[Housteau]

Thank you.  That is all very interesting.  Separate from that I know that the models above mine have a high frequency tweeter that is in parallel with the main membrane.  Since the membrane has a much larger excursion than other designs there is a tendency for the highs to be rolled off as it produces deep bass.  Mine does not have those as it is only rated to 35-40Hz.  Those seems to be the hinge frequencies where anything below that requires those tweeters.

[Rocket_Ronny]

Dave, can you play some high frequencies and see if they are coming from the entire width of the panel, or from a narrower line, or lines?

I have had the same thoughts as Jazzman regarding this issue and thought there must be some break in the stators to segment the highs into a narrower section to avoid beaming.

Rocket Ronny

[Jazzman53]

Great idea! 

This can be easily determined by ear alone, with the panel playing at low volume, and the ear very close and panning across the panel.

I noted that some of Popori models have ribbon tweeters in the edge baffle.  This I suspect is a tradeoff imposed by using very high-transformer winding ratios to achieve their remarkable efficiency. 

Upping the bias voltage raises efficiency but it's limited to about 1/2 the breakdown voltage of the air gap, and the only other option for higher output is to increase the transformer's winding ratio.   

Getting seven octaves of bandwidth from a transformer is a tall order anyway, and higher winging ratios (if used) roll off the highs, in which case the ribbon tweeters are probably necessary.

[Housteau]

I had just explained the purpose of those ribbon tweeters being in parallel.  I just know that they are in parallel and no passive components are in the main panel signal path


Yes.  I can do that high frequency test using tones
  What frequencies do you suggest using?  Currently my system is off due to tropical thunderstorms the last few days.

[Jazzman53]

I think you could detect a segment with music alone but if I were using a test tone, I think I would select a tone in the highest octave I can still hear (my hearing falls off a cliff at 5kHz) to determine whether its emitting from the entire panel (unsegmented) or just a small area (segmented). 

If the panel is symmetrically segmented, the top octaves would only be emitted from a narrow zone at the panel center line.
Most segmented ESLs are asymmetridally segmented with just one or two added segments, usually at the inboard edge of the panel.

 
Here's an example of a symmetrically segmented panel layout with 15 wire groups-- a center wire group and seven left/right paired groups on either side.  The center wire group emits the full frequency spectrum above the crossover fequency, and the outer wire groups progressively chop off the upper octaves and sequentially shift phasing, out to the panel edges.  This arrangement produces smooth trending dispersion and bends the wavefront from planar to cylindrical. The separate wire groups are colored red and black for clarity.

[Housteau]

I used pink noise at 5K.  To me the level was higher in the middle of the diaphragm and that was continuous from top to bottom.

The technology is likely different for some of the other models, such as the Hyper-sail.

[Jazzman53]

I used pink noise at 5K.  To me the level was higher in the middle of the diaphragm and that was continuous from top to bottom.

The technology is likely different for some of the other models, such as the Hyper-sail.

If only a couple of segments were added it might be more obvious to detect using a 10kHz test tone. 

In any case, if a flat panel ESL that wide were totally unsegmented, it would beam highs like a laser and put your head in a vice. 

I swore off speaker building four years ago but those Popori's are so pretty I'm lusting...

[Housteau]

If only a couple of segments were added it might be more obvious to detect using a 10kHz test tone. 

In any case, if a flat panel ESL that wide were totally unsegmented, it would beam highs like a laser and put your head in a vice. 

I swore off speaker building four years ago but those Popori's are so pretty I'm lusting...

I repeated the test at 10K and 14K, my limit.  The results were the same, more energy coming from the middle.

Try to have a listen to them if you can.