MDD Multi Delays Diffraction (Multi TL, omnidirectional, single drive, ...)

Claudio, what is the internal diameter of the PVC pipe you use?
In the 21M7 project, the internal diameter is 22 mm, the outer diameter 25 mm and the PVC thickness is 1.5 mm. The tubes are used in electrical systems, already cut at 2 or 3 meters.

For a quick test you can use a pack tape as a flange.
 

Attachments

  • 22d7 32 colla.jpg
    22d7 32 colla.jpg
    113 KB · Views: 379
  • 66c9 19 left ap.jpg
    66c9 19 left ap.jpg
    206.9 KB · Views: 372
3D printed version?
There are 3D drawings made by Elleman on other MDD projects but no working prototype.

https://www.diyaudio.com/forums/full-range/341739-34c9-mdd-range-speakers-5.html#post5916484
https://www.diyaudio.com/forums/full-range/341739-34c9-mdd-range-speakers-4.html#post5915845

It is convenient to use 3D printing only to build the connection flange between drivers and pvc pipes. With a euro you buy a 25 mm x 2 meter tube for electrical installations. It is easily cut to the desired size and you can save tens of hours of 3D printing.
 

Attachments

  • smooth1.png
    smooth1.png
    60 KB · Views: 380
  • KS3042527587.10.png
    KS3042527587.10.png
    4 KB · Views: 447
Last edited:
22C71L8 an MDD project

Premise, do not take into consideration the appearance of this new 22C71L8 project, it will change.

The acoustic characteristics are very good:
- 40 Hz to 0 dB with a 3" Faital-Pro 3FE25,
- omnidirectional on the whole audio band,
- almost full-range 40 Hz - 15 KHz,
- reduced distortion up to 80 dB SPL,
- coherent emission points in a three-dimensional volume.

link: 22C71L8
 

Attachments

  • 22C71L8 102 R.jpg
    22C71L8 102 R.jpg
    148.1 KB · Views: 136
  • 22C71L8 100 R.jpg
    22C71L8 100 R.jpg
    147.8 KB · Views: 352
  • 22C71L8 124 FR THD left.jpg
    22C71L8 124 FR THD left.jpg
    88.4 KB · Views: 358
Alien loudspeaker:) I wonder if the multiple pipes at the back are really necessary - maybe a single stuffed line or horn would give enough LF extension. I still plan to try that one day.
To extend the low frequency response I thought of both an air suspension and a rear transmission line. In both cases the response at low frequencies would improve and could be calculated with simulation programs.

I have renounced the air suspension for the generation of internal reflections and the loss of efficiency due to the masking of the rear emission.

With single TL rear loading, an acoustic load could be calculated that exactly compensates the characteristics of the speaker but could not take into account the interaction with the listening environment.

The use of multiple TL has some advantages.

The frequency response at the various points of emission in the listening room changes and you listen to the sum of signals that statistically partially compensate each other.

The emission of coherent signals in non-coplanar points increases the realism of reproduction. I use the expression: 3D effect.

The cost of the spiral sheath is minimal, 20 euros for 30 meters. Unlike a wooden TL in a specific installation with the 22C71L8 project, some waveguides can be lengthened or shortened to equalize the frequency response at a point in the listening room.
 
I finally got the 22C7.V2 up and running just before the end of 2019 :D

Previously, I was going to build the 21M7 until I saw the 22C7.V2. I deviated from the 22L7 (bass section) to use a Tang Band W4-1720 4" woofer in TL arrangement - ABS 4" plumbing pipe 55" long. I've been listening to this setup for 2 days now and am very impressed.
 

Attachments

  • P1090850.JPG
    P1090850.JPG
    113.4 KB · Views: 175
  • P1090842.JPG
    P1090842.JPG
    168.2 KB · Views: 167
  • P1090815.JPG
    P1090815.JPG
    161.6 KB · Views: 142
  • P1090814.JPG
    P1090814.JPG
    150.8 KB · Views: 149
  • P1090808.JPG
    P1090808.JPG
    169.9 KB · Views: 158
  • P1090827.JPG
    P1090827.JPG
    111.1 KB · Views: 164
  • P1090834.JPG
    P1090834.JPG
    166.7 KB · Views: 170
  • P1090838.JPG
    P1090838.JPG
    157.3 KB · Views: 195
  • P1090848.JPG
    P1090848.JPG
    122.9 KB · Views: 192
I finally got the 22C7.V2 up and running just before the end of 2019 :D
Previously, I was going to build the 21M7 until I saw the 22C7.V2. I deviated from the 22L7 (bass section) to use a Tang Band W4-1720 4" woofer in TL arrangement - ABS 4" plumbing pipe 55" long. I've been listening to this setup for 2 days now and am very impressed.

Hi Kec
I was traveling with limited Internet access. I only saw your project today.

