Open Baffle Epiphany!

[kyle]

I have a question that is relevent to all types of speakers. Have you really ever thought that the speakers you're listening to sounded just like real live music?
I've heard a lot of speakers at all kinds of price points and for me the answer is no. It may be better sounding to me or have asound quality that I like but I've never thought, Oh boy, I'm right there in the studio or on stage with the musicians.
Pick your poison, they're all right.

 

[Poultrygeist]

The best way to make a great sounding OB is to duplicate someone else's great sounding OB.

 

[mhardy6647]

What would you consider the most difficult sonic hurdles to overcome...compared to DIY-ing a conventional monopolar box design?

Based on this thread:

Not sure what example is being cited here, misleading as it may be (is this some anti-OB agenda or just ignorance?), but Emerald Physics CS series OB's get down to 20 Hz with a baffle width just under 19". And nothing too exotic going on there....inexpensive pro-audio Eminence driver and regular old flat baffle made of wood with affordable Behringer DSP.

DSP

The route I've taken, after experimenting with a number of different drivers and designs, is to stack identical multiple drivers. I do this for mains using 8" drivers, which I run full range and for bass.

For bass, here is a economical driver that works very well OB....

http://www.parts-express.com/goldwood-gw-410d-10-poly-dvc-subwoofer--290-362

A set of 4 of these per side on a 20" or so wide board do a great job down low. How high you run them depends on what you use for mains. They seem to do fine up to 250-300hz or so. The lower the cross, the better I think. To keep the crossover out of the way, not necessarily because of the driver.

$80 a side plus a 20" wide by 48" tall piece of wood seems economical to me. Then you can do mains seperately.

For mains, I've tried the replacement BOFU drivers, didn't work.

http://www.parts-express.com/grs-8fr-8-full-range-8-speaker-pioneer-type-b20fu20-51fw--292-430

To much of a cheap cardboard sound for me. But at about $50 a side for 4 drivers, I had to try due to all the buzz about the original BUFO's.

I tried those, because I kind of liked first these....
http://www.parts-express.com/goldwood-gw-8003-8-8-full-range-driver-w-whizzer--290-379

Again, 4 in a stack on about 20" width and for $80 a side, you make nice music. For the money this are pretty amazing. They do very good bass.....good enough some might not miss if you didn't bother with the bass stack, but they do lack a bit at highs. Most would want to augment these up above about 14k with something. I'd suggest a small rear facing horn driver, to get the efficiency high enough, and roll it in with a simple cap.

Ultimately, what I've done for mains is go to the Tang Band 1808's, which miss on your request for economical. They do the highs nice and do pretty good almost full range but once you augment the lows with the woofer stack, you don't give up much to many speakers to a pretty darn high price point!

The reason to do a stack of drivers, both mains and woofs, is the dynamics and the lower distortion. You listen to a single driver and it's ok. Do two and you readily hear less distortion. Go to four and you about wouldn't know your listening to the same driver as the single. And the dynamics...they just pop. This is what is missing from many of the single driver configurations I've heard running OB. Stack 'em and you get a really amazing sound for such simple construction!

All of these drivers I've mentioned get you enough efficiency to use about any amp power including very low power ones. You just have to work at setting up for a reasonable load. While I've avoided that as a problem with a current source amp, I've also tried running with a number of different voltage source amps including tube.

OB is amazing...I don't see me ever going back!

EV3

Lots of drivers.

Yes Zu and Tekton use 10 inch Eminence drivers. I had a pair of Zu Omens but sold them as every pair of OB's I've built with bass support sounded better. I liked the Zu's but couldn't justify keeping them. Their foot print was larger than my OB's and twice as heavy.

As Linkwitz says "it's hard to screw up an open baffle speaker to where it sounds worse than a box speaker".

"Bass support"

Pics, as requested:


Front: With KR-5200 and the obligatory cat...

Back: I think the offset of the driver and the multiple holes provide a nice effect. The bass dipole cancellation is spread out over a wider band. The sandbags were installed to dampen the relatively thin enclosure. I also made a little pillow under the crossover out of some quilt padding.

Sandbags and pillows

I'm not a fan of the complexity and negative impact of passive crossovers in the signal path of multi-driver open baffle speakers.

I use an inexpensive active crossover which allow use of one amp for the HF and one amp for the LF but the huge advantage comes from the fact that the signal is altered before it is amplified rather than after. I've tried passives vs actives and the difference is significant. With all the available information on this topic it's not debatable which is better.

ultumax, PM me.

