Acourate or AudioLense. Their power and functionality make miniDSP look like a child’s toy In comparison.
@Studley thanks for mentioning these. I have not tried either. The tool I've been using is Eclipse Audio FIR designer - https://eclipseaudio.com/fir-designer/ It runs on MacOS which is one of the reasons I like it. I believe the ones you mention are Windows. Have you experimented enough to compare the various SW choices?
The Bitches Brews, the Birch Dipoles and the Flanagangster designs all have FIR filters built with Eclipse software. They are embedded in the configuration files I give away, and they render the systems linear phase with outstanding impulse and step response. So they incorporate a level of response correction universally appropriate to using those speakers in any room.
I try to be conservative in my approach to DSP. I think you can endlessly chase perfection while getting further from it. A textbook perfect speaker doesn't necessarily sound as good as it "should." You can correct frequency response down to 1/100th of an octave with FIR but I think that's almost always overkill. It's against the grain of my design philosophy to "over micro manage" the speaker. There's a sense in which you need to let it do what it wants to do while keeping performance within reasonable bounds.
Also, the MiniDSP only has 4K of FIR taps across all channels, as @Yourmando pointed out, so you can only correct above 150Hz or so. Going lower than that requires an investment in more expensive DSP hardware and I'm not all that interested in increasing the complexity and most people in the forums are not either.
As far as I can tell I'm doing roughly the same kinds of correction that this gentleman incorporates in his system with Audiolense
https://audiophilestyle.com/ca/ca-a...nd-room-correction-software-walkthrough-r682/
...But my files don't include bass correction specific to my room or any other room. I don't say a great deal about this part because most people don't ask all that many questions about the FIR part of the design.
I think you can fix most room modes with notch filters in the standard MiniDSP input filter section, just by taking some measurements and averaging them. Where I feel FIR is most useful is fixing phase problems that a minimum phase filter cannot correct.
The Bitches Brews, the Birch Dipoles and the Flanagangster designs all have FIR filters built with Eclipse software. They are embedded in the configuration files I give away, and they render the systems linear phase with outstanding impulse and step response. So they incorporate a level of response correction universally appropriate to using those speakers in any room.
I try to be conservative in my approach to DSP. I think you can endlessly chase perfection while getting further from it. A textbook perfect speaker doesn't necessarily sound as good as it "should." You can correct frequency response down to 1/100th of an octave with FIR but I think that's almost always overkill. It's against the grain of my design philosophy to "over micro manage" the speaker. There's a sense in which you need to let it do what it wants to do while keeping performance within reasonable bounds.
Also, the MiniDSP only has 4K of FIR taps across all channels, as @Yourmando pointed out, so you can only correct above 150Hz or so. Going lower than that requires an investment in more expensive DSP hardware and I'm not all that interested in increasing the complexity and most people in the forums are not either.
As far as I can tell I'm doing roughly the same kinds of correction that this gentleman incorporates in his system with Audiolense
https://audiophilestyle.com/ca/ca-a...nd-room-correction-software-walkthrough-r682/
...But my files don't include bass correction specific to my room or any other room. I don't say a great deal about this part because most people don't ask all that many questions about the FIR part of the design.
I think you can fix most room modes with notch filters in the standard MiniDSP input filter section, just by taking some measurements and averaging them. Where I feel FIR is most useful is fixing phase problems that a minimum phase filter cannot correct.
Last edited:
Perry, I became aware of these software DSP solutions from various articles that Mitch Barnett (Mitchco) has had published on another forum. You can also find interviews with him on YouTube. I was completely persuaded that this was something I needed to try. Due to extensive house renovations I’ve not yet had a chance to pursue this but later this year I will be buying AudioLense and using it to create active crossover filters for a biradial horn / OB woofer hybrid speaker I will be putting together. (I already have all the components for these speakers.)
