3-way active dipole project

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hey guys
I'm just beginning my second dipole speaker project after many years I wanted to share it with you to hear your idea. the plan is to build a perfect symmetrical dipole speaker with m-frame woofers. I'm going to active it with a 2-3way DSP and some Hypex amps. I'm gonna use an Air Motion Transfer tweeter and I'll remove the rear felt pad then I'll hand the tweeter in the air using steel wires. this idea comes from DAudio speaker. I still don't know if Faital Pro 12RS2066 with fs of 41hz will work properly here or not. I have access to all Faital pro and Beyma drivers

this is my crazy sketch for now:

attachment.php


this is Daudio W1 dipole:
DAUDIO_high_end_munich_20162.jpg
 

Attachments

  • photo_2016-05-26_01-13-49.jpg
    photo_2016-05-26_01-13-49.jpg
    45.3 KB · Views: 1,536
Last edited:
Midrange loses dipole action at 1,5Hz. There might be some discontinuity with power response at xo. Not bad however, should not be a big problem for TPL's distortion. Not so sure about the alternative... (Sorry I used 320x250 baffle)
 

Attachments

  • dave123 dipolemid.jpg
    dave123 dipolemid.jpg
    138.8 KB · Views: 551
Baffle diameter is very important for dipoles, it largely determines how dipole radiation appears, where dipole loss, peak and null appear. Narrow shifts all to higher Hz, but then you need more eq in low end and the driver's distortion and Xmax are challenged.

This is the reason why 4-way constructions are better, it is very difficult to stretch a dipole driver's passband to 3 octaves without loosing constant directivity or getting high distortion.
 
Baffle diameter is very important for dipoles, it largely determines how dipole radiation appears, where dipole loss, peak and null appear. Narrow shifts all to higher Hz, but then you need more eq in low end and the driver's distortion and Xmax are challenged.

This is the reason why 4-way constructions are better, it is very difficult to stretch a dipole driver's passband to 3 octaves without loosing constant directivity or getting high distortion.


that's right
do you think adding a 3-4" driver from around 1khz to 3-4khz will be a good thing to do?
something like this:

12" woofers below 120hz
8" midbass from 120hz to 1khz
5" midrange from 1khz to 3-4khz
AMT tweeter from 3-4khz and upward

in that case I'll run everything with three amps and add a passive LP and HP between midbass and mdrange. any idea?


oh look at this interesting Faital pro driver. 5" 99db efficient!

FaitalPRO | LF Loudspeakers | M5N8-80


M5N8-80_response_8.gif


An externally hosted image should be here but it was not working when we last tested it.
 
I don't comment exact driver choices. Here are quick Edge simulations in a baffle that looks a bit like Kreskowsky's NaO baffles.
We can see that problems arise abore 2kHz, thing get tricky when max baffle width should be more narrow than the driver's own physical diameter. Tweeter's response will be nicer in reality. Kreskowsky crosses his 4" driver to tweeter at 6kHz.

it is easiest to make passive xo for highM-T

These are excellent resources. Notice the similarity in baffles, laws of physics are same in Northern California, perhaps even globally!
LX521 Description
NaO Note II RS Details
 

Attachments

  • dave123 tpl range.jpg
    dave123 tpl range.jpg
    148.1 KB · Views: 374
  • dave123 55 mid dipole range.jpg
    dave123 55 mid dipole range.jpg
    148.7 KB · Views: 376
  • dave123 12 dipole edge.jpg
    dave123 12 dipole edge.jpg
    139.6 KB · Views: 387
  • dave123 8 mid dipole.jpg
    dave123 8 mid dipole.jpg
    143.3 KB · Views: 398
I don't comment exact driver choices. Here are quick Edge simulations in a baffle that looks a bit like Kreskowsky's NaO baffles.
We can see that problems arise abore 2kHz, thing get tricky when max baffle width should be more narrow than the driver's own physical diameter. Tweeter's response will be nicer in reality. Kreskowsky crosses his 4" driver to tweeter at 6kHz.

it is easiest to make passive xo for highM-T

These are excellent resources. Notice the similarity in baffles, laws of physics are same in Northern California, perhaps even globally!
LX521 Description
NaO Note II RS Details

Tahnk you Juhazi for your help with good information
you're right having a passive XO between midrange and tweeter is better. these two driver will also draw less current so better to dedicate a separate amp to midbass driver

so here is the list of drivers for now:

2x Faital Pro 12RS1066 as subwoofers
1x Faital Pro 8PR200 as midbass/lower midrange
1x Faital Pro M5N8-80 as midrange
1x Beyma TPL-150 or Aurum Cantus AST2560 as tweeter
 
If you decide to go 4 way, I would suggest
T
M
15" midbass
and separate ripole sub
You can use cheaper (lower xmax, higher eff.) 15" drivers for the ripole sub


I think I prefer to keep the simplicity of three way and try to cross midrange and tweeter as low as possible. I'm starting to think about using a TPL-200 with more headroom and cross it around 1400hz with a steep slope
I also thought about the bridge between bassbin and top section to keep the nasty vibrations out. I'm using 12" subwoofers this will result in a wider speaker. I want to keep it compact so the bridge is not an option. I'm thinking about levitating the top section magnetically just like this: https://www.youtube.com/watch?v=74F7odDKL4Y

attachment.php
 

Attachments

  • ob2.gif
    ob2.gif
    37.4 KB · Views: 524
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.