Active crossovers.

http://www.audiokarma.org/forums/showthread.php?t=442092

You can build these without the H-frame subs (but you will need some sort of subwoofer to cover the low end) and just stick a cap before the tweeter and not need bi-amplification as the Beta can run full range. I tried it first that way but ultimately went to a dedicated amp on the tweeter and dedicated amp on the Beta (and plate amps on the H-frames) with an Ashly active crossover so I could have adjustable parameters for all the drivers. You can tailor where you want the tweeter to come in with different cap values, but you won't have control over it's output and unless your tweeter's sensitivity is right there with the Beta's it's critical to have that adjustment. Anything that gives you on-the-fly level adjustment here makes matching the output of two drivers with different sensitivity a snap.

But to just get a taste of OB without the fine tuning I say get a pair of the Betas, use those tweeters you have with a cap to tailor where they come in (I had BG Neo 3s coming in around 3.5K) and let her rip. If you like where it's going you can pursue a proper build. And I HIGHLY...let me say that again....HIGHLY recommend building the H-frame subwoofers. You will never look at a box subwoofer the same again. It will be every bit the revelation the discovery of OB is.

- Michael

H-frame subs (a flavor of bass support, I suppose).

I've built Poultrygeist's 3-way OB's that use the Beta and yes, a phase plug is used to replace the dust cap.

Phase plugs.

... that's from the first 10 pages.

I was thinking a hunk of plywood sized to provide an f3 around the desired cutoff, a high-ish Qts coax or extended range driver, and some wire.

 

[MJKing]

.... But the rear wave in a boxed speaker is not isolated either. It can import energy into the cabinet or reflected back through the driver cone. There are problems with this arrangement also. A vented enclosure has timing issues with the front wave and port output. ......

If the box is of decent construction I don't believe it should enter the equation, box vibration should not be an issue.

I have seen the statement that sound is "reflected back through the driver cone" a few times before on forums and I don't believe this is the case with a quality driver. Standing wave resonances in the cabinet can exert pressure on the back of the cone which will result in a drop in output over a very narrow band of frequencies, but I have a hard time thinking of music inside the cabinet coming back through a "transparent" cone over a wide band of frequencies, it does not make sense. I have never seen this show up in a measurement of a speaker system.

I also do not understand your statement about timing issues in a bass reflex design. The port only produces significant output at the cabinet tuning frequency, that is the whole point. When the port is producing the bass the driver output is seriously attenuated because it is essentially standing still. The systems behavior is predictable and timing issues or errors should not exist.

 

[jhoyt]

Well, I left the system playing while we were gone this evening. After getting home, I find that the sound has opened up more. There's not as much congestion. I think the drivers were just still stiff from sitting for ages. Bass has opened up as well, at least with the music I'm playing (Brian Bromberg).

In regard to placing the speakers in the corners, I was trying to get more bass by loading them. If I have some free time tomorrow I'll have to do a comparison between them as presently situated, and pulled out some. I can't pull them out much, though, because of furniture/decor placement.

This is a great fun experiment, and so far it's just cost me two cardboard boxes. :thmbsp:

 

[dnewma04]

If the box is of decent construction I don't believe it should enter the equation, box vibration should not be an issue.

I have seen the statement that sound is "reflected back through the driver cone" a few times before on forums and I don't believe this is the case with a quality driver. Standing wave resonances in the cabinet can exert pressure on the back of the cone which will result in a drop in output over a very narrow band of frequencies, but I have a hard time thinking of music inside the cabinet coming back through a "transparent" cone over a wide band of frequencies, it does not make sense. I have never seen this show up in a measurement of a speaker system.

I also do not understand your statement about timing issues in a bass reflex design. The port only produces significant output at the cabinet tuning frequency, that is the whole point. When the port is producing the bass the driver output is seriously attenuated because it is essentially standing still. The systems behavior is predictable and timing issues or errors should not exist.

I'm with you 100% on every bit of this. I think we audiophiles try to invent solutions to problems that don't necessarily exist. I'm as guilty as the next, especially in the DIY realm, but I usually try to make it clear that it's just a guess that hasn't been verified.

 

[dnewma04]

I have a question that is relevent to all types of speakers. Have you really ever thought that the speakers you're listening to sounded just like real live music?
I've heard a lot of speakers at all kinds of price points and for me the answer is no. It may be better sounding to me or have asound quality that I like but I've never thought, Oh boy, I'm right there in the studio or on stage with the musicians.
Pick your poison, they're all right.

I've heard a few do a credible job. Every one of them was big and extremely dynamic. I've never heard any small 2 way or 3 way, or even an inefficient speaker sound "live". For example, one was a Unity Horn based system with dual Lambda TD15X drivers. Big, efficient, low distortion, ridiculously low power compression, massive power handling, etc.

