Go Back   Home > Forums > >
Home Forums Rules Articles diyAudio Store Blogs Gallery Wiki Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Project: CBT array, balanced DML, delay curved
Project: CBT array, balanced DML, delay curved
Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 16th August 2019, 10:29 AM   #11
Jim Griffin is offline Jim Griffin  United States
diyAudio Member
 
Join Date: Jun 2003
Location: Tennessee
Bradley,

Your new CBT project presents an interesting implementation of how to attain equivalent performance between an actively delayed straight line array versus the time delays inherent of a curved array. Your use of DML drivers is also novel.

In your first message in this thread you stated:

"Like Don Keele's CBT-24, this array uses five shaded banks rather than constant shading. Each bank is separately controlled by a DSP channel and each actuator is powered by an amplifier tuned to the same voltage."

Keele actually used only three shading banks for his CBT24 project. The lowest 12 drivers (Bank 1) have a 0 dB weight, the next 6 higher drivers (Bank 2) have a -3.5 dB attenuation, and the top 6 drivers (Bank 3) are weighted to a -8.0 dB attenuation level. The bottom two banks achieve the weighting levels via series/parallel driver connections while the weighting for the top drivers are aided by a resistive power divider.

I also used the same weighting levels as Keele for my Modified CBT24 project as you can read at this thread:

My New Line Array--It's a Modified CBT24

Good luck with your project.

Jim

Last edited by Jim Griffin; 16th August 2019 at 10:38 AM.
  Reply With Quote
Old 16th August 2019, 04:07 PM   #12
bradleypnw is offline bradleypnw
diyAudio Member
 
Join Date: May 2019
Quote:
Originally Posted by wonderfulaudio View Post
Nice project. You ve chosen dsp delay to keep the arc angle variable, which is a good thing.
Do note that to hit a particular pattern loss frequency, the narrower the coverage angle the greater the height of the array. For narrow angles, you may need very large array, that may cause a problem with the 30 degree target for the center channel
Thanks for the information. Could you explain more about narrow dispersion angles relative to array length, please. I'm not familiar with the topic.
  Reply With Quote
Old 16th August 2019, 04:12 PM   #13
bradleypnw is offline bradleypnw
diyAudio Member
 
Join Date: May 2019
Quote:
Originally Posted by Jim Griffin View Post
I also used the same weighting levels as Keele for my Modified CBT24 project as you can read at this thread:

My New Line Array--It's a Modified CBT24
Thanks for the correction. Shading is something I don't understand in detail. I'll read your thread and adjust the CBT spreadsheets.
  Reply With Quote
Old 16th August 2019, 09:53 PM   #14
Jim Griffin is offline Jim Griffin  United States
diyAudio Member
 
Join Date: Jun 2003
Location: Tennessee
Bradley,

Keele's and my shading scheme for the CBT24 are for a Legendre function. I'm not suggesting that you change your number of shading levels as you have different number of drivers. Perhaps you can better approximate the rolloff function with additional attenuation banks.

You mentioned in your first post that you were considering other attenuation schemes. Keele's approach is to plot the attenuation function level in dB versus a linear scale of the number of drivers across the aperture. The y-axis is the attenuation amount in dB. The x-axis is scaled by the number from drivers 1 to 24 for the Legendre function which varies from 0 to 1. Thus the Legendre function slopes from 0 dB to about -12 dB down for the highest driver. For the CBT24 example the stepped approximation varies from the Legendre plot by 0 dB for the first 4 or 5 drivers but trends toward a + 2 dB error for the 12th driver. The -3.5 dB step level errs from the true function by -1.5 dB for driver 13 and goes to a +2 db error for driver 18. For drivers 19 to 24 the error varies from -2.5 dB for driver 19 to a +4 dB approximation for driver 24. Keep in mind that the lowest 12 drivers provide more sound radiation while the two strings of 6 drivers each add less volume to the mix.

Viewing the attenuation function plotted across the aperture can help you determine the number of drivers to assign to each attenuation step. Through this process we are not seeking perfection as a good approximation is adequate.

