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

Multi-Way Conventional loudspeakers with crossovers

2 way TL using SB23NBACS45 and SB29RDCN
2 way TL using SB23NBACS45 and SB29RDCN
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 27th August 2018, 11:31 PM   #21
pkitt is offline pkitt  United States
diyAudio Member
 
Join Date: Feb 2009
I modeled your 8:1 tapered TL using your T/S measurements, first with a stuffing density of 0.3 lb/ft3 (5 kg/m3), then 0.5 lb/ft3 (8 kg/m3) and have attached the two predicted anechoic system bass response curves (red lines). As always in speaker system design there are compromises, like trade offs between f3 and response smoothness.
Paul


Quote:
Originally Posted by eg92b16a View Post
Thanks Paul.

The line overall length is 1.9m. Sd1=800cm2 and Sd2=100cm2. The driver is located 30cm from the closed end. The filling is in the first 1.4m of the line at a density of 5kg/m3. Now that I know this level of fill is considered “light”, I’ll have another go with a density of 8kg/m3 to see the effect.
Cheers
Jason
Attached Images
File Type: gif TTL for SB23NBACS45-8b.gif (6.9 KB, 257 views)
File Type: gif TTL for SB23NBACS45-8c.gif (6.9 KB, 255 views)
  Reply With Quote
Old 28th August 2018, 03:39 AM   #22
eg92b16a is offline eg92b16a  Australia
diyAudio Member
 
Join Date: Feb 2018
Location: Brisbane
Quote:
Originally Posted by LineSource View Post
The Re=6.5ohm Wavecor TW030WA12 is a 30mm dome tweeter in a modest depth waveguide which both extends the low frequency range and provides controlled directivity. Looks like another option for your 2-way design.

TW030WA11 and TW030WA12 are tweeters designed for the most demanding applications, where low resonance frequency, high sensitivity and power handling, controlled dispersion, and a very wide frequency range are required.
FEATURES
• Featuring waveguide face plate for controlled dispersion, offering optimized off-axis and power response
• 30 mm voice coil design with high power handling, and low resonance frequency
• Copper clad center pole yielding very low voice coil inductance for reduced distortion and increased high frequency output
• Internal volumes for low resonance frequency and distortion
• Precision-coated textile diaphragm for improved consistency and high-frequency extension
• Optimized dome shape for ultra high frequency cutoff
• Vented voice coil former for reduced distortion and compression
• Copper-clad aluminium voice coil wire offering lower moving mass for improved efficiency and transient response
• Build-in cavities under dome/edge to equalize pressure for lower distortion and lower resonance frequency
• Flexible lead wires for higher power handling and larger excursion
• Gold plated terminals to prevent oxidation and ensure long-term reliable connection
• Delivered with foam gasket attached for hassle-free mounting and secure cabinet sealing
Looks like a good tweeter, but 115mm diameter is much bigger than the 72mm SB29RDCN. I'd like the drivers centre to centre as small as possible. Currently 155mm, would blow out with a larger diameter faceplate, and I'd have to re-design everything. But that's the fun of DIY, right.
  Reply With Quote
Old 28th August 2018, 03:47 AM   #23
eg92b16a is offline eg92b16a  Australia
diyAudio Member
 
Join Date: Feb 2018
Location: Brisbane
Quote:
Originally Posted by pkitt View Post
I modeled your 8:1 tapered TL using your T/S measurements, first with a stuffing density of 0.3 lb/ft3 (5 kg/m3), then 0.5 lb/ft3 (8 kg/m3) and have attached the two predicted anechoic system bass response curves (red lines). As always in speaker system design there are compromises, like trade offs between f3 and response smoothness.
Paul
I get it now. 4db ripple is fairly substantial in the 200 - 500hz range and we need to smooth it as much as possible without sacrificing the bottom end response.
I'll have another go with a 10:1 taper and 10kg/m3 fill and play with the driver location to see if it can be tweaked some more.

The other thing now is the design of the box. TL mouth at top of box or bottom? Design was based on the 8" driver at the top of the box and tweeter directly below with 100mm offset from centre.

