Thanks for the sims. Hopefully I'll have more time over the weekend to do some sims myself. I'm definitely thinking of using the LX521 MiniDSP woofer boost EQ as a starting point. I'm going to find a high xmax large diameter driver with T/S profile that matches the LX521 Seas subs. They are already doing what I want already so adding power and SD should give me a little more headroom in the bottom octave. I'm not looking for a particularly high Q driver as there are time domain trade offs that I'd rather not make. I'm probably going to shoot for 2'x2'x4' H frame for a pair of 18s. If Linkwitz did it and it sounds so great, I'm hoping the goal is achievable.
Based on price and xmax I'm looking at 4x18" Dayton Ultimax. Not sure about other good fits. If the drivers can move sufficient air, I can worry about the EQ and gain structure issues. Also note that these will be HPF relatively low.
Thinking out loud, I may have to consider force cancellation arrangement instead of H frame given the SD....
Sent from my iPhone using Tapatalk
Based on price and xmax I'm looking at 4x18" Dayton Ultimax. Not sure about other good fits. If the drivers can move sufficient air, I can worry about the EQ and gain structure issues. Also note that these will be HPF relatively low.
Thinking out loud, I may have to consider force cancellation arrangement instead of H frame given the SD....
Sent from my iPhone using Tapatalk
Open Baffle Subwoofer Example
Here is a guy who did an open baffle subwoofer. It took 12 15 inch drivers per side and cost $4K.
mfk-projects subwoofer
Which I accessed from this thread:
http://www.diyaudio.com/forums/subwoofers/235271-open-baffle-subwoofer.html
Without modeling it to verify it, I probably can do about the same output with a pair of 18 inch drivers in an LLT (large and low tuned ported sub).
I think that this is why most on this thread think open baffle subwoofer does not make sense.
Retsel
Here is a guy who did an open baffle subwoofer. It took 12 15 inch drivers per side and cost $4K.
mfk-projects subwoofer
Which I accessed from this thread:
http://www.diyaudio.com/forums/subwoofers/235271-open-baffle-subwoofer.html
Without modeling it to verify it, I probably can do about the same output with a pair of 18 inch drivers in an LLT (large and low tuned ported sub).
I think that this is why most on this thread think open baffle subwoofer does not make sense.
Retsel
I'm in agreement that open baffle 'subwoofers' are a waste of effort and money.
I think if you want to be able to listen at moderate volumes without disturbing people, nearfield sealed subs with linkwitz transforms are ideal.
I think if you want to be able to listen at moderate volumes without disturbing people, nearfield sealed subs with linkwitz transforms are ideal.
You say the LX521 is already doing what you want, just not loud enough. So I looked up the actual specs of that speaker.
The LX521 is 3 db down at 30 hz, a far cry from the 25 hz you said you wanted. 3 db down at 30 hz isn't really a subwoofer, it's more like a woofer. MJK's standard H baffle design also does 30 hz at 3 db down. Linkwitz isn't stupid, he realizes this is a practical limit for the low end with OB.
The LX521 isn't loud despite using a total of four 10 inch woofers with 14 mm xmax. While it's true that using four more 18s with 22 mm xmax will get you more output, it's an extremely inefficient way to add more bass. Also, with high excursion drivers comes high velocity noise. That air has to go somewhere, in the case of the Ultimax it appears it's going to go through the vented pole and probably chuff or whistle.
You say that the high q driver in MJK's design poses time domain issues you'd rather not deal with, but the amount of dsp boost this low q driver will require very well might present similar problems.
One more look at it using the Ultimax drivers in the baffle size you mentioned - 2 x 2 x 4 feet with 2 drivers per unit.
This is a sim of one unit with 2 drivers. Large coil inductance effects are not accounted for in this simple sim. Also remember, this is best case scenario with no boundary reflections (other than the floor) that can cause massive suckouts in the low bass.
Pic 1 - hornresp inputs - check my inputs for mistakes and use these inputs so you can do your own sims
Pic 2 - offset - if you don't offset you will get perfect cancellation and 0 db at all frequencies
Pic 3 - frequency response with (red) and without (blue) boost, 750 watts input
Pic 4 - excursion with and without boost, 750 watts input
For this sim most of the work was done by cutting the high end sensitivity, not boosting the low end. This allowed for a modest amount of boost to gain a flat(ish) response down to 25 hz.
