Dipole and Uframe models and discussion re' Live Edge Dipoles

[perrymarshall]

@nc535 sounds like a fantastic system, and a 5208C Beryllium paired with a 15OB350 with 400ish xover will be an outstanding combination.

 

[dayneger]

With a coax driver, baffle step for the tweeter is eliminated because the midwoofer cone acts as a waveguide. For the midwoofer, I have minimized the baffle width to little more than needed to support the driver, so it operates below the baffle step frequency. With separate midrange and tweeter, the same considerations drive you away from a baffle.

I seem to be a slow study on this aspect. With the nude coax, would you be able to EQ out the frequency response, possibly with a shift of XO to the woofer if needed (and possible)?

Or if you were willing, perhaps you could elaborate on this some more. I read Charlie's articles and some other content but am still digesting the relationships between all the variables--waveguides, baffle sizes and shapes (and thicknesses and edge treatments), driver sizes, dipole peaks, baffle step, impacts on directivity, phase, timing, impulse response, etc.... lots of conditional aspects all still in a muddle.

 

[tmuikku]

Hi, relationship is the sound wavelength. If you want nice response from a transducer in dipole config, you'd want it below diffraction problems start, which depends on size of the baffle, diffraction issues are above where wavelength and construct size relate. No matter which sized transducer you have, bigger one moves things to longer wavelength (lower in frequency), smaller to shorter (higher frequency). Roughly, doameter of transducer marks it good bandwidth high end limit.

Now, as output drops with every octave down there is about few octaves of usable bandwith from baffle sized wavelength down, if you want to maintain some SPL capability.

You could bend this ideal if you just go to diffraction and don't mind about lack of bass, you could use just one transducer in "fullrange". Usually maybe three or four different sized transducers are utilized to get full bandwidth system without much issues, transducers all working their optimal bandwidth.

Wavelength slides from 10Hz/17m to 20kHz/1.7cm, and as out physical object sizes stay constant, we must use various sized objects to maintain some "control" or fidelity, or howd you want to think about it. Static sized object works nicely for small bandwidth, until wavelength gets too long or too short in relation.

 

[dayneger]

My driver lineup is 5208C, 150B350 dipole (not U-frame), BM-11 shallow mount sealed sub. When the design settles down, I expect it to be almost a foot shorter than the Bitches Brew, half the width at top, and to have depth only at the lowest 10".

That does sound like a pretty cool combination!

I'm not familiar with the BM-11. Impressive to have a 14 mm one way xmax with just over a 3" mounting depth. The choice was driven by the desire to minimize the enclosure volume at the bottom (the bass base )?

 

[Baffless]

I also think that 80 Hz is a good crossover point to a subwoofer, or perhaps a tad lower. The basis for this is a paper by Sean Olive that I mention in my aX article:

Included in the research that Olive presents are the preferred in-room (e.g. steady-stage) response curves for a variety of listener types and loudspeakers, all on one plot. This paper is not easy to get, so instead I suggest that anyone interested download this other paper by Toole in which he cites that work and presents the same data along with some other related info in Figure 14:

Please download the article, see Figure 14, and read the accompanying text in section 4.1. The preferred responses all show an increase around ~100Hz of between 3dB and 6dB. This "bass bump" is what the listeners preferred in the steady-state in room response.

I relate the data of Figure 14 to OB/dipole systems in the following way:
A true dipole system has constant directivity, and therefore also has a constant power response.
The in-room steady-state response reported by Olive and Toole will be very similar to the power response of the loudspeaker and is not flat!
This is the same result you see time and time again from Klippel testing of loudspeaker e.g. see Erins Audio Corner, reported as the estimated in-room response.
Therefore some equalization/correction of the dipole should be made to better mimic this response. Above 100Hz I do this via EQ of the input signal to "tilt" it down.
Below 100Hz we can obtain the "bass bump" for free just by crossing over to a monopole. This is because a monopole radiates 4.75dB more power into the room than a dipole for the same on-axis SPL.
So I just cross over around 80Hz to a sealed sub. The power response will naturally rise through the crossover point by almost 5dB, exactly like the preferred listening curves from Olive. The monopole sub can also deliver a better low bass response and room pressurization at very low frequency.
This also greatly reduces the demand on the dipole woofer, which now only needs to work well down to 80Hz. By reducing the LF requirements the dipole loudspeaker can be made physically smaller, or use a smaller woofer(s), etc.

thanks, very good link. But i try to read out how subjective preferred curves are derived? by listening to various speakers with different DI? or altering room properties? yers ago i was reading some toole paper and read that they listened to different DI loudspeakers. But i didn't see any flat DI speakers there. So it seems not that scientific approach covering all. or i am missing something?