Jason
  Reply With Quote
Old 28th August 2018, 07:46 AM   #24
eg92b16a is offline eg92b16a  Australia
diyAudio Member
 
Join Date: Feb 2018
Location: Brisbane
Paul,
You're absolutely right with the 10:1 taper. It's the best all round compromise between extension and ripple removal. I've settled on a 1.8m length, 10:1 with Sd1=1000cm2 and Sd2=100cm2, driver at 26cm from closed end, 8kg/m3 fill in the first 120cm.

Another question about TL mouth placement, is it ideal to have the mouth far away from the driver or does it not matter?
  Reply With Quote
Old 28th August 2018, 08:00 AM   #25
eg92b16a is offline eg92b16a  Australia
diyAudio Member
 
Join Date: Feb 2018
Location: Brisbane
Quote:
Originally Posted by ernperkins View Post
At first glance it appears both versions do a lot more to protect the tweeter down low. It would be informative to see how close these match to actual LR4/6/8 curves. Does VirtuixCAD have that capability? I'm used to Passive Crossover Designer (PCD) where you can pick a crossover type & slope as a target and design the crossover so the driver response matches that target.

Other than that it's down to measurements (on-axis, distortion, off axis, etc.) and listening to the drivers.
I've just modelled the data into Win PCD and the woofer roll off is bang on LR6, while the tweeter is almost LR8. So should have pretty good protection.
Attached Images
File Type: gif SB TL7.gif (185.1 KB, 248 views)
File Type: gif SB TL8.gif (186.7 KB, 248 views)

Last edited by eg92b16a; 28th August 2018 at 08:06 AM. Reason: add picture
  Reply With Quote
Old 28th August 2018, 01:49 PM   #26
pkitt is offline pkitt  United States
diyAudio Member
 
Join Date: Feb 2009
I was planning on suggesting you increase the taper to 10:1 by simply increasing the beginning area to 1000 cm2, but you beat me to it. I assume when you say "TL mouth" you're referring to the line's exit, the terminus? I've never worried about the relative locations of driver and terminus. With line lengths of 5 feet or more, the line usually ends up being folded in half in its simplest form, and that results in the terminus being on the back panel, pretty far away from the driver. But, you can come up with a 2-fold, 3-segment line that would place the terminus on the baffle at the bottom. BTW, your modeling naturally assumes the line is continuous without any folds or corners, and if you do fold it, you need to model how the line's area actually changes going around the corners if you want a more accurate prediction of the performance. I don't know if the LA software allows folded lines to be modeled accurately, if at all.
Paul

Quote:
Originally Posted by eg92b16a View Post
Paul,
You're absolutely right with the 10:1 taper. It's the best all round compromise between extension and ripple removal. I've settled on a 1.8m length, 10:1 with Sd1=1000cm2 and Sd2=100cm2, driver at 26cm from closed end, 8kg/m3 fill in the first 120cm.

Another question about TL mouth placement, is it ideal to have the mouth far away from the driver or does it not matter?
  Reply With Quote
Old 28th August 2018, 08:29 PM   #27
eg92b16a is offline eg92b16a  Australia
diyAudio Member
 
Join Date: Feb 2018
Location: Brisbane
Paul,
I haven’t really looked into that side of the program, but on first looks, I don’t think it can model a folded line. Will the folds cause more ripple or lower the output?
  Reply With Quote
Old 28th August 2018, 11:29 PM   #28
pkitt is offline pkitt  United States
diyAudio Member
 
Join Date: Feb 2009
The output won't be lowered but there will be minor unwanted effects, most of which can be neutered by the physical arrangement of the angled line divider. Let's say you have a line 70" long and you fold it once, creating two halves, each nominally 35" long. Let's also say the line started out with a depth of 10" and ended with a depth of 1" for a 10:1 taper. So, the internal dimensions of the enclosure needed will be about 35"H x 11-3/4"D (and let's assume a line width of 10"), assuming the divider is 3/4" thick. With the woofer located towards the top of the enclosure in the front half of the line, most often the bottom of the divider would be located such that it equally splits the enclosure's depth at its bottom, meaning the front of the divider at the bottom and the back of the divider at the bottom are both 5.5" from the baffle and back, respectively (and the bottom of the divider is 5.5" above the bottom of the enclosure). What you end up with is a straight, non-tapered length of the line across the bottom of the cabinet which shortens the effective length of the line. To mostly mitigate this, you would instead locate the bottom of the divider so its front edge is more than 5.5" behind the baffle (6" let's say), its bottom is 5.5" above the internal bottom of the enclosure, and its back edge is less than 5.5" in front of the back panel (5" let's say). Last your modeling would then have short 3 straight short sections each with a smaller area than the previous section and more like a tapered line. Overall the line would start with area dimensions of 10"W x 10"D and end at 10"W x 1"D, with three short non-tapered sections in the middle of the line that stepped down the area from 10"W x 6"D to 10"W x 5.5"H to 10"W x 5"D. It's a lot easier to understand this with a drawing. Just for a reference, one can pretty much create the response of a continuously and smoothly tapered line by using multiple (at least 5 but more is better) constant area sections, each of which has a smaller area than the previous section.
Paul