 

[dnewma04]

I would agree if you are just picking drivers, sticking them in a convenient baffle, and then dicking around with the crossover trying to tune by ear. Many things can bite you.

But if you select appropriate drivers, measure their T/S parameters, and then run simulations to design the system (baffle and crossover) then it is not that hard to get a really great sounding result. In other words engineer the design, don't wing it. Designing an OB is much easier than designing a two way boxed speaker.

The effort you put in is easily heard in the final speaker system.

I'm with you on everything but the bit about it being easier to design than a two way boxed speaker. I think the same engineering principles can be employed to make a 2 way boxed speaker easy to build with very predictable real world performance. For me, since I've build more boxed speakers, it would be an easier task to pull off successfully. I don't doubt for a minute that your experience with OBs allows you to build one more successfully than me.

Now, if we are talking about BLHs, FLHs, TQWTs, TLs and all the things your old mathcad files make easier to predict, then by all means, an OB would be easier to pull off if for no other reason than the construction details. However, a small braced 2 way sealed/ported box can but cut out and glued up quickly and easily.

 

[dnewma04]

These points about a standard room free of treatments as adequate if not preferable as a listening environment is interesting. In some ways almost heresy to the grizzled audiophile. At a certain point I would want to believe it, but at a basic level this feels inherently wrong. Got any more information or link to why this would be the case?

Regards
Mister Pig

I detailed some of this in my response to the individual comments, but the main point is that above a certain frequency (usually around 200hz+) the way the room interacts with the sound is fairly predictable and more often than not, a typical room will tend to behave similarly. When SL mentions the sound of the room, he is referring to the reflected material. The early those reflections get to your ears, the more potential trouble they can impart. The worst situation you can have from a reflection standpoint is an early reflection arriving at your ear from the same side as the source. IOW, the sound of the right speaker bouncing off the right wall and arriving at your right ear. What's even worse is when the off axis sound is significantly different than the direct sound and in the typical 2 way system with a 5-6" woofer and a dome tweeter, the high frequency response will beam. What you hear off axis will have a significantly different sonic character than the on axis content.

For instance, here is the off axis response for one of my favorite tweeters, a hiquphon OW2.

As you can see, once you get outside of 15 degrees off axis, the sound will change significantly, so an uncontrolled early reflection can change the sound of the speaker.

With waveguides, you can create a situation where the reflected material has the same sound as the direct material and using tricks like significant toe in can virtually eliminate early reflections, promote later reflections and those reflections can have the same sonic character as the direct content. The result is a significant enhancement in the perception imaging and space.

With a dipole, the nulls at the sides tend to reduce the importance of the off axis content and even though there may be more overall reflected material, it's more consistently the same sounding material and reduces the overall rooms impact.

Next to constant directivity systems, dipoles rank as one of my favorites in this regard.

 

[tomlinmgt]

I didn't really get the answer I was looking for when I asked what are the sonic hurdles that are difficult to overcome in an OB design (MH primarily gave hardware items). So Dave, since you also see OB's as a greater challenge to get right than a box design, what is it about a box design that makes you more confident about being able to execute a successful build....besides having more experience with the type? Or worded another way...what concerns are you faced with at the initial stages of planning an OB design that you don't have to deal with when planning and executing a box design? I get the challenge of reproducing content from the first few octaves with OB, but are there conditions unique to OB's when it comes to reproducing content from say 150 Hz and up? If you understand the performance characteristics of a driver based on its TS parameters can everything you apply to a box design not apply to an OB design as well? I mean, do you really rely on anything from the box when it comes to mid/upper bass, mid range and HF reproduction? The enclosure's contribution should be fairly inert....right?

 

[dnewma04]

Getting the bass right is the biggest hurdle for me both in extension and perceived detail of the bass. Crossover requirements are similar, construction requirements are similar (I tend to overbuild both). The things Mark listed wouldn't be hurdles, necessarily for me. I embrace the right tools for the job and the first time you ever hear me embrace a single design philosophy like "I only use first order crossovers!" or "I need to have tubes in my signal path!" you have my permission to kick me square in the nuts.

When I say getting the bass right, I'm referring to getting great extension, detail, tone, and dynamics. I've yet to get great extension sans EQ, I've heard some good detail but no better than other options I've used. Tonally they are fine and they do have a slight advantage. Dynamics is where the 15A didn't make the cut for me but the AE dipole handled things significantly better.

I'm in agreement with the general principle that using a full range driver that performs well and tossing it in a board with a 15A can lead to some suprisingly positive results, but often times, I think I'm just expecting them to perform badly and just getting something acceptable is a win.

