how do people make a dipole subwoofer?

[iron_monkey]

The thread about the new adjustable qts drivers made me interested whether its possible for me to make a dipole from them. It seems so simple at first, attach the drivers to a MDF board, and attach boards to the sides making it into a 'H' or a 'U'. Like this one ---> https://www.htguide.com/forum/showth...ghlight=dipole After a brief research it seems to be difficult, or Im just not finding the right info. With other designs you can get a programs like unibox, and it do all the dimensions and graphs for you, I couldnt find anything like that for dipoles. How do you figure out the physical dimensions etc of the structure you're mounting the driver to?
All I could find was general info and specific instructions for a project like the Phoenix, from this site : http://www.linkwitzlab.com Alot of the content there is completely out of my scope; I get the impression that DIY dipoles are for people with engineering degrees due to little coverage and its unconventionalness.
Then theres the equalization, some people do it by producing their own circuit, and others use the behringer digital EQ. Is mechanical eq, like the new drivers with adjustable qts, sufficient alone?

Anyone like to briefly outline the key steps that led to their dipole creation?

[sfdoddsy]

It's true that there are fewer plop in the box type programs for dipole subs, but that doesn't mean it is hard to build one. The actual construction is easy as pie, whether you do an H-Frame or a U-Frame.

The trick is to get the EQ right, and the amount of EQ you need is dependent on the width of the baffle firstly, and secondly on the driver.

The baffle width gives a frequency called F/Equal. Below this point the speaker rolls off at 6dB an octave. The exact frquency of this point depends on the baffle width - the wider your baffle, the lower F/E will be.

H and U baffles are used to make the baffle width seem acoustically wider without taking up too much space. They do this by folding the sides back. To calculate the width of an H or U Frame, simply mentally unfold the sides.

There are two spreadsheets I used to calculate. The first is from the dipole subwoofer page here:

http://www.diysubwoofers.org/dipole/

Look for dipole.xls file.

It allows you to program in baffle width, Xmax, Qts and a bunch of other stuff to tell you where F/E is etc.

The second is SLs spreadsheet available here:

http://www.woodartistry.com/linkwitzlab/spl_max.xls

The two do give you the same results, although it may help to know the SL's spreadsheet is based on using an H-Frame where D is the depth of the frame, but overall I think the first one is a little more complete.

Once you have established F/E for your given drivers and baffle, you need to work out how you want to EQ it and how much. It is possible to do it with a passive EQ, but I think it isn't worth the trouble and strongly recommend an active crossover/EQ like the Behringer DCX2496. This allows you to fiddle and change your settings to deal with the inevitable room variables, and also to suppress the anomalies that dipoles produce higher in the frequency range. You can also do simple room EQ.

Since the sub will roll-off at 6dB an octave below F/E, you simply dial in the equivalent boost. In other words, if F/E is 80Hz, the sub will be 6dB down at 40 and 12dB down at 20Hz. The DCX allows you to program in a 12dB low shelving filter at 20Hz at 6dB octave which counteracts this for flat response.

Ideally, you would have a simple measuring system to get things just right. I use the Behringer ECM mic and either TrueRTA or ETF.

I stayed fairly close to the Phoenix and Orion in my design so most of the hard work is done, but once you get an idea of the general principles, the influence of various changes becomes easy to work out.

In a nutshell then, a wider baffle means a lower F/E which means less EQ is needed. A higher Qts also means less EQ is needed. The more EQ you need, or the louder you like to play, the more Xmax you need.

Hope this helps

Steve

[iron_monkey]

thanks for the helpful post.

Wouldnt the DCX2496 distort the signal enough to make the initial decision to go for dipole(for the goal of reducing 'box distortion') useless? As compared to feeding a signal straight to a normal box design that doesnt require EQ. There are people who have attempted to mod the DCX, but thats beyond the scope of most people.

Are there any 'severe' limitations using the DCX?

[Jack Gilvey]

The thread about the new adjustable qts drivers made me interested whether its possible for me to make a dipole from them. It seems so simple at first, attach the drivers to a MDF board, and attach boards to the sides making it into a 'H' or a 'U'.

