Hello Dean,Has anyone here tried this... running cone drivers for direct sound along with DML's for ambience/holographic/spaciousness ? I tried it last week and I can't see me going in any other direction...... at least not until the next new shinny (old) thing comes along
As Audiofrenzy reminded it, the association of DML and cones was already proposed by Yamaha and Shelley Katz under the naming "layered sound" (there is some where a paper of an experimentation they made... I will try to find it later).
This association was also mentioned in this thread, at least in the early posts and more recently.
The general feeling I read from that is the enthusiasm.
I am curious to test it.
The difference in what you tested is the DML orientation : edge facing the listening position (Some months ago, Burntcoil shown a pair of DML -without cones - in such position)
In your set up, I understand, you have 2 DMLs. correct?
In my opinion, the characteristic of the DML is the change in the IR shape with the direction. In a measurement on axis perpendicular to the panel, the IR is rather short while when the angle increases, the IR lasts longer. This is due to the elementary sources that compose the panel being no more a about the same distance of the mic.
The result is the phase relation between the different frequencies change with the direction so for the reflections on the room boundaries (floor, ceiling, walls) and, there might be not have consensus on that, with the same frequency balance (about the same FR)... No consensus because we can read the FR of DML changes quickly with position. From the measurements I have done, I have no evidence of that when doing FR at let say 1m and changing the angle.
So in having a DML and a cone, the expectation is to get the best of both : the time precision from the cone, the ambiance of the DML.
Hello Andre,
Try that one : https://www.aes.org/e-lib/browse.cfm?elib=21014
Or search for aes.org + 21014
Seems we can't access directly to the link given by Eric.
Christian
Hello André
It is a rather flat cone. Here is a FR of the JA5004 in this family.
More than 5kHz... Possible without entering in DML with such a surface ?
Thank you Audiofrenzy for the answer to the 1st question.
About dipole in bass, the front/rear cancellation is often listed as a drawback that leads to give extra power to the woofer in compensation.
The DML suffers also from such cancellation in the bass.
In the site linked by Deanznz, there is a linked to another thread about the type of woofer : Visualizing How Different Loudspeaker LF Directivity Patterns Couple to Room Modes and in that a thesis : Modal Coupling of Directional Subwoofers in Rectangular Rooms.
A bit heavy reading maybe but interesting to share this source.
There is a parallel between the room modes (Schroder frequency) and the modes in a DML (modal overlap, transition frequency) in the paper pointed by Eric.
My thinking with left-minus-right and vise versa is, if there is very little between the left and right signal, then there is very little ambience ?
I have tried switching the music player to mono output, with 2 bookshelf speakers say 4 foot apart, and a single DML panel in between them, edge facing the listening position, and the DML panel being switched on and off, I use a separate amp so I can change just the Panel volume, and with the panel on it just adds a magic to the mono. It's wonderful !
Yes... exactly my thinking too. I haven't even built proper stands for my panels so that's next.
Yes yes.... please test this out, either I'm delusional, or my bookshelf speakers are horribly sounding without the aid of DML, or DML used off axis is the secret to audio nirvana..... ;-)
It's almost like the worse the recording the bigger the positive effect when the DML is switched on. Music with phase trickery going on don't get worse with DML on, maybe a little different, but not worse, maybe better, or maybe just different.
I have tried adding ambience cone speakers, with the Main speakers positioned at the plus 30 degree and minus 30 degree position, and with the ambience at +/- 60 degrees, then at +/- 90 degrees, then at +/- 120 degrees, then on the floor at +/- 90 degrees firing up at the ceiling, with and without a time delay ranging from 5ms to 25ms. And ambience speakers placed directly behind the Mains facing rearward. I got the idea from Duke that posted in that ASR thread I linked to, he sells speakers with rear firing horn tweeters or a separate version to add to your own speakers, and he calls it
"LORA" Late-Onset Reflection Assist
https://jamesromeyn.com/speakers/space-generators/bighorn-space-generator/All combinations I tried added something good and something bad to what I was hearing. Kind of like musical static, and I didn't care for it. I trust that Dukes speakers sound good, but I could not replicate it with my limited knowledge.
But the DML panel on edge sounds wonderful.... even with all it's lumpy bumpy frequency response.
Mostly so far I use just 1, I have not made a matched pair yet, 1 is 4 mm poplar plywood 400 mm x 600 mm and one is polystyrene the same size. I just feed it what ever channel it happens to get wired to on the day.
There is so much fine tuning available to do with EQ, delay, positioning, mono/stereo, but straight out of the box it sounds better than anything I've tried before so I'm lacking motivation.... so far.All though I do have to take into account we tend to have a "honeymoon" period when we acquire something new... but please if anyone has a pair of cone speakers, a DML and 2 amplifiers do try this out, then come and tell me if I'm dreaming.
the best of both.... sounds good, thanks for you post Christian.
Dean.
Yes, from the (quick) reading of the links I posted and also from other threads on this forum, the trend is the boxed subwoofer, even in fact the closed box one, which seems to give the best balance performance/complexity/availability. To come to the previous wording, the order of choice seems to be : monopole, cardioid, dipole. I haven't seen for now comments in this classification about asymmetric dipole (attenuation of the rear wave) like in the ripole or an infinite baffle (not sure of this name... an open baffle with a rear absorbent)
Christian
Very interesting.
