Hi Dennis,
Here is a sub suggestion I believe is as good as any other:
b
Thanks for the idea...
dennis
So Pallas, why do you think the SM-96 (if I have the right one) is such a cool design?
dennis
Because it's just so elegant. The SM's (M=molded horn) take Danley's Synergy approach and collapse it into a single 5" BMS Dual Concentric driver. The concentric compression driver feeds into the mouth of the horn, and the cone of the dual concentric is fed into a bandpass cabinet that comes out through taps further down the horn.
Here's a cutaway of the three-way variant, with two 8" woofers further down the horn:
Basically, it takes the major advantages of a big Dual Concentric (symmetrical polar response, coherence) and takes away the major disadvantage of the Dual Concentric approach (treble modulation due to woofer cone movement).
Here's a cutaway of the three-way variant, with two 8" woofers further down the horn:
An externally hosted image should be here but it was not working when we last tested it.
Basically, it takes the major advantages of a big Dual Concentric (symmetrical polar response, coherence) and takes away the major disadvantage of the Dual Concentric approach (treble modulation due to woofer cone movement).
I do not think its a fair comparison. The DTS-10 is design maximum performance from 12Hz to < 40Hz so anyone looking for the "Best design" would be adding other subs anyways.
For a ULF choice its unmatched in terms of output in its performance range. People should then add other subs to smooth out the response >= 40Hz.
I don't think that's really true, though. At least without the qualifier "using two drivers." A number of sealed drivers in the same airspace (~20 cubic feet) can give you similar numbers in terms of output and efficiency, at the same price. A reasonable difference is that that much box with that many drivers is going to weigh more than a DTS-10.
Not to take away from what Danley has done - incredible efficiency and ULF performance from two relatively small drivers, with the noted cost in bandwidth - but there are certainly other ways, even cheaper ways, to the same or better result in the same space.
You have seen the measurements from 12Hz to 40Hz correct? Please post another single box design that matches one DTS-10 output and distortion in that bandwidth.
I fully understand pretty much any high output design choice considering I own an IB array and I also own the premier subwoofer (LMS5400) but at $1500 one DTS-10 is still unmatched based on the measurements and we are not even considering the extra power needed to drive a sealed design or the reduce harmonic distortion a horn design offers.
My LMS5400 sealed design cost $1500 and my IB array cost $1500 so I have valid $$$ per $$$ comparisons.
The DTS-10 just has extremely impressive measurements in that specific range (Yes they are accurate 3rd party measurements)!!
btw, I still won't own one though, it does not do < 10Hz
Last edited:
You will find when sensitivity adjusted for the different drive levels used, Josh's tests show about 8 dB less output than DSL's posted measurements.
Though the DSL DTS-10 output is impressive, that difference is hard to ignore, though when I pointed out the discrepancy to Tom Danley, that is what he chose to do..
Art Welter
Attachments
So Pallas, which horn pattern would you want for two channel stereo, in your listening room and why if possible gven the tradeoffs with this technology?
Danley will probably pick just one pattern (for economic reasons of the kit proposed), allthough I can see him picking a pattern that worked best with 5.1 or 7.1 surround sound system since he does have a market presence
there instead of maximizing it for two channel only...
dennis
At these frequencies everything is lumped parameter and in that regime there is no magic. Its all highly predictable and all designs come down to closed or ported, front and back. Each approach has its advantages and its disadvantages. Put all this into a small room and its another situation altogether and what might work best in a larger more "free" field is no longer the best choice.
I will stick by my claim that nothing is going to beat a bunch of closed boxes in terms of performance in a small room. They may cost more and weigh more, take more power, a lot of things, but they will produce excessive SPLs down to the very lowest frequencies - even DC if the room is sealed well enough.
Below the rooms first mode there is only a single paramter that determines everything and that is total net volume displacement of the cone. This means that anything that is not a monopole is going to loose (only a monopole has a non-zero net volume displacement). Ported direct radiator, dual ported, anything that is not a pure monopole will loose. At and just above the first mode the room dominates over any sub design, but monopole is still the winner in terms of source drive. Above this first mode, when the room becomes "multi-modal", the sub design becomes somewhat arbitrary because its all about location and numbers.