The colors you used give a more serious aspect to the realization. I'm glad he was impressed with the type of sound he plays.

I saw that the succession of my posts made you change the work in progress. In the first post of this thread I added my to-do list sooner or later. I have also added a summary of my MDD project guidelines.
 
Hi Kec
I was traveling with limited Internet access. I only saw your project today.

The colors you used give a more serious aspect to the realization. I'm glad he was impressed with the type of sound he plays.

I saw that the succession of my posts made you change the work in progress. In the first post of this thread I added my to-do list sooner or later. I have also added a summary of my MDD project guidelines.

Thanks Claudio. I read almost everything on your website :up: You put a lot effort into these projects and I appreciate it. I hope more people at least try some of these projects. They are pretty easy to build and the results are very satisfying. I haven't measured anything yet, but will at some point.
 
Unlike some of the "innovative" threads in this forum, the evidence looks pretty good to me and deserves more attention both for the quality the treble and the enduring issue of bass. (It is only enduring until drivers with resonance below 20 Hz show up.)

I sure wish I could hear a demo of the treble. While there are ripples in the FR, they are once again, likely a whole lot more prominent to the eye than to the ear. When an orchestra is playing, you don't have a notion of "absolute flat" in your head (like you do with pitch); all you know is what you hear and, as Toole has shown, a crude sense of colouration.

For sure, a lot of us (esp dipole enthusiasts) are placing emphasis on the musical goodness of sound filling our room and less on the point-source stereophony localization geometry of Blumlein.

About the bass, my take is that it is a smart approach to the "sequester" problem albeit limited to the DIY world, for obvious reasons.

Long pipe to sequester rear wave

B.

PS. My solution to the sequester problem was a 17-foot labyrinth:

17 foot pipe sub 12-230 Hz ±5dB
 
Last edited:
Unlike some of the "innovative" threads in this forum, the evidence looks pretty good to me and deserves more attention both for the quality the treble and the enduring issue of bass. (It is only enduring until drivers with resonance below 20 Hz show up.)

I sure wish I could hear a demo of the treble. While there are ripples in the FR, they are once again, likely a whole lot more prominent to the eye than to the ear. When an orchestra is playing, you don't have a notion of "absolute flat" in your head (like you do with pitch); all you know is what you hear and, as Toole has shown, a crude sense of colouration.

For sure, a lot of us (esp dipole enthusiasts) are placing emphasis on the musical goodness of sound filling our room and less on the point-source stereophony localization geometry of Blumlein.

About the bass, my take is that it is a smart approach to the "sequester" problem albeit limited to the DIY world, for obvious reasons.

Long pipe to sequester rear wave

B.

PS. My solution to the sequester problem was a 17-foot labyrinth:

17 foot pipe sub 12-230 Hz ±5dB
The potential of the MDD technology is greater than the measurements of the prototypes that I made. I have a clear understanding of the potential of the MDD technology which will also allow you to build Hi-End speakers, it is probably the most economical system for making high-quality loudspeakers. The acoustic energy concentrated near the transducer can be distributed in a volume with precision using waveguides and the phenomenon of diffraction, the emission is omnidirectional.

I am in difficulty if I have to judge the quality level reached by my prototypes. I have no experience with top-level devices, I don't know if and when I will make a Hi-End prototype, I don't know if and when an MDD loudspeaker will be on the market. I hope other DIYs will try and talk about their impressions. The first attempts seem to confirm that the relationships between costs, time and performance are already very good with the prototypes already published.

I agree on filling the room with sound energy, localization is a possible secondary effect. When listening to a live concert or band music, it is still difficult to locate the origin position of all sound.

The low frequency "sequester" I would call it the transfer of acoustic energy from the cone to the listening environment. With an open waveguide you cannot hope to violate the need for a length greater than L / 4, if you want to reproduce the 20 Hz you need 4-5 meters in length. For now I have reached 2.4 meters, about 40 Hz.

The unconventional aspect of the non-conventional MDD technology is mainly linked to the need to use easily available material for carpentry, plumbing and electrical systems. With a workshop and suitable software available, development times would be much shorter and the appearance would be more appealing, unfortunately they are out of budget.
 
MDD neutral cabinet

I made the second version of the 22C71L8 project, I fixed the spiral guides to a wooden table and rigid guides. It is now transportable and usable for a series of measurements with the microphone at the exit of the flexible guides.

With the graphs it is possible to calculate the mathematical average of the frequency responses which highlights the neutral behavior of the loudspeaker at low frequencies.