 

[nc535]

minimize the footprint

I seem to be a slow study on this aspect. With the nude coax, would you be able to EQ out the frequency response, possibly with a shift of XO to the woofer if needed (and possible)?

Or if you were willing, perhaps you could elaborate on this some more. I read Charlie's articles and some other content but am still digesting the relationships between all the variables--waveguides, baffle sizes and shapes (and thicknesses and edge treatments), driver sizes, dipole peaks, baffle step, impacts on directivity, phase, timing, impulse response, etc.... lots of conditional aspects all still in a muddle.

A nude coax is little different that a coax on a small or narrow baffle. The OD of the driver's frame defines the minimum baffle. My narrow baffle is less than 50% larger in width than that minimum diameter. Its just a question of degree; EQ can deal with it. The max SPL is a little bit lower, the frequency range over which you get the desired directivity extends a bit higher. The choice has more to do with aesthetics or preferred construction methodology than anything else.

In terms of comprehending the relationships between all the variables - that is what simulation is for. To tell the truth, my head is still spinning from all the variations I've run trying to optimize the 5208's response and it will take me a little time for me to sort it out.

 

[dayneger]

A nude coax is little different that a coax on a small or narrow baffle. The OD of the driver's frame defines the minimum baffle. My narrow baffle is less than 50% larger in width than that minimum diameter. Its just a question of degree; EQ can deal with it. The max SPL is a little bit lower, the frequency range over which you get the desired directivity extends a bit higher. The choice has more to do with aesthetics or preferred construction methodology than anything else.

Thanks @nc535, that helps a lot. What happens as you go from the horizontal to the vertical direction (baffle height increasing)? In any case, a baffle that narrows to toward the top sounds like a good fit for my particular "multifunctional listening space" (aka living/dining room ).

I can imagine it being a serious challenge simulating something with so many variables! It's impressive witnessing you tackle it.

Out of curiosity, what makes the 8" Radian 5308c your preferred choice vs the 10" 5210? Are you expecting to drive them separately from the SB 15OB350 or in series analogous to what @perrymarshall has implemented in different combinations?

 

[gilbodavid]

Great thread. I’ve never built a speaker before, but am reading a lot, collecting drivers and really relieved that you are also confused about the shape.

https://getmaco.macoacoustic.com have an interesting design, with no hole between the 2 drivers, but a horizontal slit between them that doesn’t open at the front. Maybe that is worth checking. Nice design as well, with waveguide I think.

Cheers David

 

[perrymarshall]

The Maco is a beautiful zen-like design and I love the stone baffle. But I think a Constant Directivity coaxial, active with DSP, is far superior to a large beamy full range with whizzer cone.

I used to design OEM car speakers for Jensen (Acura, Honda, Jeep, Chrysler, Ford) and while dual cone designs are admirable for their simplicity, the resolution you’ll get from a good compression driver far exceeds any large full range, and the CD gives fantastic imaging everywhere in the room.

 

[nc535]

Thanks @nc535, that helps a lot. What happens as you go from the horizontal to the vertical direction (baffle height increasing)? In any case, a baffle that narrows to toward the top sounds like a good fit for my particular "multifunctional listening space" (aka living/dining room ).

I can imagine it being a serious challenge simulating something with so many variables! It's impressive witnessing you tackle it.

Out of curiosity, what makes the 8" Radian 5308c your preferred choice vs the 10" 5210? Are you expecting to drive them separately from the SB 15OB350 or in series analogous to what @perrymarshall has implemented in different combinations?

I instinctively tapered the baffle thinking its width at any point needed to be not to much greater than the driver positioned there, if any. One could posit that there is a width that is optimal for the driver's intended frequency range. I haven't been looking for an optimum but simply for one that is good enough - that allows design goals to be met. Aesthetically, I don't want the baffle any wider than needed to meet acoustic and maxSPL goals.

I think 8" is the right size for this application. Smaller and it might not be able to reach down to the 15" woofer without strain. Larger and its directivity will widen more as it approaches crossover to the CD. Furthermore, the data sheet frequency response of the 8" Radian is smoother; less equalization is needed for it.