As you can see, excursion doesn't become a problem until you start boosting the low end, then excursion skyrockets. With the low end boosted this driver taps out at about 25 hz at this power level and will need a high pass filter to protect it below 25 hz. Velocity noise out of the vented pole will probably be terrible at this level.
It does appear that a pair of these units (with 4 drivers) will be able to do stated goal of 110 db at 25 hz with relative ease (add 6 db to the sim for 2 units).
Now just for a little dose of reality, here's the unfiltered response of this unit with one watt applied compared to the filtered response with a 2nd order 60 hz low pass to level out the response and cut the midband sensitivity to something flat(ish) and useful.
Blue is unfiltered, red is with the low pass applied. As you can see, the midband sensitivity is HUGE, with the peak at 150 hz being 95 db high (that's the dipole peak). It would take almost 20 db of boost to get 25 hz up to the level of 150 hz which is somewhat impractical and why I chose to mostly cut the high end and add very little boost to the low end.
At 25 hz the natural sensitivity of this design is about 78 db at 25 hz, and that's with 2 drivers with one watt. That is remarkably low. So while it does appear that you can achieve your stated goals with 4 of these drivers, it's a terribly impractical way to do it, absolutely abysmal sensitivity.
But if that's what you want to do, go for it.
This is a sim of one unit with 2 drivers. Large coil inductance effects are not accounted for in this simple sim. Also remember, this is best case scenario with no boundary reflections (other than the floor) that can cause massive suckouts in the low bass.
Pic 1 - hornresp inputs - check my inputs for mistakes and use these inputs so you can do your own sims
Pic 2 - offset - if you don't offset you will get perfect cancellation and 0 db at all frequencies
Pic 3 - frequency response with (red) and without (blue) boost, 750 watts input
Pic 4 - excursion with and without boost, 750 watts input
For this sim most of the work was done by cutting the high end sensitivity, not boosting the low end. This allowed for a modest amount of boost to gain a flat(ish) response down to 25 hz.
An externally hosted image should be here but it was not working when we last tested it.
As you can see, excursion doesn't become a problem until you start boosting the low end, then excursion skyrockets. With the low end boosted this driver taps out at about 25 hz at this power level and will need a high pass filter to protect it below 25 hz. Velocity noise out of the vented pole will probably be terrible at this level.
It does appear that a pair of these units (with 4 drivers) will be able to do stated goal of 110 db at 25 hz with relative ease (add 6 db to the sim for 2 units).
Now just for a little dose of reality, here's the unfiltered response of this unit with one watt applied compared to the filtered response with a 2nd order 60 hz low pass to level out the response and cut the midband sensitivity to something flat(ish) and useful.
Blue is unfiltered, red is with the low pass applied. As you can see, the midband sensitivity is HUGE, with the peak at 150 hz being 95 db high (that's the dipole peak). It would take almost 20 db of boost to get 25 hz up to the level of 150 hz which is somewhat impractical and why I chose to mostly cut the high end and add very little boost to the low end.
At 25 hz the natural sensitivity of this design is about 78 db at 25 hz, and that's with 2 drivers with one watt. That is remarkably low. So while it does appear that you can achieve your stated goals with 4 of these drivers, it's a terribly impractical way to do it, absolutely abysmal sensitivity.
But if that's what you want to do, go for it.
An externally hosted image should be here but it was not working when we last tested it.
If I may take the liberties to expand on this a bit ? Brian means the claim that di-poles create less problems with room boundaries----that claim is a bit over exaggerated and often repeated like the shiny bright start in the sky to go with open baffles and to hell with the nasty box. Yep. Every. Single.Time. The open baffle cult claim some sort of superior performance, when this is actually antiquated methodology that has been recycled to yet again become popular. You know, like Tulip Bulb Mania. If you were to go back in time and read the works of someone like Harry F Olson, you would discover exactly what Harry did: totally enclosing the previously accepted methodology of open backed cabinets, actually improved the quality of bass, tremendously.
It was then discovered that the front and back waves emanating from a driver would cancel, unless an infinite baffle was used to prevent this. I must say I am at a total loss to understand this recent open baffle CRAZE. If there was as much energy and interest in learning how to improve the "sound" of enclosures, to the point of "no sound" (colourations), the DIY world would be leaps and bounds ahead of where they are now.
