Help with esl hybrid

Hello All,

I have finished my esl panel Which Im very happy with, its 5kv biased, 200mm x 600mm and performs very well 700 -20k cycles. under 700 its down and under 300 its very down - due to its small size. I could add more and more panels (like my esl63's) but I want a small pair of speakers.

Im thinking of having 6" drivers under my panels

can someone recommend a suitable driver?

how should I do my cross over? I have so many ideas that now I need some input and advice.


  • blaclberrydec2010 211.jpg
    blaclberrydec2010 211.jpg
    286 KB · Views: 160
Could you test the panel in a much larger baffle and see if the drop-off is the panel or the non-enclosure? Far easier to add plywood than to add panels, even if you need to add some boost.

Could you establish if the SaranWrap is flapping at 500 Hz or where ever or where is the necessary bottom end you could play down to?

BTW, the Dayton-Wright panels look like yours. Each D-W speaker has 8 mounted on the surface of a sphere - to simulate a point source (the "origin" of the sphere). I did a DIY speaker with 6 such panels and had no trouble playing down to 100 Hz, as I recall. They panels were mounted a few inches apart on a curved board about 30 inches across by maybe 50 inches high. Nice.
Hi Bentoronto -the panel plays down to 50 cycles(where the trannys saturate), it has resonance at 70 cycles - but aparts fron the resonance its well down below 300cycles.

I want to keep the panel dipole like the quads

Chinasettawong - the spacing is 2.5mm

Calvin - im a bid fan of your work, how to a establish what my Q is?


  • blaclberrydec2010 242.jpg
    blaclberrydec2010 242.jpg
    295.6 KB · Views: 142
Hi electros,

as it seems you will end up with an XO frequency
fairly above 500Hz, i would think about a dipole or
maybe cardioid style for LF.

Subwoofer is a different aspect, but i guess
at least from 150...700 Hz a dipole or cardioid
would be a better match for your ESL.

That radiation pattern matching will make the
difference between a "box with nice dipole tweeter"
and a system sounding like a homogeneus
fullrange ESL.

As your panel resonance is fairly low,
i think a bit "baffling" the panel - even fairly
narrow side "bars" - would increase margin for
crossver frequency and slope significantly and
raise more of the LF potential of the panel.

It will at least help in the overlap region to
save membrane excursion and increase dynamic

Kind Regards
Last edited:

I´d look at the measurements, simulate and decide after that.
Every panel needs some sort of equing. With a bit of luck You may get away with a passive high-Qt second order highpass (see one of the threads here where I described such filtering). The bass may need 3rd to 4th order lowpass.
Going active you very probabely need a x-over with more than just the X-over filtering but with additional equalizer stages too (or design a custom filter to omit with the chance of generating too complex circuitry).

jauu Calvin