My S13 OB. Uniform polar response to tweeters at last!

[gainphile]

This next iteration of my dipole builds cannot happen without the following:


Keyser on Unbaffled Neo3 PDR

Cuibono on Tangband W4 dipole

Rudolf on small-baffle Tweeters

Saurav on Unbaffled Neo3

Mige0 on dipole horn

StigErik on no-baffle OB


Their dedication to perfection is unsurpassed. I really hope I haven't missed anyone for the credit.
I believe this finding is a major milestone in dipole speaker technology

[gainphile]

A bit of history first.

My first dipole was in 2007. In the beginning I thought the most difficult part was the bass. Turns out that the midrange EQ is not easy.

But then after understanding the midrange, it's actually the transition to tweeters that is problematic. The normal approach of back-to-back tweeters is a love-hate relationship. I couldn't decide whether it's actually better than single tweeter. At one stage I was considering to try waveguides/horns (shiver..)

After following closely (and quietly ) the above threads I built this prototype today:





"Tweeters" are 3" TangBand Fullrange.

... and what a beautiful polar response (un-eq for now)! Very consistent response up to 10khz. I haven't seen polar plot of well-known dipole speakers but this would be almost impossible to produce using typical back-to-back dome tweeters.



And so cheap too! Next is to build the EQ

What is surprising for me is how easy it was to set tweeter level. Just measure flat from 1m and I cycled through my music. All sounded well. Tweeter setting using back-to-back Dipole used to be very frustrating.

[planet10]

That looks like a small version of Bud Purvine's rocketship OB... except he is planning a lowther up top.

You have me inspired to go read the threads you've linked.

dave

[gainphile]

Yes, I followed BudP's plan as well. However the size of that Lowther may not allow good polar response near the last octaves. For a 6" perhaps up to 2 or 3 khz. Lower if the diameter is bigger.

What I want is a great 2" fullrange as this would extend further. I do not really need the low Fs (60Hz) of 3" speaker. I have aura nsw2 which works great with Pluto but the rear cavity presented problems as dipole.

[Rudolf]

Gainphile,
I really don't want to spoil the party, but what you show is not (yet) what the contributors mentioned above are trying to achieve IMHO.
You have got nice constant directivity up to 1 kHz, but above 3 kHz it is simply a beaming 3" TangBand. Between 1-2 kHz I see an intermittent beaming which I would relate to a less than optimal depth alignment of both drivers for that area - but I'm not sure of that.

Radiation to the back of the speaker will be VERY different from the forward too.

If you really want to follow the 'CD throughout' path you need to reconsider those "typical back-to-back dome tweeters", but changing to really tiny tweeters - or the Neo3.

Rudolf

[catapult]

Looks good, gainphile! I agree that a 2" with a small rear structure might be ideal. The dipole D would be smaller than either the Neo3 or any back to back setup I've seen.

I don't know if you can still edit your first post but Keyser is using the Neo3 (non PDR) and Saurav is using the Neo3 PDR.

[fergs1]

Greetings Gain, good to see you on the move again. The new baffles look great. Remind me what is the midrange you are using and what is your high pass on the mid(where is it crossing over to the h frame) cheers fergs

[cuibono]

Nice work Gainphile - I really enjoy this sort of shared work environment, getting to informally collaborate and inspire each other.

I like your idea, definitely innovative. I would be interested in knowing how a 2" driver goes.

Rudolf does have a point though - the driver is beaming. Even the SL's Plutos beam, and I feel it is audible, and they use a 1" driver. Something interesting though - SL addressed that in his 2.1 revision - compare the transfer functions. The original Pluto has a flat on axis response, but the later version has a rising response at higher frequencies. I would wager this it to help even the power response due to the tweeters beaming, and I bet it is a nice improvement.

Good work!

[gainphile]

Originally Posted by Rudolf

Gainphile,
I really don't want to spoil the party, but what you show is not (yet) what the contributors mentioned above are trying to achieve IMHO.
You have got nice constant directivity up to 1 kHz, but above 3 kHz it is simply a beaming 3" TangBand. Between 1-2 kHz I see an intermittent beaming which I would relate to a less than optimal depth alignment of both drivers for that area - but I'm not sure of that.

Radiation to the back of the speaker will be VERY different from the forward too.

If you really want to follow the 'CD throughout' path you need to reconsider those "typical back-to-back dome tweeters", but changing to really tiny tweeters - or the Neo3.

Rudolf

The xo is at 2.4khz, so the irregularity 1-2khz is due to the midrange being pushed quite high. As I can cross the TB very low I will see whether crossing at 1khz will fix this issue.

With dipoles, we actually need to 'utilise' the beaming properties of drivers above dipole peak. So the frequency before the dipole dip is perfectly usable. Of course we can use the "pure" CD section that is possible too, but will be very expensive in terms of the number of drivers used.

I did measure the rear radiation, and it is different, but still very smooth to high octaves.

[gainphile]

Originally Posted by catapult

Looks good, gainphile! I agree that a 2" with a small rear structure might be ideal. The dipole D would be smaller than either the Neo3 or any back to back setup I've seen.

I don't know if you can still edit your first post but Keyser is using the Neo3 (non PDR) and Saurav is using the Neo3 PDR.

The 2" I've seen so far have typical problem: the rear opening does not look good. Perhaps due to physical design constraints.

Thanks for the correction. I got confused all the time which one is which.