Quote:
Originally Posted by eg92b16a View Post
Paul,
I haven’t really looked into that side of the program, but on first looks, I don’t think it can model a folded line. Will the folds cause more ripple or lower the output?

Last edited by pkitt; 28th August 2018 at 11:31 PM.
  Reply With Quote
Old 29th August 2018, 12:27 AM   #29
eg92b16a is offline eg92b16a  Australia
diyAudio Member
 
Join Date: Feb 2018
Location: Brisbane
With regard to driver centre to centre spacing, what would be the furtherest spacing possible without messing up imaging and crossover response?
Currently the C2C spacing is 155mm with a crossover point of 1250Hz. I could even make it 150mm by shaving a little off the tweeter frame.
I'm asking because that wavecor tweeter does look really good, and available in Australia, but distance increases to 175mm at least.

Cheers.

Last edited by eg92b16a; 29th August 2018 at 12:35 AM.
  Reply With Quote
Old 29th August 2018, 12:33 AM   #30
eg92b16a is offline eg92b16a  Australia
diyAudio Member
 
Join Date: Feb 2018
Location: Brisbane
Quote:
Originally Posted by pkitt View Post
The output won't be lowered but there will be minor unwanted effects, most of which can be neutered by the physical arrangement of the angled line divider. Let's say you have a line 70" long and you fold it once, creating two halves, each nominally 35" long. Let's also say the line started out with a depth of 10" and ended with a depth of 1" for a 10:1 taper. So, the internal dimensions of the enclosure needed will be about 35"H x 11-3/4"D (and let's assume a line width of 10"), assuming the divider is 3/4" thick. With the woofer located towards the top of the enclosure in the front half of the line, most often the bottom of the divider would be located such that it equally splits the enclosure's depth at its bottom, meaning the front of the divider at the bottom and the back of the divider at the bottom are both 5.5" from the baffle and back, respectively (and the bottom of the divider is 5.5" above the bottom of the enclosure). What you end up with is a straight, non-tapered length of the line across the bottom of the cabinet which shortens the effective length of the line. To mostly mitigate this, you would instead locate the bottom of the divider so its front edge is more than 5.5" behind the baffle (6" let's say), its bottom is 5.5" above the internal bottom of the enclosure, and its back edge is less than 5.5" in front of the back panel (5" let's say). Last your modeling would then have short 3 straight short sections each with a smaller area than the previous section and more like a tapered line. Overall the line would start with area dimensions of 10"W x 10"D and end at 10"W x 1"D, with three short non-tapered sections in the middle of the line that stepped down the area from 10"W x 6"D to 10"W x 5.5"H to 10"W x 5"D. It's a lot easier to understand this with a drawing. Just for a reference, one can pretty much create the response of a continuously and smoothly tapered line by using multiple (at least 5 but more is better) constant area sections, each of which has a smaller area than the previous section.
Paul
Thanks Paul.

I have done a couple of TL subwoofers in the past and I used some 6mm ply and lots of liquid nails to create the curves around the folded sections. I started with blocks at 45 degrees to the corners and then moulded the ply around.

Cheers

Jason
  Reply With Quote

Reply


2 way TL using SB23NBACS45 and SB29RDCNHide 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


New To Site? Need Help?

All times are GMT. The time now is 01:22 AM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 15.00%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Copyright ©1999-2018 diyAudio
Wiki