More photos at the address: 22C71L8
More information on the neutral cabinet at: acustica MDD projects
 

Attachments

  • 22C71L8 FR Z 1-8 right.jpg
    22C71L8 FR Z 1-8 right.jpg
    75.5 KB · Views: 396
  • 22C71L8 142 umik1.jpg
    22C71L8 142 umik1.jpg
    212.3 KB · Views: 406
  • 22C71L8 146 right.jpg
    22C71L8 146 right.jpg
    110.5 KB · Views: 720
Member
Joined 2008
Paid Member
MDD attachment for compression drivers

I wanted to try out 3D modeling of the MDD as a proof of concept. I yet have to think about how to make the spiral, but the steps look OK. I printed that scaled for a 1" compression driver, the overall length is 85 mm. Once I have some measurements, I will post them.

As another idea, I wonder if MDD type fullrange speaker could be used in a trunk of a car, where the openings would be placed above the back seats. I need to try that one day too.
 

Attachments

  • 17C2C7FF-03EC-402B-B63C-99E7DFCD180E.jpeg
    17C2C7FF-03EC-402B-B63C-99E7DFCD180E.jpeg
    124.6 KB · Views: 365
MDD tweeter

I wanted to try out 3D modeling of the MDD as a proof of concept. I yet have to think about how to make the spiral, but the steps look OK. I printed that scaled for a 1" compression driver, the overall length is 85 mm. Once I have some measurements, I will post them.
For those wishing to try the MDD technology on multi-way systems, I propose a drawing of a possible two-way configuration. The vertical waveguides (12) mounted in front of the tweeter (5) make the speaker omnidirectional on the horizontal plane.

A possible complication could be created by positioning too high that sends the energy of the high frequencies above the listener's head.

One condition to be tested (I have no tweeter available) is the possible lack of rear primary emission. It could be avoided by making holes at the base of the waveguides, a primary wave in phase with delayed waves would be emitted. Given the times and costs of 3D printing, I would first test with rigid PVC pipes.
 

Attachments

  • fig10.jpg
    fig10.jpg
    9.1 KB · Views: 359
As another idea, I wonder if MDD type fullrange speaker could be used in a trunk of a car, where the openings would be placed above the back seats. I need to try that one day too.
For those wishing to test MDD technology in the car, a starting point can be the one in the drawing. The structure is the same as in the 22C71L8 project.

A closed box contains the FR speaker. The box is mounted under the trunk panel.

The frontal waveguides in rigid PVC exit directly into the passenger compartment through holes in the panel.

The rear spiral sheaths come out of the box, can make the necessary curves inside the trunk and exit into the passenger compartment through holes in the panel.
 

Attachments

  • carAudio.jpg
    carAudio.jpg
    38.4 KB · Views: 152
MDDHX135 an MDD project with the HX135 driver

In the MDDHX135 prototype I apply MDD technology to a discontinued Ciare HX-135 speaker. With previous projects I think I have reached the limit of the potential of the excellent 3FE25 of the Faital-Pro, I decided to experiment with the best speaker I have available.

The HX-135 cost about 10 times more than the small 3FE25 and at first glance you can perceive the difference. The first impression is that of a greater extension to the ends of the band, less distortion and greater listening pleasure. I plan to take a series of measurements within ten days.

Most of the time it was used to build a frame that supports the MDDBL rear acoustic load made with spiral sheath for electrical systems. The frame also supports the horizontally mounted speaker and the MDDFL frontal acoustic load, which is very simple to make.
 

Attachments

  • mddhx135-06.jpg
    mddhx135-06.jpg
    178.9 KB · Views: 138
MDDHX135-LV (Lockdown Version)

I made the measurements on the MDDHX135 prototype in my studio, I found myself in the situations of those who move to a better place in the theater but are worse off because of the presence of annoying neighbors. In my case I was annoyed by the excessive dimensions (1000 x 760 x 360 mm) in a 4 x 4 m room, also after 15 days the smell of the solvent contained in the impregnating agent was still excessive. I reconnected the 22C71L8 prototype and continued listening to music with that.

For the impregnating agent, I left the prototype outside without the speakers for another week. For the size I arranged. Stuck in the house for the covid-19 emergency, I could not buy and have the wood cut to size, I had finished the electrician clamps. I cut parts of the frame freehand and instead of the clamps I used rigid copper cable.

I used the only abundant resource: time. If I hadn't been stuck in the house I wouldn't have used a couple of days to get a result like the one shown in the pictures, a prototype destined to be dismantled with the normalized situation.

I moved the speaker cube back and moved the waveguides back. It is clearly seen that the frame is undersized to support all the spiral sheath. The only bulky part of the prototype is the rear one which is positioned close to the wall. The visual footprint is much smaller than before.

How does she feel? Well very well.

I will not make measurements on a prototype destined to be dismantled (I hope soon). Those already published on the same page remain valid, they also contain measurements with the microphone positioned at the outlet of each waveguide.

