David,
Thanks for telling me so much about DML. I wish I were wrong, but from my past experience with anything with an NXT panel, I am not so happy with their sound. They sound a bit thin and tiny to me. I am sure something larger will sound better such as the Podium Sound monsters, but I have never tried them myself and I cannot comment on that.
I am actually developing my own theory in psychoacoustics, it's again something related to point source and delay and decay. The NXT working principle is actually against the idea of having a pin-point source! Instead it opens a can of worm and make vibration modes with different delays and decays! Maybe that exceeds the resolution of our ears, but somehow I just feel there's something wrong with such a theory. Dr. Katz's (Podium Sound) theory seems ridiculous to me, and I am sure it's just an alternative explanation for NXT. If you knock a soundboard of a piano or a guitar or whatever that makes sound, the sound is actually quite monotone. That's the resonance you get from the soundboard. If you get an exciter mounted on that soundboard, then it will have a major SPL peak at the resonance and some minor ones at 2 fo, 4 fo and so... Therefore I believe 1. the NXT panel is not a really rigid diaphragm 2. there is some pistonic motion involved to get the sound right.
I have to go back to my study now, university exams are tough. :-(
Thanks for telling me so much about DML. I wish I were wrong, but from my past experience with anything with an NXT panel, I am not so happy with their sound. They sound a bit thin and tiny to me. I am sure something larger will sound better such as the Podium Sound monsters, but I have never tried them myself and I cannot comment on that.
I am actually developing my own theory in psychoacoustics, it's again something related to point source and delay and decay. The NXT working principle is actually against the idea of having a pin-point source! Instead it opens a can of worm and make vibration modes with different delays and decays! Maybe that exceeds the resolution of our ears, but somehow I just feel there's something wrong with such a theory. Dr. Katz's (Podium Sound) theory seems ridiculous to me, and I am sure it's just an alternative explanation for NXT. If you knock a soundboard of a piano or a guitar or whatever that makes sound, the sound is actually quite monotone. That's the resonance you get from the soundboard. If you get an exciter mounted on that soundboard, then it will have a major SPL peak at the resonance and some minor ones at 2 fo, 4 fo and so... Therefore I believe 1. the NXT panel is not a really rigid diaphragm 2. there is some pistonic motion involved to get the sound right.
I have to go back to my study now, university exams are tough. :-(
Hello Boris,
my guess is you should read some more on bending wave
transducers and especially DML, as soon as you find
some time after your exams:
e.g.
http://www.essex.ac.uk/csee/researc...ubdocs/C123 Modeling romm interaction DML.pdf
Loudspeaker and headphone handbook - Google Bücher
It is a common mistake to mix up the "apparent size of
phantom sources" with the "size of the transducers".
DML radiate rather phase coherent on axis, it is the
radiation at larger listening angles which has increasing
phase decorrelation. The latter being the key attribute to
have a mitigated room interaction when compared to
conventional pistonic speakers.
When trying to understand bending wave loudspeakers
- bending wave loudspeakers have not been invented by NXT btw. , not even the planar ones -
it is most important to fully understand the terms
"modal density" and "modal overlap".
A usual DML has to be used fairly above its fundamental
resonance in its individual range of sufficient modal
overlap to achieve a high reproduction quality.
It is also important to distinguish between the bending
wave transducer principle and the most common
implementations available.
Most implementations do not aim for high quality
reproduction but are designed as cheap and unobtrusive
"easy listening" devices.
As there are many parameters to be chosen well for
achieving high quality reproduction with bending wave
designs, there is a great variety in quality and usable
bandwidth in currently available designs.
Real high quality designs are very few and as they are
not widespread you have much higher probabilty for most
listeners in having listening experience with high quality
conventional speakers than with high quality
bending wave designs.
For this reason up to now most statements on comparison of
audible quality are based on comparing "apples with potatoes"
unfortunately. Even with conventional pistonic speakers
we would not compare a 1$ wideband speaker with an
acknowledged high end system ...