You do need a Windows machine to initially run AudioLense and create the filters. However the filters are then loaded into a suitable convolver, such as the ones built into JRiver, or HQ Player, or Roon. Mitch has also developed his own convolver package called Hang Loose Convolver. I know for a fact that JRiver can run on a Mac (and perhaps the others can as well?) so if you had at least temporary access to a Windows laptop then you could generate the filters but then use them in a convolver running on a Mac.
You do need a Windows machine to initially run AudioLense and create the filters. However the filters are then loaded into a suitable convolver, such as the ones built into JRiver, or HQ Player, or Roon. Mitch has also developed his own convolver package called Hang Loose Convolver. I know for a fact that JRiver can run on a Mac (and perhaps the others can as well?) so if you had at least temporary access to a Windows laptop then you could generate the filters but then use them in a convolver running on a Mac.
FWIW, I've been using Acourate for years. Amazing piece of software. The learning curve is significant as it is a very flexible and therefore not too automated piece of software. Other programs are easier, but don't allow the flexibility Acourate does. I haven't used Audiolense, but read great things about that one too.
Uli Brueggemann, the man behind Acourate, is great to work with and has spent time 1:1 with me solving specific issues.
I don't see myself ever going back to uncorrected systems (with software as sophisticated as Acourate, for example).
I have exchanged with Mitch as well. Great guy; lot's of knowledge. In fact his early article on computeraudiophile.com about room correction walkthrough with Acourate and later the article on turning the system active using Acourate, is what got me going down this path. Highly recommend Mitch too.
Uli Brueggemann, the man behind Acourate, is great to work with and has spent time 1:1 with me solving specific issues.
I don't see myself ever going back to uncorrected systems (with software as sophisticated as Acourate, for example).
I have exchanged with Mitch as well. Great guy; lot's of knowledge. In fact his early article on computeraudiophile.com about room correction walkthrough with Acourate and later the article on turning the system active using Acourate, is what got me going down this path. Highly recommend Mitch too.
I just want to say I love your work, @perrymarshall ! I have also been making live-edge open baffle speakers, but yours have a level of refinement and sophistication to aspire to. Great minds think alike!?
http://natursound.com
http://natursound.com
@gregulator Gorgeous work! I l especially love the photos of your work-in-progress carpentry. Very nice to meet you, let's stay in touch.
I was on the way to the airport yesterday, so I had to cut short my replies.
What a coincidence with the Stereophile article. I been on a pro audio & coaxial kick lately as well in my research for the 2 builds I’m looking to do. That might have been how I originally found your Bitches Brew design (and/or maybe it was the dipole angle). I used to think a using pro coax seemed too simple to get good sound, but I've gotten over that. There seem to be plenty designs that simply use a single pro coaxial driver. Danley's first product was a 8" coax in a ~20" horn enclosure. I've seen more and more pro coaxial designs marketed to the hifi market lately as well.
One design I came across that measures well and seemed relatively simple to build (something similar) is this Taipuu DSP way:
Talk about refrigerators. Its a 15" BMS coax and 18" subwoofer in a sealed enclosure. I believe Kimmosto of VituixCAD fame was involved with this Finnish design. They also have a design with "just" a 15" coax in a bass reflex box. Danley's new Hyperion for the Hifi market is also a refrigerator and is an active DSP design with linear phase. That one probably costs as much as a small house.
I've also been looking into Scott Hinson's MEH/Unity horn design for the theater room, where construction is simplified slightly because it uses a coax compression driver that can cross over as low as 300hz, so you can just jump straight to 2 10" woofers in the horn vs having to cram in a bunch of mid drivers as well.
What these designs and yours all have in common is they are about 20" wide and have controlled directivity down to about 100hz due to their size. (Well, yours is more "constant" directivity.)
And finally, I've been looking into CBT line array designs, which I may use as surrounds in the movie room. I wouldn't bother trying to make those, as the cabinet design and electronics required would be a bit way too much work for me--I'd just buy one of the JBL models.
Maybe I should collect the whole set: dipoles, coaxial horn, and CBT
Back to Bitches Brew and even bass. On the plane, I thought of perhaps a simpler way to iterate. This is all optional of course. There's nothing wrong with the speaker design, it's the room we're looking to tame, if necessary.