It's getting too late to make any sense, I think.

 

[Mister Pig]

If the box is of decent construction I don't believe it should enter the equation, box vibration should not be an issue.

I have seen the statement that sound is "reflected back through the driver cone" a few times before on forums and I don't believe this is the case with a quality driver. Standing wave resonances in the cabinet can exert pressure on the back of the cone which will result in a drop in output over a very narrow band of frequencies, but I have a hard time thinking of music inside the cabinet coming back through a "transparent" cone over a wide band of frequencies, it does not make sense. I have never seen this show up in a measurement of a speaker system.

I also do not understand your statement about timing issues in a bass reflex design. The port only produces significant output at the cabinet tuning frequency, that is the whole point. When the port is producing the bass the driver output is seriously attenuated because it is essentially standing still. The systems behavior is predictable and timing issues or errors should not exist.

I m just a simple pig and hobbyist, so my understanding of these principals is pretty much a layman....ok laypig terms. I had come across this passage regarding vented bass that describes my understanding of the trade offs of vented bass designs. Now I certainly may have a limited understanding, and perhaps these issues are not as profound as this passage makes them be, and I am always willing to learn more on the subject than I know...which probably wouldnt take long.

When this vent air mass/box air springiness resonance is so chosen as to lie lower in frequency than the natural resonance frequency of the bass driver, an interesting phenomenon happens: the backwave of the bass driver sound emission is inverted in polarity for the frequency range between the two resonances. Since the backwave is already in opposite polarity with the front wave, this inversion brings the two emissions in phase (although the vent emission is lagging by one wave period) and therefore they reinforce each other. This has the useful purpose of producing higher output (for any given driver excursion compared to a closed box) or, conversely, a similar output with a smaller excursion (which means less driver distortion). The penalty incurred for this reinforcement is time smearing: in essence the vent resonance augments main driver output by imposing a "resonant tail" on it. For frequencies above the natural resonance of the driver, the reflex alignment has no influence.

In terms of cabinet designs and materials, over the years there seems to be many philosophies over what works the best, and what is practical, cost effective, and even desirable in a commercial speaker. Rogers developed the light mass and quick dispersion design with the LS3/5A as one method. The rigid and high mass cabinet that raises the resonant point of the cabinet, and then using dampening materials to deal with the remaining resonances is another. Companies like Green Mountain Audio use synthetic cast cabinets to do this, as did Mourdant Short, and some companies used wood and sand combinations, while others like YG Acoustics use aluminum or aluminum and wood, other companies concrete. Some companies such Mission Audio wanted their cabinets to resonate a bit and fill the bottom end in, and Micheal Green speakers had rods that went through the cabinets allowing the owner to allow more or less cabinet resonance to "tune" the speaker. The basic point is that they are trying to find ways to minimize or control cabinet resonances, and since they have gone to such exhaustive means....they feel that this issue is salient to the overall performance of their speaker design.

Now I do not claim to be any kind of deep thinker, or grizzled speaker builder. But I have heard a decent amount of speakers, and spent my time trying to hear alternatives in design that are friendly to lower power amplifier designs. From what I have experienced, OB speakers seem to have defining characteristics that neither vented or acoustic suspension speakers have. Of course they deal with the back wave differently, and that may be the majority of the difference. But both acoustic suspension and vented designs have to take into account that back wave, and utilize or minimize the effects of that energy. Each implementation will sound different it seems to me....But like I say I am just a hobbyist, and so my understanding of these issues mat very likely be incomplete.

Regards
Mister Pig

 

[Mister Pig]

I detailed some of this in my response to the individual comments, but the main point is that above a certain frequency (usually around 200hz+) the way the room interacts with the sound is fairly predictable and more often than not, a typical room will tend to behave similarly. When SL mentions the sound of the room, he is referring to the reflected material. The early those reflections get to your ears, the more potential trouble they can impart. The worst situation you can have from a reflection standpoint is an early reflection arriving at your ear from the same side as the source. IOW, the sound of the right speaker bouncing off the right wall and arriving at your right ear. What's even worse is when the off axis sound is significantly different than the direct sound and in the typical 2 way system with a 5-6" woofer and a dome tweeter, the high frequency response will beam. What you hear off axis will have a significantly different sonic character than the on axis content.

For instance, here is the off axis response for one of my favorite tweeters, a hiquphon OW2.

As you can see, once you get outside of 15 degrees off axis, the sound will change significantly, so an uncontrolled early reflection can change the sound of the speaker.