My very first attempts had very little baffle across the top of the coax driver. I found I needed to provide more space between the rim of the driver and the top edge of the baffle to prevent the vertical dipole null from being tilted forward. In my latest iteration, 36 mm seems to work. The prior art that I reviewed that had zero baffle around the entire top half of its coax driver seems like a bad idea from this perspective.

In a conventional baffle with no extra holes, no dipole cancellation can occur on the floor side of drivers. With the holes in the baffle I showed earlier the vertical directivity is definitely improved by holes to allow that cancellation. The attached drawing shows the difference. The differences aren't great and I am able to equalize them away in the system simulation but I had to work harder on the manual equalization for the no holes in baffle version. The end result was just as good until the 2nd decimal place of Vituix's preference rating.

 

[dayneger]

I think 8" is the right size for this application. Smaller and it might not be able to reach down to the 15" woofer without strain. Larger and its directivity will widen more as it approaches crossover to the CD. Furthermore, the data sheet frequency response of the 8" Radian is smoother; less equalization is needed for it.

I was thinking about the XO to the 15" and remembered that @perrymarshall used 200 Hz to the 18" on the Live Edge unless being played at mind-blowing levels, at which point he suggested crossing a bit higher to get the load off the 8" Radian.

Were you considering the XO around 350 Hz due to less support from the narrower baffle? Or was it that without wings on the side of the 15" SB you don't have tunnel resonance issues to be concerned with, and there were other advantages to running the 15" higher? You mentioned something about the floor null?

Last night I was doing some casual reading and was very surprised to be reminded of the fact that human speech has fundamentals between about 85-155 Hz for men and 165-255 Hz for women... way way lower than I'd have imagined, with the highest soprano barely cracking 1000 Hz. Of course there are lots of harmonics in our voices. Interesting to see how much the bass driver actually covers.

 

[nc535]

My latest sims give a good idea of how low the 5208 can reach with the narrow baffle I'm providing it. The answer of course depends on how loud you want to play. The SPL is only 6 db lower at 100 Hz than at 400 Hz .
.

Those are the SPL levels at 2.83V pk, 1m. We need to look at excursion and driver power limits before reaching a conclusion.

I next created a simulation in which I equalized the 5208 for an XO at 100 hz with the source cranked up to 110 db SPL

I then opened a Vituix Enclosure simulation of the 5208 and transferred the filters and source voltage from the above simulation to the enclosure simulation.

The 5208 can't reach down to 100 Hz open baffle at such a high SPL. In the top right curve, you can see its crossing Xmax at around 270 Hz. We have to add a crossover high pass filter to see where the limit is. I next applied LR24 high pass filters at 200, 300, and 400 Hz. You can read the maximum excursions on the graphs.

first 200 Hz

300Hz

400Hz

The XO is a tradoff between the range over which we have a point source and managing the depth of the floor reflection null. Center to center distance also matters and i'm pushing the limit at 400 Hz. The Vertical polar map shows a >20 degree vertical window so it should be OK.

Clearly this driver in OB, can be used down to 200 Hz. At 400 Hz, its not the least bit strained.

Here are the floor nulls for those crossover points. These are taken from my Vituix system simulation. Its EQed optimally for 400 Hz. I did no re-EQ for the other XO frequencies, should changed the high pass and low pass filter frequencies. The orange trace is the in-room response. What I mean by the floor null is the first dip in the orange trace. The remaining dips are comb filtering with the floor, which likely won't show up in an in-room measurement. The miniscule dip at 400 Hz speaks for itself.

200Hz

300Hz

400 Hz

 

[mark100]

Hey guys, don't want to hijack the thread, or go into off topic too deep.
The OB posts have me thinking about an unusual project...a combination OB and unity/synergy.

One of my favorite builds, showing the back side is the is this guy.....


The four 4" mids are already "open baffle" in that they radiate backwards openly, albeit at a 45 degree angle. (horn is 90 degree H)
I'm thinking about putting two 15" drivers on the horn walls, butting up close to the mids, and being open back just like the mids.
What do you think?

 

[nc535]

Iirc, that large horn controls directives down to 300hz or so. If so, leaving the 4ndf34s open won't create dipole-like directivity. Might add some ambience but that should include HF. You probably could get a dipole effect from the 15s, plus save some weight. In a small horn, ob mids might well allow the directivity to be controlled to a lower frequency

One thing I would like to do if I ever finish this project is to model one of your syn_×' and see what the polars look like.