- Just build the full Bitches Brew. If the bass is even across all of your desired listening positions, you're done! Ideally, you're looking for within +/-1db seat to seat variation from your bass region target curve, as level difference in the bass region of 1db are quite audible.
- If you are having room mode issues in the bass region, treat the subwoofer portion of the dipole as if they are separate subwoofers. You're using 2 "separate" subs and "bass managing" them using a tool like MSO to form a single virtual mono sub. Unfortunately, then you move from bi-amp to tri-amp, so you'd drop the 6db passive series crossover from the coax woofer to sub and use DSP to crossover to the "separate" subwoofers, which happen to be built into the same baffle. You'd still need a Flex 2x8 or similar for the extra outputs, and you'd use a tool like MSO to optimize the 2 subs and the "splice" between the dipole tops and the dipole subs. So you still have something that is the same physical Bitches Brew design with more DSP complexity for bass room correction.
- If that didn't do the trick, then add 1 or 2 more separate subwoofers (such as sealed), located in room corners, or wherever you can make them work, as long as they are not near the subwoofers built into the Bitches Brew. These separate subwoofers don't have to be huge, because they are just augmenting the bass where there are modal suckouts (unless you are pulling home theater duty, in which case you might want to add huge subs anyway). You'd re-run the MSO optimization with the full 3-4 subs in the "pool" of subwoofers.
I think I may use a process like this, maybe starting with a tri-amped version so I just need to vary DSP settings to experiment. I'll have spare amps to use, and I can always go back to the original bi-amp design if I find I don't meed additional subs.
Here's Anthony Grimani's talk about Bass Management I mentioned earlier. This is pt 3, which talks about the tuning process, but it's all automated with a tool like MSO. It gets cooking around 14mins in. There are good tutorials online and on YouTube for MSO.
He said he coined the term "Bass Management" ~25 years ago, so there's that
@Yourmando I like the Finnish design. I feel the top section should be dipole, with full bandwidth firing on both sides. But I bet they sound fantastic.
If I had to take a simple 2-way to a desert island I'd strongly consider a B&C 15CXN88 in a winged Open Baffle similar to my original 12" Faital Dipoles. They would only go down to 40Hz but they would have that fantastic ambient dipole sound.
RE: Bitches Brew, all the options you mention are viable if you're willing to deal with the added complexity. I have a MiniDSP 2x8 (old version) and could easily run this as a 3- or 4-way, I have plenty of amps. I just didn't feel like that approach was elegant. Too many moving parts. Plus the 2x8 can't support FIR DSP. I could kill all the birds with one stone, and do as much bass EQ as necessary, with 2 amps. The new Flex 2x8 has the FIR which is a big plus. I also think it has lower noise.
Speaking of which - The MiniDSP 2x4HD has a high enough noise floor that I have to add 12dB passive attenuators between the DSP and the amp, otherwise I hear hiss coming from the 106dB SPL tweeters.
I have a sneaking suspicion that if I switched the 110Hz xover to active triamp, the upper bass and lower midrange might tighten up. That's because in the current design at the xover frequency, the mid and subs are wired in series and I have to think that muddies up the sound a bit. That said there's nothing that shows up in the distortion plots; at low power it's around 0.3%.
I could always add my Dayton 12" subs and they would augment the bass from 15-25Hz where the dipole "clutch is slipping rapidly" at the rate of 18dB/octave, but that would only make a difference in 1% of my recordings. And again, more complexity.
If I had to take a simple 2-way to a desert island I'd strongly consider a B&C 15CXN88 in a winged Open Baffle similar to my original 12" Faital Dipoles. They would only go down to 40Hz but they would have that fantastic ambient dipole sound.
