The Candelas: An Experiment In Burning The Transmission Line At Both Ends

[Taterworks]

Hi, everyone.

I've had a set of 4 Pioneer 4-inch full-range drivers (PE #269-569) sitting on the shelf in my basement for the last three or four years, and I'd been scratching my head over what to do with them. Lately, I had an idea that arose out of some thoughts about Danley's tapped horn and Jensen's Transflex enclosure. I thought, what if a single-ended pipe is driven both from near the open end and from near the closed end?

With the Danley tapped horn and Jensen Transflex, the idea is that the pipe undergoes both quarter-wave and half-wave resonant modes, which fill in the nulls created by the other's resonant action and produces a broadband resonator (with self-damping, in the case of the tapped horn). But I was curious to know whether that would work with two drivers, facing the exterior of the enclosure.





And here's the result, minus its eventual painted finish and grilles. It looks like I've given in to a certain sort of Bose design mentality, but I didn't want to go all-out on drivers for an experimental speaker. The 'Candela' is a single-fold, un-stuffed 47 Hz pipe, with open-cell foam lining the tube behind both drivers. The rear driver can be run full-range, or filtered with a corner frequency of 500 Hz to provide BSC.

The result isn't the most musical midbass I've ever heard, but it's not muddy, and these speakers have surprised me with their ability to get loud on just a little power, with little cone excursion. They do benefit from about 5dB of equalization centered at 65 Hz, to overcome the drivers' suspension compliance through that range. If I've truly succeeded in creating a broadband resonator, this seems like the sort of thing Bose would kick themselves for not thinking of.

Can anyone see any technical problems with what I've attempted here (besides the use of drivers with a high Qts)? The terminus doesn't unleash the torrent of air that I might have expected it to, so I wonder whether the rear driver is somehow killing internal standing waves that could otherwise be contributing to output. I can supply an impedance plot if that would be helpful in determining what's actually happening inside these speakers.

[Iain McNeill]

do you have any measurements? impedance and SPL either indoor or outdoor would be interesting as long as you tell us how it was measured.

It's an interesting idea with a great intuitive possibility but I for one couldn't say yea or nay based on a picture.

Some comparisons of one speaker, the other and then both would speak volumes...but this is a PITA to collect, I know too well.

[Graham Maynard]

Hi Taterworks.

Did something similar myself;-
http://fullrangedriver.com/forum/viewtopic.php?id=1761

Cheers ...... Graham.

[Scottmoose]

Got any dimensions?

[Graham Maynard]

I had the line length equal to wavelength of lowest desired response divided by 4, and constant line cross sectional area equal to driver csa.
Very simple, and with midbass + tweeter gave a surprisingly good performance with powerful output easily down to 1/2 driver Fs.

However my efforts went to OB because you can still hear the internal reflections if you try to use a lighweight/thin fullrange driver cone, though of course this happens with all cabinets which develop internal frequency dependent pressures.

From memory my bipoles were about 51" high, 9" deep and 10.25" wide, and the info is in those threads somewhere.


Cheers ..... Graham.

[LineArray]

I was inspired by grahams design to "burn the
candle at both ends" too.

@taterworks:

I can report similar astonishing effects concerning
cancellation of standing waves and midbass quality.


The concept is very very interesting !


Cheers

[LineArray]

internal subdivision:

[LineArray]

sorry for the desaster, heres a collection of pics:

[Taterworks]

I also forgot to mention -- in the case of the Candela design, the driver near the terminus end of the line is shifted up one-fifth of the line length toward the closed end, to begin cancelling standing waves near about 200 Hz and up.

Impedance Plot:

[LineArray]

Hi Taterworks,

since in my variant the terminus is mass loaded there
is an effective shift of the front driver versus the
terminus too, since the pipe is acoustically longer than
geometrical.

I am curious to see your impedance plot ...

Kind regards