In my personal opinion the bending wave loudspeaker is
able to overcome literally every problem we have with
pistonic speakers:
- frequency dependent directivity
- disturbing phase coherent early reflections in common
(especially non professional) listening rooms
- need for multiway systems for coverage of the audio range
to name the obvious only.
Kind Regards
my guess is you should read some more on bending wave
transducers and especially DML, as soon as you find
some time after your exams:
e.g.
http://www.essex.ac.uk/csee/researc...ubdocs/C123 Modeling romm interaction DML.pdf
Loudspeaker and headphone handbook - Google Bücher
It is a common mistake to mix up the "apparent size of
phantom sources" with the "size of the transducers".
DML radiate rather phase coherent on axis, it is the
radiation at larger listening angles which has increasing
phase decorrelation. The latter being the key attribute to
have a mitigated room interaction when compared to
conventional pistonic speakers.
When trying to understand bending wave loudspeakers
- bending wave loudspeakers have not been invented by NXT btw. , not even the planar ones -
it is most important to fully understand the terms
"modal density" and "modal overlap".
A usual DML has to be used fairly above its fundamental
resonance in its individual range of sufficient modal
overlap to achieve a high reproduction quality.
It is also important to distinguish between the bending
wave transducer principle and the most common
implementations available.
Most implementations do not aim for high quality
reproduction but are designed as cheap and unobtrusive
"easy listening" devices.
As there are many parameters to be chosen well for
achieving high quality reproduction with bending wave
designs, there is a great variety in quality and usable
bandwidth in currently available designs.
Real high quality designs are very few and as they are
not widespread you have much higher probabilty for most
listeners in having listening experience with high quality
conventional speakers than with high quality
bending wave designs.
For this reason up to now most statements on comparison of
audible quality are based on comparing "apples with potatoes"
unfortunately. Even with conventional pistonic speakers
we would not compare a 1$ wideband speaker with an
acknowledged high end system ...
In my personal opinion the bending wave loudspeaker is
able to overcome literally every problem we have with
pistonic speakers:
- frequency dependent directivity
- disturbing phase coherent early reflections in common
(especially non professional) listening rooms
- need for multiway systems for coverage of the audio range
to name the obvious only.
Kind Regards
Last edited:
This is a statement which may be hard to believe for
listeners not having experienced the reproduction
with high quality DML ...
Nevertheless i can nothing but agree to that description.
And please let me add:
It is even possible to achieve that in a common living
room, not only in professionally treated listening rooms.
Even DML benefit from rooms with high modal density
at low frequencies and moderate reverberation but
even in "common" rooms the performance "mimicks"
room acoustics being better apparently as it would
be experienced when using pistonic speakers in the
same room.
The usual tradeoff between phantom source localization
at the one hand and "spaceousness" / "envelopment"
at the other hand is not as pronounced or mutually exclusive
as with pistonic speakers. You can have both at the same
time to a larger extent.
Kind Regards
Last edited:
Alright guys, thanks for telling me so much about DML and their benefits. I think we should get back to the DIY aspect and the possible whirlpool effect that I am trying to create by using a mylar sheet together with an NXT exciter. 
Next time I may try to make a high quality DML and I hope I was wrong about them.
I have to get back to my studies. Exam's approaching!
Cheers,
Boris
Next time I may try to make a high quality DML and I hope I was wrong about them.
I have to get back to my studies. Exam's approaching!
Cheers,
Boris
Hi,
Same here, they were outstanding for AV, clear diction for all listeners.
They also made simple live recordings stand out from studio stuff.
Regarding frequency response, all I can say is the small ones I had seemed
to be optimised for speech and vocals, not too good elsewhere, the idea
full digital EQ is really needed for their best makes a lot of sense to me.
rgds, sreten.
Hi!