RE: Bitches Brew, all the options you mention are viable if you're willing to deal with the added complexity. I have a MiniDSP 2x8 (old version) and could easily run this as a 3- or 4-way, I have plenty of amps. I just didn't feel like that approach was elegant. Too many moving parts. Plus the 2x8 can't support FIR DSP. I could kill all the birds with one stone, and do as much bass EQ as necessary, with 2 amps. The new Flex 2x8 has the FIR which is a big plus. I also think it has lower noise.
Speaking of which - The MiniDSP 2x4HD has a high enough noise floor that I have to add 12dB passive attenuators between the DSP and the amp, otherwise I hear hiss coming from the 106dB SPL tweeters.
I have a sneaking suspicion that if I switched the 110Hz xover to active triamp, the upper bass and lower midrange might tighten up. That's because in the current design at the xover frequency, the mid and subs are wired in series and I have to think that muddies up the sound a bit. That said there's nothing that shows up in the distortion plots; at low power it's around 0.3%.
I could always add my Dayton 12" subs and they would augment the bass from 15-25Hz where the dipole "clutch is slipping rapidly" at the rate of 18dB/octave, but that would only make a difference in 1% of my recordings. And again, more complexity.
@perrymarshall yes, I bet the Finnish design would sounds great with a Dipole top. Just for aesthetics, I think I'd rather have them either 100% dipole or 100% in a box. The slab dipole looks great, and the sealed design has cool retro look.
I agree with you on complexity. I'd rather keep it simple, and only add complexity if necessary. Room correction pretty much always adds complexity whether either more DSP, more gear, room treatments, etc. I like a listening room / living room to look like a normal room and not a studio, and to keep the system as minimal as possible.
I'll keep your notes about the lower crossover in mind. Yes, for music, I wouldn't bother augmenting the bass from 15-25Hz either, or fixing modal room problems if they occurred that low. I'd be using the Brews just for music as well.
I agree with you on complexity. I'd rather keep it simple, and only add complexity if necessary. Room correction pretty much always adds complexity whether either more DSP, more gear, room treatments, etc. I like a listening room / living room to look like a normal room and not a studio, and to keep the system as minimal as possible.
Yes, the great think about the new Flex devices is that they are reference quality preamps (with remote) & DACs with extremely now noise, in addition to DSP processors. I thought my old Benchmark DAC1 was great some years ago, but this measures better than even the latest Benchmark DAC3 is in the same tier as some of the best products on the market (ASR review).
I'll keep your notes about the lower crossover in mind. Yes, for music, I wouldn't bother augmenting the bass from 15-25Hz either, or fixing modal room problems if they occurred that low. I'd be using the Brews just for music as well.
@Yourmando if you are interested in MEHs you should have a look at the latest iteration of Joseph Crowe’s Nighthawk speakers. They are not strictly MEH as such but you will see why I’ve mentioned them when you see them. I believe he will be testing them and reporting on their performance in the not too distant future. He has very high aspirations for their performance so I’m awaiting his report with great interest.
This thread has me thinking of a hybrid version of the Beyond The Ariel project, with two stacked Altec/GPA 416 15" drivers, with the lower one in a 3.9 cubic foot/110 liter sealed box, and the upper in a dipole, but retaining the 4" left and right edge radius of the current project. To prevent the overall height from getting out of hand, the bass bin would be 18" high, the upper dipole section 18" high as well, with the Yuichi A290 topping both of them (the A290 wood horn is 11"/280mm high).
Interesting thought for a Phase II of the BtA project. Compared to a single 15" driver, the series network divides the passband into two pieces, above and below 100 Hz, so IM distortion is greatly reduced, and system headroom increased. Box coloration in the critical 100 to 800 Hz band is nearly eliminated, and using the series crossover with the same midbass driver above and below assures a seamless transition in the otherwise awkward 100 Hz band.
Lots to think about here! Thank you for the great ideas in your project!
Interesting thought for a Phase II of the BtA project. Compared to a single 15" driver, the series network divides the passband into two pieces, above and below 100 Hz, so IM distortion is greatly reduced, and system headroom increased. Box coloration in the critical 100 to 800 Hz band is nearly eliminated, and using the series crossover with the same midbass driver above and below assures a seamless transition in the otherwise awkward 100 Hz band.
