My open baffle dipole with Beyma TPL-150

[skorpion]

Rudolf,

Got it ! Thanks !




When I talked of 7 mm (in absolute terms) cone movement in post #327 I applied the rule of thumb multiplication with SQR(2) to get a reasonable measure including most all of natural spread. I would think this is a reasonable way of dealing with measures like this.

/Erling

[Rudolf]

Originally Posted by skorpion

Got it ! Thanks !

Hi Erling,
Sorry, I should have been a bit more detailed in my description.
Like this:

[mige0]

Originally Posted by cuibono

Michael, could you rotate your 'thong baffle' 45deg in the baffle plane and do a set of polars? So the frame looks like a diamond (corners point up/down)?

You are thinking along the same lines!


Will take some time to do a little bit of playing around to possibly find a solution to the 4kHz polar peak.

But anyway, I already consider the current "thong baffle" to be exceptionally good - haven't seen such smoothly extended polars measured elsewhere for a speaker that size ! - have you ?

Michael

[gainphile]

Great stuff guys. These are my measurements of Seas L21 (W22's poorer cousin) on narrow 24 cm baffle. Outdoor and ungated. They are very uniform up to 2k+, may go outside again and measure it with no baffles ...

With outdoor measurement the -6db/oct dipole droop clearly revealed.


no smoothing:




with smoothing:

[mige0]

looks excellent, gainphile - thanks for sharing!

Michael

[cport]

Originally Posted by mige0

Below my polar measurements of a 8" and 6" in a 30cm / 12" wide baffle and without any baffle taken indoor with a gating of ~4ms (so forget anything below ~300Hz).
8" is the Jantzen JA8008
6" is the B&C 6PEV13

To make even more clear what Sitg Erik and Cuibono have outlined I normalised FR to on axis response.

What we see is a virtually perfect equalised an axis response and the corespondent polars at 10, 20, 30, 40, 50, 60 deg




Above the 8" Jantzen JA8000 in a 30cm / 12" wide OB





Above the 8" Jantzen JA8000 in a 30cm / 12" without any baffle





Above the 6" B&C 6PEV13 in a 30cm / 12" wide OB





Above the 6" B&C 6PEV13 in a 30cm / 12" without any baffle



Uups – the top end polar's are way better without a baffle (as already been said)
Comparing the two different speaker size' - both without any baffle - it comes as a surprise that the 8" is *smoother* (up to ~2500kHz) than the 6" !
seem to be perfectly reasonable to still cross the 8" at ~ 1500Hz

Comparing the two speaker in the same baffle the 8" speaker clearly wins by a large margin.
To be fair, we possibly would have to measure the 6" in a proportionally smaller baffle.

Also can be seen the dipole –6dB at 60deg isn't true for *both* speakers. I think this is due to the same reason as 90deg never measures 0dB – simply the asymmetry due to the motor structure.


One conclusion seems to be a safe bet, taking a smaller speaker does not necessarily turn out in smoother top end dipole polars - its individual for any speaker and subject to measurement

What can't be seen here is that the penalty is roughly a - 6dB SPL loss for the nude speaker variant

Michael

Hi Michael:

Would it be possible for you to do the same set of polar measurements from the rear of the driver?

Regards,

Charles

[mige0]

Originally Posted by cport

Hi Michael:

Would it be possible for you to do the same set of polar measurements from the rear of the driver?

Regards,

Charles

Sorry - I will not take the effort - it took me almost a full weekend to share above measurements.

There are some rear OB measurements around - also from SL or linked at Rudolfs page (scroll down)
Welche Lautsprecher für OB?

On the other hand - I might come along with some more interesting measurement regarding optimization of OB polar behaviour with big low-mid units - looks interesting for now but needs time consuming refinement and checking back.

Michael

[mige0]

Originally Posted by mige0

On the other hand - I might come along with some more interesting measurement regarding optimization of OB polar behaviour with big low-mid units - looks interesting for now but needs time consuming refinement and checking back.

Sadly further improvement of polar pattern in OB turned out to be a dead end.

Front to rear asymmetry seems to be too big with usual speakers to let me establish any useful prediction form my trials.




Here – as an inspiration - where I actually was heading







#########


But at least - I can share measurements of an additional 8" speaker (well known Dynaudio 21W54) that makes clear that even comparable speakers show *widely* different polar behaviour in OB – not only - but especially in the sensible 1kHz region.




Above the ancient 8" Dynaudio 21W54 measured without any baffle (and intact surround !).




Above again the 8" Jantzen JA8000 without any baffle for direct comparison


Michael

[gainphile]

Hi Michael, thanks for that!

What about the tweeter polar response you've been working on, the "dipole horn"?

I think tweeter frequency is stil the achiless heel of typical dipole setup.

[LineSource]

Which compromise is more desirable:
1) fewer crossover circuits from using a single driver for the midrange like the Linkwitz Orion, and accepting degraded polar response; or

2) improved polar response from using several size optimized speakers, and accepting tone and phase effects of multiple crossover circults over spaced speakers?


Some 8" and 10" midbass drivers appear to have better polar reponse curves than the more common 6" midranges up to 1Khz. I favor the dynamics and coherence of a single 8" or 10" midrange and I always watch for the best speaker curves and baffle designs... like mige0's Dynaudio 21W54 better than expected polar measurements.