 

[CinnamonRolls]

You can investigate this design too.



https://www.diyaudio.com/community/...-design-the-practical-way.354772/post-7517489

 

[nc535]

That one we sort of know to be good, if not the details since it came out of ATH. Only concern is how much damage is done to the polars by the mid taps.

A simple conical like Mark builds, and I built about 7 yrs ago, has poor polars by comparison. I think when he sees them he'll start planning on incorporating a secondary flare in the basic structure, instead of as an add-on.

 

[mark100]

Iirc, that large horn controls directives down to 300hz or so. If so, leaving the 4ndf34s open won't create dipole-like directivity. Might add some ambience but that should include HF. You probably could get a dipole effect from the 15s, plus save some weight. In a small horn, ob mids might well allow the directivity to be controlled to a lower frequency

One thing I would like to do if I ever finish this project is to model one of your syn_×' and see what the polars look like.

Thx Jack, i've been figuring the minor backwards radiation of the small mids is almost immaterial...they only cover about 300-900Hz.
I'm wondering about the impact of the 15's. I can open up the top 12", but not the bottom.
And then, i wonder if some rear firing >1000Hz driver would be in order to get deeper into OB territory. (CD/horn range)

A simple conical like Mark builds, and I built about 7 yrs ago, has poor polars by comparison. I think when he sees them he'll start planning on incorporating a secondary flare in the basic structure, instead of as an add-on.

I've incorporated secondary flares in the basic structure on at least three synergy versions.
Two of the most refined had removeable secondary's, so I could measure and listen compare
Measurements get better with the secondarys

That said, and maybe I'm deaf...but i think we might be swallowing the blue pill when accepting the current craze on polars, especially polars beyond listening window

 

[dayneger]

The XO is a tradoff between the range over which we have a point source and managing the depth of the floor reflection null. Center to center distance also matters and i'm pushing the limit at 400 Hz. The Vertical polar map shows a >20 degree vertical window so it should be OK.

Cool to see the analysis. So, with @perrymarshall's Live Edge I believe he's crossing over at 200 Hz with a... hmmm, not sure which slope, the BB were super shallow but much bigger coax... is he getting a lot more support from the wider baffle and EQing out the floor reflection null (but possibly with less lateral directivity due to the wider baffle?)? Or is something else going on?

What exactly does the center to center distance mean... linear vertical distance between the drivers, with 400 Hz starting to get high enough for vertical directivity to suffer?

Sorry for all the questions! I'm finding this super interesting but the sheer number of variable relationships is still hard to get a grip on.

 

[pelanj]

My two prototypes have open backs for mids and LF. Yellow one is 2x 3" and 1x18" at the bottom, the green one 4x 3" and 2x 12". It is because I was too lazy to build the box around them - and I quite like the open backs. Even the planar tweeter can be open backed. I would say there is added ambience due to the open backs. I also equalized the bottom end so both play down to 30 Hz at my home listening volume.

 

[nc535]

I've incorporated secondary flares in the basic structure on at least three synergy versions.
Two of the most refined had removeable secondary's, so I could measure and listen compare
Measurements get better with the secondarys

That said, and maybe I'm deaf...but i think we might be swallowing the blue pill when accepting the current craze on polars, especially polars beyond listening window

There is nothing wrong with "perfect" polars. I don't think its a craze; it's a mountain that just became climbable so of course we are climbing it.

The question is whether or not perfect polars are necessary. Its not so much whether our ears can discriminate between the sounds of perfect/imperfect polars but whether or not differences are so great as to affect our enjoyment of the music. I do know that one can get the sound of a simple conical equalized quite enjoyably within a listening window. the more controlled the directivity, the wider that listening window can be.

polars of a simple, axisymmetric conical can be produced in HornResp quite easily and I just happen to have a model around. The DI has wild variations. Comparing on axis with 20 degrees off axis shows a 5db difference. This is a potential problem for me in this design as well because the coax's mid cone is a conical. Hopefully the baffle tames this to some extent and i can get something of a termination by rear mounting the driver on a thick baffle and beveling in front of the cone.

It would be interesting to see how much better that would be with a secondary flare; a question that ABEC should be able to answer.

I wish I could replace the coax driver with a small ROSSE derived (ATH) synergy with open back mids. I'm still hoping that someone with more CAD and 3D printing skills than me will design one I can use.