Four Vifa p13wh00-08s and two Scan Speak d2905-950000 - Page 3 - diyAudio
Go Back   Home > Forums > Loudspeakers > Multi-Way
Home Forums Rules Articles diyAudio Store Gallery Wiki Blogs Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Multi-Way Conventional loudspeakers with crossovers

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 9th February 2017, 04:46 PM   #21
RajkoM is offline RajkoM  Bosnia and Herzegovina
diyAudio Member
 
RajkoM's Avatar
 
Join Date: Feb 2009
Quote:
Originally Posted by chrisb View Post
Lynn, for those of us with less time to wade through the noise of thousands of forum posts, could you share your opinion on the latest generations of very affordable DSP based crossover approaches? I almost said "solutions", but that might seem leading.

+1
  Reply With Quote
Old 10th February 2017, 07:50 AM   #22
suzyj is offline suzyj  Australia
diyAudio Member
 
suzyj's Avatar
 
Join Date: Jan 2006
Location: Western Australia
Quote:
Originally Posted by Lynn Olson View Post
The Ariel crossover is 4th-order acoustic at 3.8 kHz, so adjust your active crossover accordingly. The tweeter will never need more than 20 watts, more likely 10 watts in the real world, so you might consider a Class A option for the HF amplifier.
Firstly, thanks for a cool design. It's already providing plenty of enjoyment, and that's just cutting bits of wood up.

Regarding the crossover, I'm thinking a pair of 3rd order Bessels implemented with a discrete opamp I've been playing with. I'll add some baffle step correction to the bass too.

I don't really do class-a amps, except for work. The amp I'll use for this is a little lateral-fet output, matched jfet input class-ab.
__________________
http://www.suzyj.net
  Reply With Quote
Old 11th February 2017, 11:42 PM   #23
diyAudio Member
 
Lynn Olson's Avatar
 
Join Date: Jul 2004
Location: Northern Colorado
If you like the Ariel "sound", the key components are:

1) Very flat drivers with outstanding impulse response. By this I mean clean decay characteristics in the first 5 mSec, not the rise-time.

2) An acoustic 4th-order crossover that maintains driver-to-driver phase match across the crossover region within 10 degrees. Odd-order crossovers (1st and 3rd order) typically have a 90-degree phase divergence between LF and HF drivers in the crossover region, while carefully-tuned even-order crossovers can reach 5 to 10 degrees phase-match. (I'm speaking of the acoustic, not electrical crossover.) Not everyone can hear phase divergence between drivers, but I can usually hear it from a considerable distance, and don't like it.

3) Asymmetric round-overs on the left and right edges of the cabinet that are fairly large radii (1.25" and 0.75", if I recall right). I built the first versions with symmetric and smaller radii, and the image quality was nowhere near as good, and coloration in the voice region was noticeably greater. Diffraction problems are hard to measure (unless very severe), but I find in my experience they are very audible, and have a substantial effect on overall spatial impression.

Last edited by Lynn Olson; 11th February 2017 at 11:52 PM.
  Reply With Quote
Old 13th February 2017, 10:01 AM   #24
suzyj is offline suzyj  Australia
diyAudio Member
 
suzyj's Avatar
 
Join Date: Jan 2006
Location: Western Australia
I have no understanding of the term "acoustic crossover". Google wasn't particularly helpful. A reference would be greatly appreciated.

My understanding of filters is that each pole provides 45 degrees phase shift at cutoff, so a 2-pole filter will be 90 degrees, a 3-pole one 135 degrees, a 4-pole filter 180. Matching the phase at cutoff between high-pass and low-pass is simply a matter of matching the cutoff frequencies, a doddle with an active filter as you can chuck the whole thing into spice and simulate it.

Or am I missing something fundamental (and no doubt non-electronic).

Also, I've been building boats of late. a 1.25" radius seems awfully sharp to me :P
__________________
http://www.suzyj.net

Last edited by suzyj; 13th February 2017 at 10:09 AM.
  Reply With Quote
Old 13th February 2017, 10:15 AM   #25
system7 is offline system7  United Kingdom
diyAudio Member
 
system7's Avatar
 
Join Date: Aug 2008
Location: Portsmouth UK
Acoustic slopes refer to the sum of the driver's mechanical and filter's electrical slopes after filtering.

Tweeters tend to have a natural rolloff anyway. Less so with the 5" drivers. The object is to operate both drivers in their linear region.

