Dipole (-ish?) surrounds

[tiroth]

This is a new speaker I've been working on to serve as the surround left/right pair in HT. They are quite listenable as stereo speakers but I'm not sure they are in the hifi realm. I also think they will match my mains pretty well without needing to buy a ton of ribbon tweeters.

Tweeter is Monsoon planar driver
Mid is AP130M0
Woofer is Infinity 7" (Vifa)

You can't tell from the pic, but the mid/woof complement are mirrored on the back. Currently running the woofers push+push, the mid/high dipole.

It's an interesting thing to walk around the speaker...standing in the dipole null it is as if the singer is just around the corner in another room. Move two feet in front and everything opens up. I think this behavior bodes well for HT functionality...it shouldn't allow localization of effects, but it does deliver ambience.

My quest continues...I still have a lot of work to do to dial the crossover in, but I did some listening tonight and the concept seems sound. If I ever finish the design I'll do a proper write-up.

[tiroth]

Detail of the tweeter. I had to make something up in terms of mounting.

Initially I thought this could be diffraction city, but the response is actually better in the center of the box than at the front. I think the vertical beam is so tight that diffraction from surfaces in the plane of the driver is much stronger than those orthogonal to it.

[kelticwizard]

Originally posted by tiroth

Tweeter is Monsoon planar driver
Mid is AP130M0
Woofer is Infinity 7" (Vifa)

You can't tell from the pic, but the mid/woof complement are mirrored on the back. Currently running the woofers push+push, the mid/high dipole.

So let me get this straight. You have an Infinity 7" woofer and an Audax 5" mid on the front of the cabinet, the same on the back of the cabinet, and a planar tweeter on top of the cabinet in the middle.

Your 7" woofers are push-push. That means the front and back woofer both go outward at the same time, and both go inward at the same time. This is a bipole arrangement.

However, your 5" mids are different. When the front mid goes outward, , the back mid goes inward?

In other words, you have a dipole arrangement with two speakers instead of one?

So if your front woofer and mid are hooked up in phase with each other, then your your back woofer and mid are hooked up out of phase with each other?

I always thought the dipole idea was just done with a single speaker or array to prevent having to box the speakers. I have never seen it done with two woofers in enclosures the way you are doing it.

What are you gaining from this approach? Why not just have a bipolar woofer, a bipolar mid, and a dipole tweeter?

Or for that matter, if the tweeters are not too expensive, have you considered buying two of them, putting them into an enclosure on topof the main enclosure, and making the tweeters bipole as well?

[tiroth]

I want dipole, not bipole, because of null region. This is designed to be used off-axis!

Since using a single mid or pair of mids didn't really have much impact on the enclosure size, going with two allows me to
1. lower excursion significantly
2. match existing sealed mids in other speakers
3. flexibility to switch to monopolar

If you look at HT dipoles in general, I don't think any are the classic DIY open-baffles. The primary reason is likely the excursion requirement, as most use small drivers.

It may be odd, but for $35 in drivers it works quite well for its purpose.

[kelticwizard]

Originally posted by tiroth
I want dipole, not bipole, because of null region. This is designed to be used off-axis!

....going with two allows me to
1. lower excursion significantly

If I understand correctly, you are not lowering excursion requirements by using two.

Let me try to visualize this. Your front mid has an enclosure, and your back mid also has an enclosure. Or perhaps you have one long enclosure connecting them. Doesn't matter

When you hook the back mid up so it goes inward when the front is going outward, the two mids together are acting as one 5" speaker. If you hooked them up in phase, as a bipole, you would be lowering excursion requirements by 50%.

But as a dipole, you are not lowering excursion requirements at all. Together, the two act as one 5" cone, going back and forth together as a unit.

However, if this arrangement gives you the null pattern you are looking for, then it would make sense. I've never favored dipoles because of the cancellation factor, but others like them and they do seem to be the coming thing. Besides, in an enclosure 1 foot wide, over 500 Hz or so I guess cancellation doesn't happen. Where do you plan to cross the mids over from the woofer?

[tiroth]

I'd like to cross around 400-500Hz. The xover is set to 400Hz based on intitial testing but I need to remeasure and tweak now that I have the final enclosure. You may well be correct about excursion; I did not seriously consider open baffle at any point due to my other requirements.

The null is the entire reason for this project; because of my room the surrounds have to be placed practically on top of the listening area. This results in easy localization of effects. Going dipole is a good way to get ambience only.

[paulspencer]

KW is right - there should not be an excursion advantage to dual driver dipoles. It would be quite feasible to make open baffle - tweeter/mid then crossing to monopole lower midrange/bass.

However, small cheap drivers aren't always suitable for OB. You really need a larger driver with an open basket or the magnet and basket alter the sound out the rear. His arrangement means the front and rear radiation will be the same.

A common surround design is with a monopole midbass with side firing fullrange drivers - dipole. You need monopole for bass and lower midrange. The fullrange drivers mean simplicity (not too many drivers) and for frequencies that are most critically directional, dipole radiation is useful.

You can also experiment with the angle of the dipole drivers, which alters the size of the null. Often the shape is a trapezoid.

[tiroth]

Another difference is that Qtc can be controlled by the size of the enclosure. OB, the only option is equalization to change Q. In this case, either would work, but there are a lot of drivers with unsuitably low Qts.

[paulspencer]

Tiroth, I'm not sure I agree with you. If you have the drivers configured as dipole then the box won't be pressurised. One driver moves into the box as the other moves out by the same amount, yielding no net pressure change inside the box. Hence talking about Q is irrelevant as the box does not provide damping. You do have the dipole mids in their own chamber, don't you?

As I understand, you should treat the drivers as if they are open baffle, except that:

1) you are getting a perfect mirror image - both poles will have the midrange response from the front of the driver.

2) Also, since you have two drivers, there are of course different power / impedance requirements.

[tiroth]

They don't share an enclosure--otherwise it wouldn't be possible to switch to monopole.