Now in a free field or large venue, then some tricks can be performed that favor one design over another and sometimes a non-monopole is a better choice.
Hi Dennis,
Here is a sub suggestion I believe is as good as any other:
b
These are insane! The best 18" subwoofers I know. They can shift huge amounts of air without audible distortion. I used to have two of these in closed boxes and EQ'ed flat to below 20 hz.
When somebody offered me a good price for them, I sold them. But I still miss them sometimes.
I owned those 18" subwoofers too but the lacked the Xmax I needed. I consider them woofers from then subwoofers. The LMS5400 or LMS5100 choices are still the ultimate subwoofer choice that I have found anywhere. When we can put a LMS5400 in a sealed 4 cuft box, EQ it down below 15Hz and have clean output we have found something special.
Was it a 2.83V vs 1W thing? I remember some of the discussion about that but I was never concerned.
My focus is on Max clean output and the DTS-10 is up there with the best low frequency drivers. In its performance bandwidth. With a difference in power requirements.
I can't remember the last time I listened in two-channel stereo, outside of the car or headphones. Pretty much all my home music enjoyment is multichannel, if not from a discrete mix then via DPL2 or DTS Neo6 matrixing.
That said, I think a narrower pattern than typical (in the midrange my current 12" Tannoys are ~90deg, so let's say 60deg) would be good. But 90deg would be fine, too.
The DSL DTS-10 spec is 100 dB sensitivity 2.83 v into a nominal 4 ohm, which should be about 97 dB one watt one meter (2v). Since the actual impedance dips lower than 4 ohms, perhaps subtract another dB or two.
Josh Ricci’s one watt DTS-10 tests show a 87.3 dB average.
That is quite a large difference, even after the voltage difference is accounted for.
Hi Keyser,
Agree...Its a very good driver.. I'm not starting a 'P'-contest..there are IMO more than enough of stupid SPL bla..bla Power bla..bla
threads in the Subwoofers forum where you can find drivers that can be..I hate this phrase, : ..'Pushed harder'.. by injecting currents large enough to weld stainless steel..See the submitted picture of such a driver compared with the BMS 18" driver...No doubt on which one I would choose...
Here is more:
http://www.diyaudio.com/forums/subw...excursion-subwoofers-horns-6.html#post2618925
and
http://www.diyaudio.com/forums/subw...excursion-subwoofers-horns-6.html#post2618956
b
Attachments
All of that is true if we're focused on the frequency domain. But if we look at tapped horns in the *time* domain, they have an advantage over ported boxes. Tapped horns have superior group delay, because they're a quarter wave resonator instead of a helmholtz resonator.
Of course, we're still limited by displacement. Tapped horns are not magic, and the displacement of the woofer will generally determine maximum output. And given the same woofer, the sealed box will be smaller. Therefore, if space is a limitation, and power is not, a sealed box will probably give you more output in a given footprint.
But I still like tapped horns, because their group delay performance is rather spectacular. And if we consider that ported boxes are the most ubiquitous subwoofer variant, the tapped horn is a significant improvement.
Comparing a sealed box or a single reflex bandpass to a tapped horn gets trickier, because they're better behaved in the time domain. A lot will depend on whether the box types are tailored for the bandwidth in which they are playing.
Hi John
I am unconvinced that group delay at these frequencies is a real issue, but I don;t dispute that the tapped horns may be lower.
I have been around and around on this issue and I have conmcluded that "monopole" is the way to go. I say "monopole" because that encompases the designs that I favor most. The passive radiator subs that I am selling now are by far the best LF sourecs for small rooms that I have made to date. They are actually very small and handle a huge amount of power, so you can EQ them well below their passband and almost as much as you would care to do. They are ideal in the multi-sub/DCX2496 implimentation where power and EQ does all the work. All you need are some monpoles that can handle the power.