A lot of people aim for Linkwitz-Riley 4 aka LR4 which is 24dB/octave slopes. If it all works well, you get good phase alignment on positive polarity. It's just the maths of it. LR2 is shallower, at 12dB/octave and is negative polarity when time aligned. BW3 18dB/octave is another target slope that people use, Lynn doesn't like it. I do.
__________________
Best Regards from Steve in Portsmouth, UK.
  Reply With Quote
Old 13th February 2017, 10:54 AM   #26
diyAudio Member
 
Lynn Olson's Avatar
 
Join Date: Jul 2004
Location: Northern Colorado
To amplify system7's point, the acoustic slopes are the vector sum of the electrical properties of the filter (or active EQ) and the electromechanical properties of the driver. Vector sum means the phases add together; if the EQ is providing 90 degrees of phase rotation at the crossover frequency (this would fairly typical of a 2nd-order filter), then additional rolloffs in the driver can add another 45 degrees to 90 degrees of phase rotation to the overall acoustical characteristic.

The phase situation gets much more complicated if the driver has peaks, even small ones, in the crossover region. The peak adds phase rotation in one direction on one side of the peak, and phase rotation in the other direction on the other side of the peak. This is true if the peak is minimum-phase; if it is non-minimum phase, the phase response decouples from the amplitude response, and really requires a digital FIR filter to straighten out the phase and amplitude domains. If the peak is minimum-phase, a notch filter, accurately tuned to the mechanical peak, will correct both amplitude and phase domains at the same time, and allow the rest of the crossover to operate accurately.

When I see non-minimum phase behavior from a driver, that's a hint the driver has left the linear range of operation and is in breakup mode, where it doesn't belong. Non-minimum phase operation from a driver, which is pretty common in whizzer-cone drivers (in the mechanical-crossover region), is not a good sign, and frequently beyond the scope of what passive filtration or conventional EQ can accomplish.

As you can see from all the above, it makes it a whole lot simpler if the drivers mind their manners in the crossover region. It gets a lot more complex if FR wrinkles are present in the drivers in the crossover region. Frankly, many high-end designers these days don't seem to care if the crossover region has wrinkly phase and amplitude response, which is where the rough, incoherent sound of many high-end speakers comes from. The most direct test on quick audition is pink-noise or applause; if it's harsh, raucous or tinny-sounding, there are serious crossover problems.

Last edited by Lynn Olson; 13th February 2017 at 11:08 AM.
  Reply With Quote
Old 13th February 2017, 12:57 PM   #27
Lojzek is offline Lojzek  Croatia
diyAudio Member
 
Lojzek's Avatar
 
Join Date: Feb 2012
Location: Croatia
Maybe more clear to say that electrical filter response vector and driver acoustical response vector are being multiplied where their phases are added and their magnitudes multiplied.
__________________
Mankind is created for eternity.
  Reply With Quote
Old 13th February 2017, 09:38 PM   #28
suzyj is offline suzyj  Australia
diyAudio Member
 
suzyj's Avatar
 
Join Date: Jan 2006
Location: Western Australia
Thanks for the detailed clarification. That makes perfect sense. I see the manufacturers don't publish phase info for their drivers, so it's yet another thing that needs to be measured.

The tweeter appears to have a second-order roll off with a cutoff of about 1.8 KHz. Roll off with the mid-woofer isn't as straightforward, as there's different cutoffs for on-axis and off-axis. Is the 3.8KHz crossover frequency in order to stay clear of the tweeter roll off to keep the phase down?

To my mind (thinking purely from the perspective of matching the phase between drivers), there are discrete combinations of filter order and cutoff frequency that will yield accurate phase match at cutoff.
__________________
http://www.suzyj.net
  Reply With Quote
Old 14th February 2017, 09:05 AM   #29
diyAudio Member
 
Join Date: May 2015
Location: Melbourne
Suzy
Towards the bottom of the page;
DIY Audio Articles
  Reply With Quote
Old 14th February 2017, 11:16 AM   #30
suzyj is offline suzyj  Australia
diyAudio Member
 
suzyj's Avatar
 
Join Date: Jan 2006
Location: Western Australia
Thanks Tromperie!
__________________
http://www.suzyj.net
  Reply With Quote

Reply


Hide 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
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
WTB: Scan-Speak D2905/9500 Tweeters kimchee411 Swap Meet 6 11th September 2011 03:30 PM
SCAN-SPEAK R2904/700000 vs. D2905/990000 tweeters Sony Multi-Way 30 28th May 2011 09:24 AM
WTB: Scan-Speak D2905 tweeters (UK) irb Swap Meet 0 14th March 2007 09:01 PM
OLD vs NEW - Scan Speak D2905/9900 vs D3004/6600 danyele Multi-Way 5 14th March 2007 05:50 PM
Scan-speak D2905/950000 walker Multi-Way 5 18th November 2001 11:26 AM


New To Site? Need Help?

All times are GMT. The time now is 03:13 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2017 DragonByte Technologies Ltd.
Copyright 1999-2017 diyAudio

Content Relevant URLs by vBSEO 3.3.2
Wiki