An OB system doesn't eliminate or really to any extend reduce room boundary related problems. At least in my experience but it is not always easy to make an apples-to-apples comparison because OB systems are best used away from walls (1.5m away for example) and this is not typically where most people would set up a floorstanding speaker. Locating a speaker away from walls will reduce some resonances that happen near corners, and maybe that is why OB has garnered its reputation for less room mode excitement.
An OB system DEFINITELY eliminates cabinet resonances, however, and that is a major coloration of the large majority of "boxed' systems that I have listened to including those that I have built.
IMO the major strength of a properly set up OB system is the wonderful sound stage. The space should be large enough to move the system out from the walls, and should be a bit under-damped acoustically speaking because with the OB system the room reflections actually HELP build the sound stage that you hear at the listening location.
While it is a good idea to strive for dead enclosures, there is still one place from which sound escapes that is very difficult to "fix" - the driver cone itself. Unless you have a large stuffed box behind a cone that adsorbs 100% of the backwave sound will re-radiate thru the cone into the room. A large dead enclosures like that ends up being quite heavy and are difficult to handle and install. Ever try to move a 250 lbs box? I once tried to build a prototype enclosure with mass-loaded walls and after I was done I just destroyed it because it was totally impractical. An OB is easy to make and can be simple and lightweight. I agree that bass requires large baffles or TL-type structures, but if you transition to a CB system around 100Hz the dimensions of the OB are really not large at all.
I really don't think that you can invoke stuff that Olson did in the 50's as some kind of rationale for what is done today. Driver performance and technology was completely different back then and there was not a great understanding of how to get a loudspeaker to have some desired response at the time. Olson was a pioneer in many ways, but he did not have the tools and hardware that we have today and was just trying to make the best of what was available at the time.
Thanks for the sim showing that it is possible to reach the goal with 4x18" drivers.
As far as dipoles go, I'm a believer in what Linkwitz writes about late reflections with uniform polar response allow the brain to forget about the room. Psychoacoustics backs it up.
Dipole bass is completely different in my experience. It just seems more lifelike. I find that visitors are kind of split WRT LX521 bass. Some people look at the 4 10" drivers and expect it to sound like monopole woofers of equal size and expect room gain to create a downhill sloping response. Others, like musicians can't get over how real a kick drum or upright bass sounds... It's what I enjoy. Adding a monopole subwoofer just removes what I enjoy most. I have one I switched in and back out very quickly.... Is this a waste, I don't know yet. Just don't be quick to dismiss dipole bass because it has a set of challenges. Many good things are challenging imho🙂. If you are throwing dance parties, I'd definitely look elsewhere....
Sent from my iPhone using Tapatalk
As far as dipoles go, I'm a believer in what Linkwitz writes about late reflections with uniform polar response allow the brain to forget about the room. Psychoacoustics backs it up.
Dipole bass is completely different in my experience. It just seems more lifelike. I find that visitors are kind of split WRT LX521 bass. Some people look at the 4 10" drivers and expect it to sound like monopole woofers of equal size and expect room gain to create a downhill sloping response. Others, like musicians can't get over how real a kick drum or upright bass sounds... It's what I enjoy. Adding a monopole subwoofer just removes what I enjoy most. I have one I switched in and back out very quickly.... Is this a waste, I don't know yet. Just don't be quick to dismiss dipole bass because it has a set of challenges. Many good things are challenging imho🙂. If you are throwing dance parties, I'd definitely look elsewhere....
Sent from my iPhone using Tapatalk
I've actually tried it, as have most (if not all) of the people that have responded here. I find it completely impractical below the room's lowest mode, as I stated earlier.
OB is very nice and completely practical down to the 1/4 wavelength of your typical 18 inch baffle. But under about 400 hz or so it becomes increasingly impractical at an exponential rate.
Despite the challenges it's relatively easy to get response flat down to 50 hz at reasonable spl as Linkwitz, MJK and Kreskovsky have shown. But going much lower than that, and especially going lower than the room will support is a waste of money and space.
If you want to throw 4 high excursion 18 inch drivers at the problem and get ~ the same response you would get from a single small regular sub design that's fine. Just don't forget the very real issues of pole vent velocity noise and boundary reflection cancellations.