I am unconvinced that group delay at these frequencies is a real issue, but I don;t dispute that the tapped horns may be lower.
I have been around and around on this issue and I have conmcluded that "monopole" is the way to go. I say "monopole" because that encompases the designs that I favor most. The passive radiator subs that I am selling now are by far the best LF sourecs for small rooms that I have made to date. They are actually very small and handle a huge amount of power, so you can EQ them well below their passband and almost as much as you would care to do. They are ideal in the multi-sub/DCX2496 implimentation where power and EQ does all the work. All you need are some monpoles that can handle the power.
An externally hosted image should be here but it was not working when we last tested it.
Yeah that was one of my 'eureka' moments, when it occurred to me that tapped horns are basically limited by the displacement of the woofer. That's the reason that my latest one uses these ridiculously high displacement eights. A lot of people want to use large woofers in their subs, but it's actually counter-productive in a tapped horn because the box size is directly tied to the cone size. So a small-coned woofer with tons of displacement is ideal. Even efficiency is all but irrelevant.
The real challenge with a tapped horn is trying to outperform a sealed box with the same footprint. Because with a sealed box, you have the luxury of going big on cone size AND displacement, because box size doesn't get exponenentially larger as you move from an eight to a ten to a twelve. And contrary to popular belief, there isn't anything "magical" about a tapped horn. Assuming that you have plenty of power and a sturdy woofer, displacement is going to limit you long before anything else will, whether it's sealed, tapped horn, etc... IMHO, the reason that tapped horns *seem* to be so efficient is that they're very well suited to high efficiency drivers. For instance, if I put a woofer with an FS of 50hz in a tapped horn, it's trivial to get it to play to 35hz. You can't do that with a sealed box. So tapped horns like high efficiency drivers, and that makes them *seem* louder. But once you throw a big amplifier into the mix, the limiting factor becomes displacement, not efficiency. If someone is power limited, or has space to burn, I think tapped horns are a no-brainer. For instance, if you have a concert venue, there's almost no good reason NOT to run tapped horns. Concerts are always going to have limited power and nearly unlimited space.
Looks like you've taken that approach, but made it a bandpass to reduce distortion, displacement, and bandwidth? Then added a passive radiator, because they take up a LOT less space than a port with comparable output?
In other words, if your sub was a conventional bandpass, the port of the subwoofer could easily exceed the size of the subwoofer itself, which then introduces problems with spikes in the frequency response and the group delay, due to the port acting like both a helmholtz and a quarter wave resonator.
Polk did a lot of work on this in the nineties. They used to sell a single reflex bandpass sub that used two small woofers and a passive radiator in lieu of a port. I can't find a pic of it online; I think it came out in 1992 or so.
A few years later they switched over to ports instead of passive radiators, but inserted a diffuser into the port. The diffuser reduces turbulence, so that they can use a port size which is much smaller than ideal.
I did a quick patent search, couldn't find the exact patent. Here's one for an in-wall sub which builds on some of the ideas. As you can imagine, fitting a sub between two studs in the wall creates all kinds of challenges.
patent 4903300
Compact and efficient sub-woofer ... - Google Patents
Last edited:
What gets missed a lot with bandpass is that the drivers resonance in its rear enclosure (the largest volume) is at the center of the passband. In a rear ported enclosure it is at the lower edge. This means that for equivalent box volume and cone excursion, the bandpass goes lower than the same driver and box volume in a ported configuration. But more important than that is that it is a monopole. the same driver in a closed box of the same volume has no where near the same LF capability, although it is a monopole. Seems to me the choice is obvious.
Yes, ports do get large if they are to work well, with low loss and low noise. The PR is ideal in this regard. That, and they are a PITA to assemble. The PR is trivial. The enclosure is one piece.
Yes, ports do get large if they are to work well, with low loss and low noise. The PR is ideal in this regard. That, and they are a PITA to assemble. The PR is trivial. The enclosure is one piece.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.