Some design ideas for Dipole/OB needed?

[Mayhem13]

Re: Some design ideas for Dipole/OB needed?

I know there's some dipole builds out there (emerald Physics?) that emply a WG but i can't recall any summed response measurements of any kind. AJINFLA had a build using it as well. The only real way to find out is to just build and test it i guess?

This thread has brought out some excellent discussions though, noting suggestions from both the simple and inexpensive to pricey 4way solutions. I bet there's a middle ground here somewhere.

Although....i remain confident that the Eminence 3012HO will make an excellent dipole midrange driver...it seems that the first part of this journey should be the WG/baffle design.

Is there any software/formula's to calculate the dipole peak in respect to baffle size and driver?

[brianpowers27]

Re: Some design ideas for Dipole/OB needed?

Is there any software/formula's to calculate the dipole peak in respect to baffle size and driver?

1. The edge will simulate the baffle.
2. Jeff B.s boundary and room simulator could generate a dipole transfer functino and then could be run against a WBCD plot. The result could then be input to his PCD.

Note: I suggest a 4 way but it doesn't have to be expensive. The bottom of the 4 way can be crossed via the receiver. Since most of us have subs, this can be a negated cost. I was able to build a tapering 3 way as described in the previous posts for around $200, which is a very reasonable cost. The dipole is only asked to perform to 40hz and therefore will handle dinosaur footsteps.

[fernandov]

Re: Some design ideas for Dipole/OB needed?

I just finished a comprehensive dipole experiment called the Diminui Project which set out to design a smaller, room friendly OB and explore more cost effective means of bi-amplification, crossovers and equalization. You may find some of the information helpful to you in your own project, since we had a few of the same goals in mind.

The Diminui seem to be more forgiving with regard to room boundaries. I currently have mine only about 18" from the rear wall and they sound great, with no loss in stereo image specificity or intelligibility.

My project is documented at http://home.comcast.net/~cunundrum

Ed

[Mayhem13]

Re: Some design ideas for Dipole/OB needed?

1. The edge will simulate the baffle.
2. Jeff B.s boundary and room simulator could generate a dipole transfer functino and then could be run against a WBCD plot. The result could then be input to his PCD.

Note: I suggest a 4 way but it doesn't have to be expensive. The bottom of the 4 way can be crossed via the receiver. Since most of us have subs, this can be a negated cost. I was able to build a tapering 3 way as described in the previous posts for around $200, which is a very reasonable cost. The dipole is only asked to perform to 40hz and therefore will handle dinosaur footsteps.

Will any of the above locate the dipole peak(s)? I thought there might be a formula in relation to baffle width/SD/ and freq range which would more or less locate the 'peak' so to speak. With this info, driver considerations could be narrowed and appropriate XO ranges determined and examined for compatability.

I do have the ability to cross a passive three way to an active bottom as i'm using and AVR with Pre-outs and an NHT B-20 sub amp with 2nd order HP AND LP functions available and selectable from 40-200hz. Defeat on the LP becomes 3rd order at 240hz...need be. The setup you suggest would require another stereo power amp but that's doable.....maybe a Crown with advanced DSP functionality to get a good HP filter function with overlapp to the acoustic rolloff.....once i locate that but i think i'm correct in assuming -6db starting from below the dipole peak?

[Mayhem13]

Re: Some design ideas for Dipole/OB needed?

I just finished a comprehensive dipole experiment called the Diminui Project which set out to design a smaller, room friendly OB and explore more cost effective means of bi-amplification, crossovers and equalization. You may find some of the information helpful to you in your own project, since we had a few of the same goals in mind.

The Diminui seem to be more forgiving with regard to room boundaries. I currently have mine only about 18" from the rear wall and they sound great, with no loss in stereo image specificity or intelligibility.

My project is documented at http://home.comcast.net/~cunundrum

Ed

Thanx Ed, i'll take a look tonight and get back to you.

[martin]

Re: Some design ideas for Dipole/OB needed?

Will any of the above locate the dipole peak(s)? I thought there might be a formula in relation to baffle width/SD/ and freq range which would more or less locate the 'peak' so to speak. With this info, driver considerations could be narrowed and appropriate XO ranges determined and examined for compatability.

I think my MathCad worksheets will do most of what you describe. You can run simulations with an active or passive crossover. There are pdf files you can review to see if the worksheets meet your needs.

[Deward Hastings]

Re: Some design ideas for Dipole/OB needed?

Is there any software/formula's to calculate the dipole peak in respect to baffle size and driver?

John K's A,D,C, Dipole software may do what you want . . .

http://www.musicanddesign.com/A_B_C_Dipole.html

[johnk...]

Re: Some design ideas for Dipole/OB needed?