Having tried dipole subs, I think multiple subs of just about any other type are the way forward. Spreading multiple subs around the room will give just as flat (if not flatter) in room response than a couple of dipole subs and do it with much higher efficiency.
Efficiency?
Hi
I as far as I heard, if you want to have the same output one octave lower you need 4 times the displacement. Is that true??
If it is true, you will need to go from 4 x 10" woofers to 4-5 X 18" woofers to be able to go from 50 Hz to 25 Hz with the same output.
I don't think efficiency is that big an issue in these days. You can get really cheap power amplifiers for PA nowadays.
best regards
uwe
PS I forgot. What about using a W- profile instead of H- or U-frame
Hi
I as far as I heard, if you want to have the same output one octave lower you need 4 times the displacement. Is that true??
If it is true, you will need to go from 4 x 10" woofers to 4-5 X 18" woofers to be able to go from 50 Hz to 25 Hz with the same output.
I don't think efficiency is that big an issue in these days. You can get really cheap power amplifiers for PA nowadays.
best regards
uwe
PS I forgot. What about using a W- profile instead of H- or U-frame
Last edited:
MJK seemed pretty darn happy with his Goldman 18 inch H frames. GR Research claims pretty good subwoofer results with what amounts to a double stack of H Frame servo controlled 12 inch subs from Rythmik, and not just to 25Hz but down into the nether regions. Nelson Pass also has some helpful ideas with his slot loaded subwoofer. Jazzman seemed pretty happy with his Ripoles. The point is there are multiple ways to skin the subwoofer cat and what works best might well be room specific. I've heard some pretty good sealed subwoofers but I have to say the realism of live music seems to come across best with dipoles. Although I haven't actually heard either an infinite baffle or a multi sub swarm set up which no doubt could also be great. Are we having fun yet 🙂
The quality of sound emanating from di-poles, or open baffle bass systems, seems to lack the very fundamentals that constitute "live sound". You are hearing mostly over tones, but there is no "weight" or "heft" to the notes.
Open baffles do not sound "real" to me, at all. Musicians ? I had a band member, most notably a drummer, over to my place to listen, and his comments were that in no way, shape, or form, could an open baffle deliver the kind of sound he had just heard. And for a reference to what he heard, I use ppsl sealed mid bass, and infinite baffle subs.
hello Charlie, it's nice to hear from you. It's seems to me like you are expanding this discussion beyond the topic, maybe ?
From above, " I agree that bass requires large baffles or TL-type structures, but if you transition to a CB system around 100Hz the dimensions of the OB are really not large at all. " So, essentially what you are saying is, that it makes more sense to use a CB (closed box) under 100Hz. Yes ?
With respect to Olson, the laws of physics have not changed in the last 70 years. I have heard this far-reaching argument before, several times. It seems that the modern day stance is, there now exists speaker drivers specifically suited to open baffle bass, and have even heard that s-t-r-e-t-c-h-e-d to the point of the wild claims that "modern day" open baffle drivers don't even need e.q. really ? Never before in all my years as a "speaker builder" have I witnessed the degree of denial, that I have, when it comes to dealing with the open baffle cult. Front-to-back wave cancellation will continue to exists in this world, no matter how many times you say "ALAKAZAM"
MJK's design is fine, it just doesn't go very loud, as my sim showed. It maxes out at about 1 watt.
GR Research isn't breaking any laws of physics. I'm not familiar with their OB sub or their claims but if you simulate it you will find that it's a low sensitivity design that doesn't go very loud.
Pass's OB monstrosity isn't anything special, although he does make some weird claims about it.
Ripole, dipole, U frame, H frame, N frame, V frame are all slight variations on the same concept, that concept being very low sensitivity in the low bass. There's no way around that, and I'm sure if I spent some time researching I could find a thread that had people hating their dipoles for every thread that you can find with people loving them.
I think the drivers for the Carver Amazing OB speakers were build specifically for open baffle use. 12" drivers with very light but solid cones (Styrofoam?), low Fs, high Qts (I think as much as 2.0 on the later versions) and pretty decent Xmax of around 12mm or so. They could be used un-eq'd, on a basically flat baffle.
Yeah, Scott, if you read MJK's work you will see that very high qts drivers can be used to good effect with a very simple passive crossover that amounts to just a small amount of signal shaping to get a flat response. As Brain said, the Carver OB did the same thing.