With waveguides, you can create a situation where the reflected material has the same sound as the direct material and using tricks like significant toe in can virtually eliminate early reflections, promote later reflections and those reflections can have the same sonic character as the direct content. The result is a significant enhancement in the perception imaging and space.

With a dipole, the nulls at the sides tend to reduce the importance of the off axis content and even though there may be more overall reflected material, it's more consistently the same sounding material and reduces the overall rooms impact.

Next to constant directivity systems, dipoles rank as one of my favorites in this regard.

So what we are talking about are the known issues of room acoustics, and the problems created by the first reflection points and how they create timing issues with the direct radiated energy from the speaker. This being an issue with many typical drivers, especially domes.

So your preferred solution is CD drivers, and then waveguides, which will minimize the early reflection issues, and provide controlled radiation patterns towards the listening area. The trade off being a smaller sweet spot for listening.

I have spent several years using the larger sized full range drivers, and they tend to have controlled radiation patterns when compared to a standard dome and cone speaker, so I have experienced some of what you are describing. But the speakers have other limitations that tend to grate on me over time, and are not a long term solution at least for me. i do remember spending several months with some VMPS 626 JR speakers that Brian Cheney had implemented a constant directivity wave guide cover for his speakers. Personally I didn't care for what it did to his speakers, and liked them better with the panels taken off. Now DIY Sound Group has a pretty large selection of waveguide horns from SEOS and Auto Tech that look interesting. Maybe one day a DIY project with them and an H frame woofer package would be worth exploring.

Regards
Mister Pig

 

[Poultrygeist]

The Emerald Physics Nemesis with four Alpha 15A's.

These are pretty much in keeping with Linkwitz's conclusion that "for extreme SPL requirements the number of drivers could get very large and therefore below 50 hz they are more economically replaced by sealed box subwoofers".

I've used double Alphas per side which is a gracious plenty in a 15X15 sized room.

 

[Poultrygeist]

I would agree if you are just picking drivers, sticking them in a convenient baffle, and then dicking around with the crossover trying to tune by ear. Many things can bite you.

But if you select appropriate drivers, measure their T/S parameters, and then run simulations to design the system (baffle and crossover) then it is not that hard to get a really great sounding result. In other words engineer the design, don't wing it. Designing an OB is much easier than designing a two way boxed speaker.

The effort you put in is easily heard in the final speaker system.

I heard these at Horn Fest and was blown away. It's obvious that baffle design had nothing to do with their sound. As long as the bass is augmented by pro woofers I have a hard time hearing differences between baffle sizes.

 

[MJKing]

When this vent air mass/box air springiness resonance is so chosen as to lie lower in frequency than the natural resonance frequency of the bass driver, an interesting phenomenon happens: the backwave of the bass driver sound emission is inverted in polarity for the frequency range between the two resonances. Since the backwave is already in opposite polarity with the front wave, this inversion brings the two emissions in phase (although the vent emission is lagging by one wave period) and therefore they reinforce each other. This has the useful purpose of producing higher output (for any given driver excursion compared to a closed box) or, conversely, a similar output with a smaller excursion (which means less driver distortion). The penalty incurred for this reinforcement is time smearing: in essence the vent resonance augments main driver output by imposing a "resonant tail" on it. For frequencies above the natural resonance of the driver, the reflex alignment has no influence.

That is some premium techno babble. My guess is the person who wrote that did not do the math involved in designing a bass reflex speaker. Don't blindly believe everything you read on the Internet.

 

[evsentry3]

The Emerald Physics Nemesis with four Alpha 15A's.

These are pretty much in keeping with Linkwitz's conclusion that "for extreme SPL requirements the number of drivers could get very large and therefore below 50 hz they are more economically replaced by sealed box subwoofers".

I've used double Alphas per side which is a gracious plenty in a 15X15 sized room.

Oh I remember these! Maybe 4-5 years ago at RMAF. Strong OB sound qualities. Hearing these may have been one of my bigger steps to the conclusion that OB could go above a simple single flat board!

They were too big for the room. Held back just a bit by a laptop source into a Behringer crossover...but POWWW, they showed potential!

All you need is a big enough room and a very forgiving and understanding wife!

EV3

 

[evsentry3]

That is some premium techno babble. My guess is the person who wrote that did not do the math involved in designing a bass reflex speaker. Don't blindly believe everything you read on the Internet.

I hope you don't have a TM on that phrase! Premium techno babble. A lot of that around IMHO. A great term for when all the flowery white paper conclusions just don't at all match the result of pressing the play button!

EV3

 

[tomlinmgt]

I see some designs use front mounting on the baffle, others use rear. What determines which is optimal?

 

[dnewma04]

Are you referring to how the woofers are mounted to the baffle?