Not "essential" . . . ORION originally had a single tweeter, after all. But properly positioned in a good room the addition of the second tweeter makes a significant, obvious improvement . . . not in "clarity", and not in localization (which may or may not be the same as "imaging" to some people), but in the creation of a believable "soundstage" or overall "sonic image". In this my experience agrees completely with SL's comments at the Linkwitzlab site.

I thought ORION were the best speakers I'd ever heard with only one tweeter. Adding the second made them better. I'm not convinced that the rear radiation could not have been filled in with a less expensive driver (after hearing PLUTO I'm not convinced Millennium are needed in front, either), but at this point I have no reason, or desire, to experiment. I'm pretty sure a less expensive driver wouldn't sound *better*, so there's no reason to change.

"Something new" might change that, of course, but I haven't heard it yet . . .

Edited to add that if your front wall is "dead", or the speakers are placed too close (less than four feet) to it, a rear tweeter will add little or nothing to the presentation, and may even detract from the clarity of the sound.

I go on vacation and dipole hell breaks loose.

One point I might add that I haven't seen discussed is that it is important to remember when combining the crossover electrical response with the driver response, particularly the midrange, the electrical response is not position dependent but the dipole peak will move higher in frequency as you move off axis left or right. This can be compensated for by careful consideration of baffle width and driver directionality, and is one reason to favor narrower baffles. With wider baffles the driver remains more omni-directional at the dipole peak, and above, and off axis peaks in the response as well as on axis nulls above the dipole peak come into the picture. My experience with wider baffles (some of you may recall my post of my 33" wide x 42" tall system a while back) indicated that it didn't compare in openness and clarity to the narrow baffle system such as the NaO II, Mini, Orion and upcoming NaO Note. I have some discussion of baffles and driver directionality and power response that may be of interest.

Also, rear tweeters aren't just about rear radiation. They are also about trying to correct or control the off axis bulge in the radiated SPL in the region of the mid/tweeter crossover and above. Even with a rear tweeter, without careful design considerations there is still significant bulging of the off axis response and resulting side wall reflections are uneven in spectral content. This is true of both my NaO II design and the Orion (with rear tweeter). The polar response of the NaO II shows this behavior between 2 and 8 K Hz. SL's Orion is similar. I have spend considerable effort improving this in the NaO Note.

Dipole, cardioid and monopole bass is another issue. They all couple to the room differently and at the lowest frequency monopole are required in any event. I won't enter the argument here because it is more a matter of preference and room conditions than superiority of one format over another. I would disagree with one point Deward made earlier regarding the integration of a dipole midrange with a monopole woofers. If the crossover is of the LR type then it is true that through the crossover the response will morph from dipole to cardioid to monopole. I think Deward said that this sound "strange". It may sound a little different than what you would hear with a dipole woofer mated to a dipole mid, or even a monopole woofer mated to a monopole mid, but when the speakers are position correctly for a crossover point around 100Hz the effects is to eliminate or reduce the potential for a null in the mid bass response which results from the reflection of the rear wall (behind the speaker). On the other hand, if the crossover is of the odd order Butterwirth type the transition from dipole mid to monopole woofer does not go through the cardioid pattern. These transitions are shown here.

[Deward Hastings]

Re: Some design ideas for Dipole/OB needed?

Also, rear tweeters aren't just about rear radiation. They are also about trying to correct or control the off axis bulge in the radiated SPL in the region of the mid/tweeter crossover and above. Even with a rear tweeter, without careful design considerations there is still significant bulging of the off axis response and resulting side wall reflections are uneven in spectral content.

Yes. Dipole cancellation from the rear tweeter is . . . incomplete . . . so there is some off-axis "bloom" in the range between crossover and tweeter beaming. The deviation is a nit in comparison to the elephant of inconstant directivity shown by most "box" loudspeakers, but it is there. My room is quite dead at 75-105 degrees off (horizontal) axis (pocket door on one side, window drapes on the other, so that "bloom" is significantly reduced (essentially inaudible) in practice.

The "strange" sound I refer to at the WM crossover is also room and position dependent . . . there are typically room modes in the crossover region which can either benefit from or be accentuated by the apparent dip in power response. In fact at the level of refinement that a good full-range dipole can reach we are perhaps left considering *mostly* room effects . . . they certainly cannot be ignored in the discussion.

I note that we use "front wall" differently . . . I use the same convention as SL, placing the "front wall" forward of the listener and behind the loudspeaker, and the "rear wall" behind the listener. That is, in my usage "front" and "rear" are referenced to the listener, not the loudspeaker, unless indicated otherwise (as in "speaker rear wave", which would be reflected by the front wall).

.

[johnk...]

Re: Some design ideas for Dipole/OB needed?

Yes. Dipole cancellation from the rear tweeter is . . . incomplete . . . so there is some off-axis "bloom" in the range between crossover and tweeter beaming. The deviation is a nit in comparison to the elephant of inconstant directivity shown by most "box" loudspeakers, but it is there. My room is quite dead at 75-105 degrees off (horizontal) axis (pocket door on one side, window drapes on the other, so that "bloom" is significantly reduced (essentially inaudible) in practice.

