PIEZO NXT type panel

[LineArray]

With some simple EQing the frequency response could look
rather good ...

How big is the window ?

Thickness of glass ?

Only one glass Layer ?



Regards

[jzagaja]

110x110cm, 2mm thick. 50Hz resonates with a window frame. I can't figure out what's this yellow element in series with exciter. It doesn't have capacitance and inductance is unstable. Second electrolytic capacitor is 82uF. Sounds with slight echo.

[mkstat]

Hello Oliver!

Thank you for the critical revision of my prior post and correcting the severe mistake in it, as it is essential.


Also my prior statement in this post:
"Basicly there is a gain of efficiency above coincidence frequency."
in context with
"The output from some pistonic action is not considered here."
is a little bit sloppy/ misleading too.

Obove coincidence frequency output is gained from bending wave operation.

Below coincidence frequency things get even more complicated. IMO (and as you have noted before) the panel can be seen as an array of multiple dipole sources all having a common orientation but different phase relation. The output from the panel below coincidence is due to pistonic action and not from bending wave operation anymore. This is important because below coincidence phase cancelation effects definitely come into play and the panel size now even more plays a major role regarding LF output. The influence of panel size on modal density is of course still effective.

One problem is that below coincidence the excursion of the exciters' voicecoil has to increase quite fast with decreasing frequency to keep the acoustical output constant. The available exciters are not designed to exhibit large linear excursion. Especially the panels' lowest modes will drive the voicecoil out of the linear excursion range resulting in drastical reduced maximum output respectively high distortion figures clearly audible over the whole range.
Another problem is if we are useing high stiffness materials, the Q-factor of modes (especially the lowest ones) will be very high resulting in a peaky response with very high excursions, so damping becomes meaningfull for FR and reducing excursion.

In any way it should be avoided to excite the lowest modes of the panel because the described effect is most dramatic here (the energy of the modes decreases with rising order), otherwise the maximum possible undistorted output will be comparably low.

Understanding this is IMO one of the key factors when a as small as possible panel size with high output capabilities is the goal.

Please correct me if there is something wrong in this thoughts.


P.S.
I know there are many people in this forums who think this is all just uninteresting technical babbling and don't see the point.
I just like to understand as much as possible of the interactions of the design factors involved and I've learned much (also from "mistakes") since I joined the discussion.
My intention is in helping to define some sort of "practical design corridor".
My point is that I don't have to try out a square wheel when I knew before by envisioning it in my mind, it won't work. My aim is to help saving precious time and money and to accelerate progress in this field, independent of each individual design goal. I'm convinced that there are many possible choices of appropriate panel materials/compositions depending on what someone wants to achieve. It's all about choosing the right compromises in the distinct design. I certainly advocate any attemps in finding meaningful measurement procedures as they can give insight in problems which simply can't be captured by ear. Measurment has always been and will always be an important tool for me just as important as a set of screwdrivers.
Again - IMO something as the "ultimate" panel material for all purposes just can not exist - this would be some sort of closed-minded/ dogmatic point of view and will IMO slowdown progress respectively keep us turning in circles.

with kind regards
Markus

[mkstat]

Originally Posted by jzagaja

... I can't figure out what's this yellow element in series with exciter. It doesn't have capacitance and inductance is unstable. ...

It's a polyswitch for protection of the transducers.
Like these: Parts-Express.com:*Raychem RXE110 1.10A Polyswitch | temperature polyswitch fuses fuse

regards
Markus

[LineArray]

Originally Posted by mkstat

...
One problem is that below coincidence the excursion of the exciters' voicecoil has to increase quite fast with decreasing frequency to keep the acoustical output constant. The available exciters are not designed to exhibit large linear excursion.
...

Hello Markus,

i think i understand your search for a "design corridor" and
came to similar conclusions so far.


Kind regards

[mkstat]

Hello Oliver,

regarding the Aerogel topic discussed some time ago - this is not completely out of sight. There is a hydrophobic granulate material comercially available used for window insulation which is also not too expensive. So it could probably be used in a composite material to contribute to the desired behaviour in a panel.

regards
Markus

[LineArray]

Originally Posted by mkstat

...
So it could probably be used in a composite material
to contribute to the desired behaviour in a panel.
...

Hi Markus,

yes why not !
Most likely every good panel material will be a composite
(possibly developed by some audio fanatic) ...

By looking at composites used in the aircraft industry
or in sports equipment, we can see e.g. certain parameters
varying across the size of structures.

The making of musical instruments is full of such
techiques too. Different material properties are
combined to get closer to a desired behaviour.

There are myriads of possibilities.


Concerning your former post:

Originally Posted by mkstat

...
In any way it should be avoided to excite the lowest modes of the panel
...

I just like to add that most of the techiques
used for bass reproduction are resonators.

Examples:

- a dynamic speaker in a closed cabinet is a second order system.

- an open baffled speaker is a second order system often aligned
to have Qts from 0.7 ... 2 to compensate baffle rolloff.
Which not necessarily causes bass response to be "boomy" or low quality ...

- a dynamic speaker in a br cabinet is a fourth order system.
The Qm of the br cabinet itself can easily be > 10 !
In the case of a br cabinet it is the Thiele & Small alignment, which
leads us to controlled and auditively acceptable behaviour.

- Transmissionline cabinets e.g. can have similar behaviour like
DML (-> impedance plot). The resonating structure is a pipe filled
with air, exhibiting resonances according to length, form and kind
of porting at both ends.

Those examples should illustrate, that a resonator with high Q
involved in a system is not a problem in itself necessarily.

It depends on how the whole system is designed and aligned as a
network, including all coupled compliances, mass reactances and
resistances.

Kind regads

[mkstat]

Hello Oliver

I'm aware of the fact that high Q-systems are a very common practice in the loudspeaker business and of most of the examples you mentioned. Often it's a good compromise to keep the systems compact. The other thing is that when room acoustics come into play, IMO many of the side effects of this high Q systems are blurred anyway.

Over the years i had the opportunity to gain much experience regarding sounddesign of large scale high-end sound reinforcement systems (for up to 25000 people) with various different subwoofer approaches and source designs (different stacks, sub arcs, cardioid configurations,...). High Q systems are common here, as high Q has an interrelation to high efficiency. However - high Q-factors in the speaker design is clearly audible to me especially in open air situations and not always in a way that i'm delighted from the performance.
Personly i don't have to like this approach too much and as a designer i have the opportunity to push the compromise away from this. In the end it's a matter of what is required and personal taste.

with kind regards
Markus

[sled108]

Well, this is certainly a new Bit of DIY distraction...I just got 2 pairs of the Parts Express Specials for $10....That, and one 20" x 30" sheet of Black Elmers Foam Board for $5 at the local Stationary Store..Cut in Half....And how I have a Pair of very Different speakers..The first thing you need to do, is pull the Clear Plastic Covers off the Drivers..That, and a goodly amount of properly applied Masking Tape...And..Go Figure..They actually work. Maybe a bit Light on the Bass, but they're only 15" x 20". A nice size to get the Idea of the Potential of these....Pretty much the Sound of a Full Range Single Driver Speaker, but way Cheaper and so many DIY Options..Many more hours of Fussing ahead.

[LineArray]

Hi sled108,

how does that sturdy board sound, when hit with the knuckle ?

Can you please try to describe the sound or compare it with

other materials ?


Kind regards