Lots to think about here! Thank you for the great ideas in your project!
Those Nighthawk/1349 horns are really different - in a good way. Will read up more if I can find info.
Do you know what kind of expansion shape the midbass horns are? Nice compact design for 80-350Hz.
Lynn, the speaker system I am putting together centres around Joseph Crowe’s ES 290 biradial horns which are an evolution of the Yuich 290. My plan was to use AE Dipole woofers below that on OB (or even baffle-less) but having heard the Qualio speakers I mentioned in an earlier post I‘m wondering whether I should also consider a bass reflex cab to go up to 100hz or so. As a consequence I’m very interested in what you propose above! Please do post more on this as your thinking develops.
What really drew me about this thread (and build) is the poster actually solved the problem of a wideband cardioid (or close-enough cardioid). The measurements prove it, never mind the simplified theory, which could be wrong anyway.
It never occurred to me to have identical upper and lower drivers, then knitting them together with a series crossover at 100 Hz. That solves so many problems at once. I normally avoid series crossovers because of impedance interaction, but in this case, that is desirable, and a good thing.
From my perspective, the choice between vented and closed-box is optional ... depends on the driver. The Altec/GPA 416 and 515's are short-throw drivers because the voice coil is underhung (VC is shorter than the magnetic gap, while nearly all other bass drivers are overhung, with VC longer than the gap). Underhung might seem like a bizarre choice, but it means as long as the VC remains in the gap, the driver is very linear because the field lines within the gap are straight. By contrast, with much more common overhung VC, the turns that are outside the gap see curved field lines, called "fringing flux". It usually takes exotic design to linearize this stray flux field.
Long story short, the Altec family of underhung drivers should be excursion controlled, which a closed box does. By contrast, a vented box has uncontrolled excursion below the box frequency ... the driver is effectively in free air below Fb, with very large excursions and almost no acoustic output. If a vented box is combined with a matched highpass filter, as in a B6 alignment, then excursion is as well controlled as a closed box, so that is a good alternative. Most bass drivers these days are designed with vented box use in mind, so the spider has a progressive limiting action to protect the driver from extreme LF signals. (Rule of thumb: if you see the driver moving, the VC has left the gap and it is distorting.)
So I'm using the 416 in a 3.9 cubic foot/110 liter closed box, and relying on room lift to bring up the under-50 Hz range. People who have built this alignment report strong bass through 35 Hz, which is good for a 97 dB/watt/meter efficient driver.
The hybrid approach looks widely usable. It creates a composite driver with interesting properties. Half of this driver is in a closed or vented box that is too small to support box modes (or panel resonances) below 100 Hz, and the other half is in free air on a panel with small side wings (small relative to the wavelengths used by the upper driver). So the ideal driver to be split in two is a wideband midbass driver, not a subwoofer, nor a Lowther or Fostex full-range driver. Fortunately, there are a lot of good midbass drivers out there.
I should add I have a subjective prejudice against 3-way designs. I've done a few myself and didn't like the results. I prefer 2-ways, while acknowledging their limits. The hybrid approach isn't really a 2.5, nor a 3-way, nor a subwoofer. It's not any of these things. As long as the upper and lower driver are the same, and the series crossover is around 100 Hz, it's a new thing, not quite like the others.
Could this approach be used in a Linkwitz-like open baffle with an (equalized) MTM array on top, and the hybrid bass below? Sure, why not, but that commits to digital equalization, which some like, and others don't. But the hybrid bass at least gets rid of the requirement for massive and power-wasting EQ in the bass range. That was my big problem with the original Linkwitz design ... 20 dB of EQ required for the bass drivers, which operated in the below-baffle-peak range, and in turn required heavy EQ in the working passband.
It never occurred to me to have identical upper and lower drivers, then knitting them together with a series crossover at 100 Hz. That solves so many problems at once. I normally avoid series crossovers because of impedance interaction, but in this case, that is desirable, and a good thing.