I know the Yamaha polystyrene cone is used in bass amps. Other than that, I had always seen it in conjunction with mids and tweeters for hi-fi applications. Maybe Yamaha stumbled on DML without knowing what it was and subsequently ignored it.
Some more material I collected about the Yamaha in the pdf.
Extract : the crossover of one of the speaker of this NS family. The elephant ear speaker is not limited in its bandwidth. The medium is in LF.
@jaxboy
Hello Jaxboy
Here are pictures of my canvas.
To summarize :
Hope it will help.
Hello Jaxboy
Here are pictures of my canvas.
To summarize :
- I put a tension in the canvas thanks to the triangular keys a each corner of the frame
- I applied 1 layer of watered wood glue at the back of the canvas
- The pad is made of 2 layers of 1mm balsa. One 15x10cm then a 10x10cm with the grain in perpendicular direction of the first one
- I applied a second layer of watered wood glue including over the pad (I don't remember if I applied it also at the exciter position...)
- The exciter was glued to the pad with epoxy in dead center position
- For the spine, it is a simple piece of wood. The back of the exciter is aligned with the back of the canvas frame so no additional blocks were needed. there is a double side tape between the exciter and the spine.
Hope it will help.
That's quite sophisticated. When I built my panel with polypropylene, I was inspired by the Yamaha contour. First I put the exciter at the original location but I got a big hump in the FR. I found a location with a better balance but much more off center.
Some questions about Bertagni :
- Is the thickness into the "ear elephant" contour constant ?
- do you know the thickness in this central area? Of the thicker peripheral area?
- there is a damping or coating material on the front side; Do you know more about it?
My canvas looks in effect exactly like yours, as I used yours as my model, and I really like them. Instead of balsa, though, I used a similar-sized piece of paneling that I sanded on both sides so it was just a hair thicker than just the center ply.
Thanks for that! Did you mean in the second wiring diagram (page 545) that you would have both the box and the DML speakers in front of the listener? How about using the DMLs as rear speakers? If I try that as my rear channels, would you recommend having the speakers connected as two I's or as an X? My seat is essentially between the box speakers and will have to remain that way.
Last edited:
Thanks for providing a working link Christian.
That paper led me to this one by the same authors:
https://pubs.aip.org/asa/poma/artic...Measures-of-vibrational-localization-on-point
This paper is one of the most intriguing (and mercifully short!) papers I've come across in a long time. This figure is central to the paper:
And they end up splitting the frequency response into two regions (figure below), "modal" below a crossover frequency and "statistical" above the crossover frequency.
It reminds me of a discussion we had on this thread some time back concerning the different frequency regimes of a DML. I wanted to look back at that but couldn't figure out how to find it. Any body think they can? (Christian??).
I enjoyed that the paper is so short, but one thing I wish the authors wrote more about was the implications of this crossover frequency on the characteristics of the sound produced in each regime, say in terms of directionality, phase (correlated/uncorrelated), etc.
Eric
^ this reminds me of:From a Yamaha ad.
Who would even think about listening to sound with degeneration!?
from a PDF you posted here:
https://www.diyaudio.com/community/...s-as-a-full-range-speaker.272576/post-7458376
I actually touched YAMAHA NS18
it is not thick
Judging by finger touch, it’s about 5mm?
Because this is an antique speaker
So I'm embarrassed to keep touching it
Because it's not mine
I forgot to touch the middle part
But I guess it’s thick in the middle
At least visually it has ribs
I don't think so
Because YAMAHA is a large group
In the audio department, they have a group of truly full-time professional R&D personnel.
Compared with them, we can only be considered amateurs.
Have fun~
So my DML is currently only used in 1xx ~ 5xxx Hz
Then "DML low frequency" is limited to greater than 100Hz
"DML high frequency" has no limit
Then mix it directly with the paper cone tweeter
There are also Inverted speakers responsible for heavy bass
Does this count as "layered sound"?
I personally think that DML is mainly responsible for the mid-frequency part.
It should be the most consistent with its structural benefits
Feel sorry
Recently I have been preparing to play "coffee roasting"
So DML has not been updated for the time being.
Just keep listening like this and it feels pretty good
I will start a new DML test after the coffee roaster is assembled and tested to a certain stage.
Hello Eric
I will try to find a time to make research (no sure of the result...) and for sure to read this paper.
What I understand behind "crossover frequency" or "transition frequency" is it is the same concept as "Schroeder frequency" for acoustic in a room.
Below this frequency, the separation of the mode is important so when a frequency is played, only one mode is working. The resulting FR is at the image of the modes with peaks and valleys
Above this frequency, several modes will play for a given input frequency. The result will be a smoother frequency response (sort of mitigation between modes).
One can read the transition frequency is a statistic approach showing where the modes have enough overlapping to smooth the FR.
So not a strong physical definition but a metric to describe the behavior of the panel (the room).
The recommendation in the paper you linked (not this last one, the one before) is to have this transition frequency as low as possible.
As it is directly linked to how the modes are located over the frequency range, the drivers (design parameters) are the panel dimensions, the panel ratio, the material choice, the boundary conditions (suspension).
For a room, there is a relation between the reverberation time which is quite easy to mesure and the Schroeder frequency. For a panel, I don't know...
Christian