The big bump at (or just above) fs that you get with these high qts drivers in IB serves to boost the low end to flat(ish) in OB. But that's not free, it comes at the expense of a whole lot of displacement. So whether you use high qts drivers or a whack of eq boost it's generally the same for required displacement. But the former can be done with a simple passive crossover, the latter needs a more complicated active crossover with dsp.
And of course any dipole that plays low bass will require tons more displacement than and other design, whether sealed, ported, bandpass, horn, or anything else.
The big bump at (or just above) fs that you get with these high qts drivers in IB serves to boost the low end to flat(ish) in OB. But that's not free, it comes at the expense of a whole lot of displacement. So whether you use high qts drivers or a whack of eq boost it's generally the same for required displacement. But the former can be done with a simple passive crossover, the latter needs a more complicated active crossover with dsp.
And of course any dipole that plays low bass will require tons more displacement than and other design, whether sealed, ported, bandpass, horn, or anything else.
Hello Brian. touche. you found an example of a woofer made to be used on a flat baffle. I wonder what the crossover of the CAL looked like ?
In the final analysis though, this is a velocity based device and will still not offer
the same performance as what a pressured bass will do.
yes, open baffle sub is just plain stupid idea
That's exactly the point: Obviously, everyone is entitled to their opinion and be able to enjoy what they like, in terms of sound quality. What bothers me the most, is that somehow, at the service of the open baffle craze, this fad has been exhaulted as the new coming, and to be dammed with that nasty ole box. I would most inclined to agree -- a bad box sounds really really bad.
It must be that box, but not the closed woofer methodology. Therein lies the problem. There also exists a mass mis-understanding of how loudspeakers work; most evident in the proposal of an "open baffle sub".
That's exactly the point: Obviously, everyone is entitled to their opinion and be able to enjoy what they like, in terms of sound quality. What bothers me the most, is that somehow, at the service of the open baffle craze, this fad has been exhaulted as the new coming, and to be dammed with that nasty ole box. I would most inclined to agree -- a bad box sounds really really bad.
It must be that box, but not the closed woofer methodology. Therein lies the problem. There also exists a mass mis-understanding of how loudspeakers work; most evident in the proposal of an "open baffle sub".
Thanks Charlie for sharing your experience with OB sub building and listening. I hope you can advise as follows: I have midwoofers that acoustically roll off soon below 70Hz. I have a pair of Rythmik 12" sealed servo subs, which I haven’t had a chance to hear yet. They get very good reviews despite their low cost, but I really don’t want to use them right below my superb sounding midwoofers and my main amps. Instead, I want to build a pair of passive subs and drive them with a standalone power amp, which should have considerably better sound quality than the plate amps in the Rythmiks (eg. Bryston).
At the same time, I am torn between building another pair of sealed subs or instead going with OB subs.
How would, say, a pair of 18” drivers in h-frames, dual 12” or 15” drivers in W-frames or dual 12” or 15” drivers in slot-loaded frames differ in performance?
My room is only slightly larger than about 14 ft x 19, and my priority is sonic realism not high SPLs.
Furthermore, would it be good to use my Rythmik subs blended below the OB subs?
I presume that any coloration from the plate amps would be much less audible by having the Rythmiks playing "behind" the OB subs; that is, only down in the 40Hz to 25Hz region. And using 4 subs should help produce a much more boom-free low end response, with or without doing room electronically and acoustically achieved EQ.
Which OB frame might you recommend for my situation?
My sources are all digital files from my laptop or desktop Windows computers. I have an external DAC with balanced and unbalanced outputs, the latter feeding my 300B SET amps that drive my main speakers.
The Rythmik subs have plate amps with these controls.
http://www.rythmikaudio.com/download/XLR2_sealed_quickguide.pdf
But what solution would you use for low pass filtering to blend the passive OB subs with the midwoofers?
And to adjust delay (phase)? miniDSP?
https://www.minidsp.com/products/opendrc-series/summary-table
http://www.audiovero.de/en/acouratedrc.php
https://www.minidsp.com/products/dirac-series
Dspeaker?
DSPeaker-Anti-Mode 2.0 Dual Core
DSPeaker-Anti-Mode 8033
Or would you use a purely analog solution for bass management?
- Status
- Not open for further replies.