The "strange" sound I refer to at the WM crossover is also room and position dependent . . . there are typically room modes in the crossover region which can either benefit from or be accentuated by the apparent dip in power response. In fact at the level of refinement that a good full-range dipole can reach we are perhaps left considering *mostly* room effects . . . they certainly cannot be ignored in the discussion.

I note that we use "front wall" differently . . . I use the same convention as SL, placing the "front wall" forward of the listener and behind the loudspeaker, and the "rear wall" behind the listener. That is, in my usage "front" and "rear" are referenced to the listener, not the loudspeaker, unless indicated otherwise (as in "speaker rear wave", which would be reflected by the front wall).

.

Front and rear are always confusing which is why I usually add, "behind the speaker". The entire low frequency thing is really a mess. Cardioid, monopole or dipole, all that can really be said is that they couple to room modes differently. Caridoids couple most strongly and cardioid woofers are always less sensitive to room position and orientation. Then comes monopoles and lastly dipoles. Dipoles couple the weakest but how they couple depends most strongly on position and orientation. But the problem is that most rooms are not "acoustically rectangular". Typical home construction yields walls that are boarding on acoustically transparent to low frequency. That is they are mostly reflective and transmissive with little absorptivity. Hallways and other opening to adjacent room lead to additional resonances. About the only thing that can be said for sure about dipole and cardioid woofers is that for the same on axis anechoic response they radiate less acoustic power than a monopole. That may fit in the plus or minus column depending on the room, application... Bottom line is that if you want to get serious about clean bass response there are better ways to do it, like the double bass array.

[Mayhem13]

Re: Some design ideas for Dipole/OB needed?

Thanx for the responses and discussions.......lot's of excellent topics worthy of serious considerations. I've read the links posted as well and find a 're-read' is in order after researching some of the included deifinitions for a better understanding of the topic.

I'd like to comment and get some feedback on the monopole bass topic raised by John. I'm fortunate to have what may be an acoustic rectangular room and sealed box monopole but stereo subwoofers (RSS315HF's) work very well in the left and right front soundstage with a third helper 15" mono sealed in the rear corner. I am of course currently using point source mains above. The ONLY part of my system that i'm actually content with is the LF/Sub which gives smooth, adequate output into the upper teens. I suspect that a successfull coupling of dipole mid(bass) is possible with an active network to my existing LF system?......via a two way dipole MT top end. BUT the goals i've set forth may preclude the use of a 6.5-8" midwoofer or pair as i'm looking for significant dynamic range for HT use as well.

Last night i began simple construction of a tall narrow baffle (8"x48") to test some drivers i already have just to get a feel for dipole mids and the directivity of a WG mounted dome. There's a SEAS 27TDFC mounted to a 6.5" shallow WG mated to a SEAS CA18RLY 6.5" midbass. Originally in the boxed version the woofer overlapped the WG by .5" placing the CTC spacing at 6.3 inches with a passive LR2 crossover at 2khz which i'll use here just for testing as BSC from the original won't be correct on the open baffle.

[jonasz]

Re: Some design ideas for Dipole/OB needed?

Bottom line is that if you want to get serious about clean bass response there are better ways to do it, like the double bass array.

John, can you give some more information about the double bass array or maybe provide a link. I have eight XLS12 I was gonna put in SL's Thor subs but maybe there are better alternatives out there? Double bass array is something new to me...

[SpacePatrolman]

Re: Some design ideas for Dipole/OB needed?

I've never heard any type of OB speaker so don't have an opinion on them and I read only the OP's first post and a couple more after that, but when I saw the part about whether or not to use a rear-facing tweeter, I immediately thought of Jamo's "R" series which don't use a rear-facing tweeter. So strictly for informational purposes:

http://www.jamo.com/na-en/products/r-909-description/

There's also a version that uses 12" woofers instead of fifteens.

[Mayhem13]

Re: Some design ideas for Dipole/OB needed?

Interesting as i've never noticed before but the tweater is waveguide loaded on both the 907 and 909........hmmmmmm

[ttan98]

Re: Some design ideas for Dipole/OB needed?

Interesting as i've never noticed before but the tweater is waveguide loaded on both the 907 and 909........hmmmmmm

I have seen an actual JAMO 909 playing live(demo), I can tell you there is NO waveguide on the tweeter.

BTW a new Jamo 909 will cost you US$18,000 a pair in Australia, cheaper in the States. This will give you some motivation to build your own pair, I am building a pair based on a similar config. ie TMWW, but using different drivers, drivers alone will cost me less than $1,000 and plenty of DIY woodwork. Initial testing is very promising and I am finalizing my wood work. Too much detail to put here.

Cheers happy building.