I have been using some drivers from the old inflatable speakers attached to 2ft x 4ft 3mm correlux boards from B&Q and the results are astounding. Visitors ask if what they're hearing is electrostatics . . .
Mangers are intriguing beasts - rather disappointing treble however, but they clearly use the star shaped foam to absorb radial travelling waves . . .
Best wishes
David P
I have been using some drivers from the old inflatable speakers attached to 2ft x 4ft 3mm correlux boards from B&Q and the results are astounding. Visitors ask if what they're hearing is electrostatics . . .
Mangers are intriguing beasts - rather disappointing treble however, but they clearly use the star shaped foam to absorb radial travelling waves . . .
Best wishes
David P
Hi, I severely doubt any of them have heard Electrostatics, rgds, sreten.
Hi Kach22i,
Drum skin should work pretty well, I guess, but I would need a long summer to experiment with it. I guess acrylic would be a bit too stiff, and I suspect woven materials would have quite a bit of self damping, which may actually be a good thing, but I need to find a low density woven material to ensure reasonable efficiency.
Drum skin should work pretty well, I guess, but I would need a long summer to experiment with it. I guess acrylic would be a bit too stiff, and I suspect woven materials would have quite a bit of self damping, which may actually be a good thing, but I need to find a low density woven material to ensure reasonable efficiency.
my understanding is that the vibrating panel has to be light and stiff, limp things simply won't work, as the exciter has to be supported. I have had a brief look at a Vifa exciter (Europe Audio) and a piece of A3 foam board (basically stiffened paper) and the frequency response looked like the Alps!
Stiff and light really means a metal or metal/plastic/metal composite, or carbon fibre or glass fibre.
Stiff and light really means a metal or metal/plastic/metal composite, or carbon fibre or glass fibre.
Some aluminum channels JB Welded to the exciter, the channels then supported by a frame or blocking of some kind would remove the requirement that the spider legs be used at all and that the panel be of a stiff self supporting material.
At least that is where I would start.
Maybe I'll have some time this summer to find out first hand.
At least that is where I would start.
Maybe I'll have some time this summer to find out first hand.
Hi borispm,
your idea is like an ordinary moving coil 'speaker drive unit, but with a 'limp' cone? Can't see it working. Its not what the exciters were designed for. The electrostatics are 'driven' all over the panel, whereas your limp panel would only be driven from the centre.
As I am sure you are aware, the exciter works by exciting resonances in the panel, how they expect that to give a flattish frequency response, I don't know.
I use my exciter to induce vibrations in panels and turntable plinths, to measure the effect of damping, see here: http://qualia.webs.com/
your idea is like an ordinary moving coil 'speaker drive unit, but with a 'limp' cone? Can't see it working. Its not what the exciters were designed for. The electrostatics are 'driven' all over the panel, whereas your limp panel would only be driven from the centre.
As I am sure you are aware, the exciter works by exciting resonances in the panel, how they expect that to give a flattish frequency response, I don't know.
I use my exciter to induce vibrations in panels and turntable plinths, to measure the effect of damping, see here: http://qualia.webs.com/
Hi Cats Squirrel,
In fact, I don't know what to expect... I know those exciters are not built for that, and I do not expect anything under 300Hz from those... They should be better than those 1" "full range" drivers though, although I suspect the decay would be pretty slow for the panels...
And yes, with the conventional wisdom, such a panel should have a few large major resonance peaks and the overtones will be of lower SPL, but those commercial systems I've heard before seem to work reasonably well from 300Hz to 10kHz.
In fact, I don't know what to expect... I know those exciters are not built for that, and I do not expect anything under 300Hz from those... They should be better than those 1" "full range" drivers though, although I suspect the decay would be pretty slow for the panels...
And yes, with the conventional wisdom, such a panel should have a few large major resonance peaks and the overtones will be of lower SPL, but those commercial systems I've heard before seem to work reasonably well from 300Hz to 10kHz.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.