Jim

Last edited by Jim Griffin; 16th August 2019 at 10:01 PM.
  Reply With Quote
Old 16th August 2019, 10:28 PM   #15
bradleypnw is offline bradleypnw
diyAudio Member
 
Join Date: May 2019
Quote:
Originally Posted by Jim Griffin View Post
Keele's approach is to plot the attenuation function level in dB versus a linear scale of the number of drivers across the aperture. The y-axis is the attenuation amount in dB. The x-axis is scaled by the number from drivers 1 to 24 for the Legendre function which varies from 0 to 1. Thus the Legendre function slopes from 0 dB to about -12 dB down for the highest driver.
Thank you for the explanation. I'll need to figure that out for the Cosine and Chebychev functions.
  Reply With Quote
Old 16th August 2019, 10:31 PM   #16
bradleypnw is offline bradleypnw
diyAudio Member
 
Join Date: May 2019
Got my protractor out and measured the array axis frequency response. Mic placed 2 meters away at one meter high for the axis measurements.

First is Legendre shading, 36. Raw and 1/24 smoothed.
Attached Images
File Type: jpg Legendre - raw.JPG (641.4 KB, 51 views)
File Type: jpg Legendre - smoothed.JPG (556.3 KB, 48 views)

Last edited by bradleypnw; 16th August 2019 at 10:44 PM.
  Reply With Quote
Old 16th August 2019, 10:33 PM   #17
bradleypnw is offline bradleypnw
diyAudio Member
 
Join Date: May 2019
Cosine shading, 47. Not sure if this is really cosine shading, I still need to figure that out.
Attached Images
File Type: jpg Cosine - raw.JPG (634.5 KB, 46 views)
File Type: jpg Cosine - smoothed.JPG (558.3 KB, 47 views)

Last edited by bradleypnw; 16th August 2019 at 10:44 PM.
  Reply With Quote
Old 16th August 2019, 10:34 PM   #18
bradleypnw is offline bradleypnw
diyAudio Member
 
Join Date: May 2019
Finally, Chebychev shading, 25. Again, I need to go back and study Keele's shading techniques to make sure I'm shading these correctly.
Attached Images
File Type: jpg Chebychev - raw.JPG (632.2 KB, 49 views)
File Type: jpg Chebychev - Smoothed.JPG (556.9 KB, 12 views)

Last edited by bradleypnw; 16th August 2019 at 10:44 PM.
  Reply With Quote
Old 17th August 2019, 03:32 AM   #19
bradleypnw is offline bradleypnw
diyAudio Member
 
Join Date: May 2019
The 2K hump was killing me so I had to EQ it out. It made the music a lot more pleasant and I can increase the volume now but I still reflexively flinch when familiar songs hit the 2K frequencies.
Attached Images
File Type: jpg 2Ksmooth.JPG (398.7 KB, 18 views)
  Reply With Quote
Old Yesterday, 09:35 PM   #20
bradleypnw is offline bradleypnw
diyAudio Member
 
Join Date: May 2019
Vertical frequency response measurements. Mic six feet away from speaker: 6, 27, 50, 71, and 95 inches high. In a room with a 96 inch ceiling. I separated out measurements for 6 inches vs 96 inches and 27 inches vs 71 inches. Finally, a 1/3rd smoothed graph.

Chebychev shading.

I also included an impulse response for the vertical measurements.
Attached Images
File Type: jpg Vertical - Step Impulse.JPG (563.6 KB, 6 views)
File Type: jpg Vertical - 27 vs 71.JPG (630.5 KB, 6 views)
File Type: jpg Vertical - Chebychev2.JPG (643.4 KB, 6 views)
File Type: jpg Vertical - All.JPG (702.1 KB, 5 views)
File Type: jpg Vertical - one third smoothed.JPG (586.6 KB, 7 views)
  Reply With Quote

Reply


Project: CBT array, balanced DML, delay curvedHide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
COTS end fire sub array delay options? jphil Subwoofers 2 9th November 2015 06:36 PM
Visaton Line-Array with curved front koserthal Full Range 18 8th December 2011 12:37 PM
Starting a curved baffle dipole woofer array project Few Multi-Way 42 19th May 2010 01:47 PM
Curved line array wilbur-x Multi-Way 0 11th February 2008 12:48 PM
Another curved side project Great Dane Multi-Way 0 2nd January 2004 02:00 PM


New To Site? Need Help?

All times are GMT. The time now is 06:33 PM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 15.00%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Copyright ©1999-2019 diyAudio
Wiki