More photos at the links:
MDDHX135
MDDHX135
 

Attachments

  • mddhx135-35.jpg
    mddhx135-35.jpg
    111.6 KB · Views: 130
MDDHX135-42

The MDDHX135-42 prototype uses the same drivers and waveguides as the previous ones. It is more compact, the base measures 42 x 26 cm, the height of about one meter. It now has a suitable footprint for use in my studio. The construction method is that of the previous LV version, I reused what I had available: woods and rigid wire.

The sound is very good, like the previous MDDHX135 prototypes.

More photos at the link: MDDHX135
 

Attachments

  • mddhx135-FL07.jpg
    mddhx135-FL07.jpg
    41.7 KB · Views: 102
  • mddhx135 FR FL waveguide.jpg
    mddhx135 FR FL waveguide.jpg
    68.9 KB · Views: 91
  • mddhx135-BL12.jpg
    mddhx135-BL12.jpg
    33.2 KB · Views: 93
  • mddhx135 FR BL waveguide.jpg
    mddhx135 FR BL waveguide.jpg
    69.3 KB · Views: 93
  • mddhx135-l52.jpg
    mddhx135-l52.jpg
    135.3 KB · Views: 123
I'm struck by the similarity of this to Harry Olsen's (RCA) original design to increase microphone directivity, by placing a bundle of tubes of progressive length in front of the microphone element. That design was superseded by the now familiar interference-tube "shotgun" microphone, which substitutes a single tube with multiple holes or slots along its length for the bundle of tubes. Some designs use a single tube formed from a material which provides a specific acoustic resistance through the wall of the tube, such as a tight-weave metal mesh.

Experimenters here might similarly try a single pipe with holes or slots along its length, a tube which is "acoustically porous" formed from a mesh, or a tube with a slot running along its length.. which then begins to bear similarity to Karlson tube designs.

This image too large to attach-
https://digilab.libs.uga.edu/scl/files/original/2dc753ad6cd22e015cdfd7387eecdd2a.jpg
 

Attachments

  • 96160f4a767e373c9ef7ad9bb1a99f76.jpg
    96160f4a767e373c9ef7ad9bb1a99f76.jpg
    22.8 KB · Views: 245
  • D-99098.2.jpg.596ec0d233fa9a7b741a03491cbb17af.jpg
    D-99098.2.jpg.596ec0d233fa9a7b741a03491cbb17af.jpg
    164.4 KB · Views: 110
  • Like
Reactions: 1 user
Harry Olsen's (RCA) microphone

I'm struck by the similarity of this to Harry Olsen's (RCA) original design to increase microphone directivity, by placing a bundle of tubes of progressive length in front of the microphone element. That design was superseded by the now familiar interference-tube "shotgun" microphone, which substitutes a single tube with multiple holes or slots along its length for the bundle of tubes. Some designs use a single tube formed from a material which provides a specific acoustic resistance through the wall of the tube, such as a tight-weave metal mesh.

Experimenters here might similarly try a single pipe with holes or slots along its length, a tube which is "acoustically porous" formed from a mesh, or a tube with a slot running along its length.. which then begins to bear similarity to Karlson tube designs.

This image too large to attach-
https://digilab.libs.uga.edu/scl/files/original/2dc753ad6cd22e015cdfd7387eecdd2a.jpg
Hi Gutbucket
thanks for reporting a microphone that I didn't know existed. The structure of some MDD speakers (22C71L8 - 21M7 - 21FL7) and this microphone are very similar, it is the mode of use that makes them very different.

With MDD technology, most of the sound energy emitted by the speaker enters the waveguides, passes through them and is reissued by acoustic diffraction with coherent, delayed and omnidirectional wave fronts.

In Harry Olsen's (RCA) microphone a small part of the energy of the sound fronts in the environment enters the waveguides, only that emitted by a source in line with the waveguides can be detected by the transducer. Sound energy does not come out of the waveguides. The result is an extremely directional microphone.

The presence of multiple guides in the MDD technology and in the Harry Olsen's microphone (RCA) derives from the fact that a single waveguide has minimum and maximum frequency response in the frequency domain. The frequencies of the minima and maxima depend on the length of the waveguides, with more waveguides of appropriate length the minima and maxima compensate each other and a linear frequency response is obtained.

The single tube with multiple holes is a configuration that I have already tried and proposed in the patent application (now abandoned) in 2015. With a single tube in the speakers there are two problems. Resonances related to the length of the tube make the response at low frequencies irregular. The individual exit holes disturb the flat wave fronts that cross the waveguide. With multiple waveguides, the only point of interaction between the sound sources is in the compression chamber in front of the speaker cone.
 

Attachments

  • mddhx135 FR FL waveguide.jpg
    mddhx135 FR FL waveguide.jpg
    68.9 KB · Views: 101
  • fig2fig3.jpg
    fig2fig3.jpg
    15.3 KB · Views: 72