From my perspective, the choice between vented and closed-box is optional ... depends on the driver. The Altec/GPA 416 and 515's are short-throw drivers because the voice coil is underhung (VC is shorter than the magnetic gap, while nearly all other bass drivers are overhung, with VC longer than the gap). Underhung might seem like a bizarre choice, but it means as long as the VC remains in the gap, the driver is very linear because the field lines within the gap are straight. By contrast, with much more common overhung VC, the turns that are outside the gap see curved field lines, called "fringing flux". It usually takes exotic design to linearize this stray flux field.
Long story short, the Altec family of underhung drivers should be excursion controlled, which a closed box does. By contrast, a vented box has uncontrolled excursion below the box frequency ... the driver is effectively in free air below Fb, with very large excursions and almost no acoustic output. If a vented box is combined with a matched highpass filter, as in a B6 alignment, then excursion is as well controlled as a closed box, so that is a good alternative. Most bass drivers these days are designed with vented box use in mind, so the spider has a progressive limiting action to protect the driver from extreme LF signals. (Rule of thumb: if you see the driver moving, the VC has left the gap and it is distorting.)
So I'm using the 416 in a 3.9 cubic foot/110 liter closed box, and relying on room lift to bring up the under-50 Hz range. People who have built this alignment report strong bass through 35 Hz, which is good for a 97 dB/watt/meter efficient driver.
The hybrid approach looks widely usable. It creates a composite driver with interesting properties. Half of this driver is in a closed or vented box that is too small to support box modes (or panel resonances) below 100 Hz, and the other half is in free air on a panel with small side wings (small relative to the wavelengths used by the upper driver). So the ideal driver to be split in two is a wideband midbass driver, not a subwoofer, nor a Lowther or Fostex full-range driver. Fortunately, there are a lot of good midbass drivers out there.
I should add I have a subjective prejudice against 3-way designs. I've done a few myself and didn't like the results. I prefer 2-ways, while acknowledging their limits. The hybrid approach isn't really a 2.5, nor a 3-way, nor a subwoofer. It's not any of these things. As long as the upper and lower driver are the same, and the series crossover is around 100 Hz, it's a new thing, not quite like the others.
Could this approach be used in a Linkwitz-like open baffle with an (equalized) MTM array on top, and the hybrid bass below? Sure, why not, but that commits to digital equalization, which some like, and others don't. But the hybrid bass at least gets rid of the requirement for massive and power-wasting EQ in the bass range. That was my big problem with the original Linkwitz design ... 20 dB of EQ required for the bass drivers, which operated in the below-baffle-peak range, and in turn required heavy EQ in the working passband.
Last edited:
@Lynn Olson IMHO the dual woofer configuration sounds really great and NOT boxy. It came about from two things:
1) This design, which I made 2 years ago and just posted about yesterday:
https://www.diyaudio.com/community/threads/ultimate-open-baffle-gallery.123512/post-7293467
A friend said "Wouldn't it be easier to just use 2 woofers and run one of them reflex and the other one dipole?"
So I thought of doing these.
2) My friend wanted a system similar to my Bitches Brews but getting high SPL rock & roll bass from a dipole was not going to happen on his budget, so the Flanagangster was the result. You can push them really hard for high SPLs, they play loud down to 35 Hz, and you still get the wide open sound of OB.
1) This design, which I made 2 years ago and just posted about yesterday:
https://www.diyaudio.com/community/threads/ultimate-open-baffle-gallery.123512/post-7293467
A friend said "Wouldn't it be easier to just use 2 woofers and run one of them reflex and the other one dipole?"
So I thought of doing these.
2) My friend wanted a system similar to my Bitches Brews but getting high SPL rock & roll bass from a dipole was not going to happen on his budget, so the Flanagangster was the result. You can push them really hard for high SPLs, they play loud down to 35 Hz, and you still get the wide open sound of OB.
Last edited:
@Lynn Olson I would also love to try this with an acoustic suspension woofer instead of reflex.