That's all I did. A couple Adire DPL12's on an H-frame with one voice coil on each left open for a resultant Qts of ~0.8. The higher Qts helps counteract the rolloff since I didn't use eq. Even done this crudely (as opposed to Steve's beauties), easily the best sub/bass I've heard. All other subs I've done (sealed/IB, vented/PR) sound much more like one another than they do this one. Not a lot of output, but I got an in-room response measured with my corrected RS SPL meter -6dB @ 20Hz.

[sfdoddsy]

There are two main reasons for going dipole, IMHO. Firstly to reduce the influence of the room as dipoles produces noticeably fewer room resonances, and secondly for the airiness that having no box produces and rear radiation provides.

Provided you follow the correct procedures for using the DCX, which basically means feeding it as high a level signal as possible I doubt you'll be unhappy. I've used three different digital crossover/EQs and haven't had any sonic problems with any of them. If there were any tiny audio problems that slipped past my battered ears, the advantages of dipole operation, active crossovers, room EQ and flexibility more than drown them out.

I'm not a hair shirt audiophile like some people (although I used to be) and I did bypass tests when I first moved to using digital crossovers. If there was any 'distortion' I couldn't hear it.

Cheers

Steve

[Dennis H]

Hey Steve,

I read somewhere else that you're using Panasonic digital receivers for your amps these days. Are you doing the volume control with the preamp or with the receivers?

[sfdoddsy]

I use the volume control from the preamp, with analog attenuators in front of the Panasonics. I could use the volume control on the Panasonics to control the volume I suppose, I just haven't tried it. I'm worried about signal to noise and the complications of using one remote to control two amplifiers. Anyone care to reassure me?

Still, I guess I should give it a try.

Cheers

Steve

[Dennis H]

Well Steve, you're the pioneer with this stuff so I don't think my reassurances would mean much. In theory, the S/N should be better leaving the preamp turned up and cutting volume at the receivers. Or maybe not - who knows with these digital gizmos. Give it a try and see if you can hear any difference at low volumes.

It's good to know you can do it the simple way without it sounding bad. That certainly makes life easier and gives you a lot more amp choices.

[Al Garay]

Any experiences using a U-frame compared to W- or H-frame?

[Dennis H]

Steve should be able to answer that one. As I recall, he tried both U and H frames in his early fiddling.

[iron_monkey]

I just thought, why do I need a behringer (which will cost a hefty $700AU imported) when I have a PC?!! A major proportion of my music is stored on the computer as wavs or mp3s. I have a quiet 2.4 Pentium4. I dont need many channels as I listen to stereo music, so I can get a quality M-Audio delta sound card for $250AU, and this program(which is FREE) called a Brutefir which is basically the software version of the behringer. Far more cost effective. But then this starts to sound not so easy, this talk of FIR/IIR filters

[sfdoddsy]

I tried both U and H frames. I didn't do full measurements, but the U frame seemed to give fuller bass, ie needed less EQ. I went for the H-Frame cos of SL's distortion tests. Once my current bout indolence subsides I may try it again and do some more scientific testing.

Ironmonkey,

The Behringer is of course basically a computer in a box so you could duplicate its functions on a PC, although it would be a hassle.

Cheers

Steve

[SonicBooMan]

Hey Steve,

I sent you an email earlier in the week about the Orion (from heymanode@yahoo.com). Did you get my email? I'm a novice in DIY hi-fi and am learning as much as possible on how to build the Phoenix, then placing them on top of the H-frame woofers. Or I could build the Orion, similar to what you did (MTM + separate woofers). The active crossover that Mr. Linkwitz sells is quite expensive, I think, and requires me to open the box to change EQ? I like the idea of using the digital crossover (such as the Behringer)--it's cheaper and much more flexible, according to you.

Thanks,

SonicBooMan
P.S. If others out there who have advice on building the Phoenix/Orion, please advise. Thanks.

[Davey]

BooMan,

I think you have a basic misunderstanding of the Linkwitz designs....they're dedicated designs. There isn't any flexibility or EQ changing capabilities....on purpose. And the crossover is actually cheaper than alternatives.
However, if you want to do some experimenting with a variety of speaker projects then a DSP-based crossover like the Behringer would be great.

I've built the Phoenix, Orion, and a variety of other dipole systems. I can probably help you out with most questions.

Cheers,

Davey.

[SonicBooMan]

Thanks for the offer, Davey. I can certainly use your advice and guidance. I sent you an email concerning the Orion/Phoenix.

Thanks,

SonicBooMan