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

Open Source Monkey Box
Open Source Monkey Box
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 12th January 2019, 07:20 AM   #641
Juhazi is offline Juhazi  Finland
diyAudio Member
 
Juhazi's Avatar
 
Join Date: Mar 2007
Location: Jyväskylä, Finland
Since there is a minidsp available and set up, you can quite easily test different xo topologies and save 4 configurations on the board. Then it is easy to change config and listen to them.

Small wiggles in response, be it from a driver or xo don't necessarily sound bad at all. It depends very much on how it behaves off-axis. It might take several days and different kind of music to find the best config for a specific multi-way speaker. 2-ways are piece of cake compared to 3-way, especially when midrange has narrow passband.
__________________
Radikal aktivist AINOgradient speaker project
  Reply With Quote
Old 12th January 2019, 10:40 AM   #642
KaffiMann is offline KaffiMann  Norway
Fanatic
diyAudio Member
 
KaffiMann's Avatar
 
Join Date: Sep 2009
Location: Grua, not far from Oslo.
Open Source Monkey Box
Haven't tried mixed order filters yet. It was what saved me when I had my big 4 way setup some years ago.

Edit:
Very certain I've posted this earlier in the thread, but it was a big help to me in understanding some of the important basics of filters.
A Bessel Filter Crossover, and Its Relation to Others

Edit2:
Just one specific example:
Quote:
Rane's Bessel Crossover at -3 dB
Rane's Bessel crossover in Halogen is set for phase match between low-pass and high-pass. This minimizes lobing due to driver separation, and also results in a pretty flat combined response. Another popular option is to have the magnitude response -3 dB at the design frequency. If -3 dB is desired at the setting, the frequency settings need to be changed by particular factors.

You will need to enter separate values for low-pass and high-pass: multiply the low-pass and high-pass frequencies...
For example, for a second-order low-pass and high-pass set to 1000 Hz, set the low-pass to 1272 Hz and the high-pass to 786 Hz.
It is a very good read, a "must" for everyone IMO.
__________________
FC152

Last edited by KaffiMann; 12th January 2019 at 10:47 AM.
  Reply With Quote
Old 12th January 2019, 10:45 AM   #643
Paul Vancluysen is offline Paul Vancluysen  Belgium
diyAudio Member
 
Join Date: Mar 2009
Location: Heusden-Zolder
Quote:
Originally Posted by BYRTT View Post
Can't stop thinking it has been there all the time simulation or not including if one use optimal textbook smooth responses and is inherent non ideal compromise to filter type even if we had not any AC timing plus spacing mismatch, some plot scales mask the fenomen but its there in all of them below.
These EL3 targets are a compromise between ripple and the width of the overlap band between the drivers. It is possible to make the sum response flat, but with a wider driver overlap. In my targets I choosed for the smallest overlap and the lowest ripple. Different elliptical configurations are possibe, but I didn't tried them out yet.
  Reply With Quote
Old 12th January 2019, 12:19 PM   #644
Boden is offline Boden  Netherlands
diyAudio Member
 
Join Date: Mar 2010
Hello Paul,


Why not keep the EL3 Lowpass for the woofer, but ad an LCR suck-out to deal with the 1-5kHz "mountain". In LEAP, that would be a breeze .
  Reply With Quote
Old 12th January 2019, 12:45 PM   #645
mbrennwa is offline mbrennwa  Switzerland
diyAudio Member
 
mbrennwa's Avatar
 
Join Date: Mar 2005
Location: Zurich, Switzerland
Quote:
Originally Posted by Juhazi View Post
I am accustomed to LR symmetrical filters, I have tried duelund, first order, Bessel etc. and S.Harsch with minidsp and Hypex, but I just don't get them working and sounding right. Steep filters like LR48 are more forgiving to make response look right, but sound is eery instead of airy like with poor timing.
Quote:
Originally Posted by KaffiMann View Post
... I see. What about 4th order BW for woof/mid?
Quote:
Originally Posted by Paul Vancluysen View Post
...with the tweaked woofer filter the outband reduction between 1 and 2 kHz is 5 dB worse than the original targets.
In the plot the SPL of the modified filter and the original targets.
I did see it while making the digital version of the tweaked filter, but I thought it was your idea to do the modification like that.

Maybe it is better to reduce the woofer outband conform the targets.
Quote:
Originally Posted by BYRTT View Post
Can't stop thinking it has been there all the time simulation or not...
Quote:
Originally Posted by KaffiMann View Post
Haven't tried mixed order filters yet.
Quote:
Originally Posted by Boden View Post
Why not keep the EL3 Lowpass for the woofer, but ad an LCR suck-out to deal with the 1-5kHz "mountain". In LEAP, that would be a breeze .
All in all, we seem to agree that we need to revise the current x-over design in some way. I did see the leakage of the woofer filter in the simulation, but didn't think it would actually be noticable in the summed response. Since no one complained about it, I just didn't care (let alone check it in the simulation...). Lesson learned!

I'd suggest to try the following in order to proceed from here:
  1. Keep the current filter concept, but reduce the leakage of the EL3 woofer low pass.
  2. Change the woofer/midrange x-over to something similar to a Bessel or Butterworth type with monotonic slopes. The slopes wouldn't be as steep at the elliptic ones, though.
  3. Change the entire x-over to Bessel or Butterworth type filters.

I will try option (1) in my Vituix simulation, but it would be good if someone else would also have a look at this, too.

Would option (2) (mixed filter types) be very wrong? Technical issues? Sacrilege?

Also, other options and ideas are welcome!
  Reply With Quote
Old 12th January 2019, 07:01 PM   #646
KaffiMann is offline KaffiMann  Norway
Fanatic
diyAudio Member
 
KaffiMann's Avatar
 
Join Date: Sep 2009
Location: Grua, not far from Oslo.
Open Source Monkey Box
I would go for option 2, the mid has a sharper roll off, and does not need to change from 3rd order elliptic, but the woofer should maybe change to 4th order BW.
Offsetting xo points should be mandatory IMO, take into account the drivers response, textbook filters are for textbook examples, IE: perfect response imaginary drivers.
With software you can just schew the targeted xo point until the response changes to your liking.
__________________
FC152
  Reply With Quote
Old 13th January 2019, 07:51 PM   #647
Paul Vancluysen is offline Paul Vancluysen  Belgium
diyAudio Member
 
Join Date: Mar 2009
Location: Heusden-Zolder
Quote:
Originally Posted by Boden View Post
Hello Paul,


Why not keep the EL3 Lowpass for the woofer, but ad an LCR suck-out to deal with the 1-5kHz "mountain". In LEAP, that would be a breeze .
Hi Eelco

That is an option yes, but it will make the passive filter version more complex. Then also a fifth order elliptical version can be considered with -55 dB outband reduction, but it all means many components.
It is also possible to improve the sum response flatness, with a wider overlap between the drivers, I will have a look at it...
  Reply With Quote
Old 13th January 2019, 09:16 PM   #648
Paul Vancluysen is offline Paul Vancluysen  Belgium
diyAudio Member
 
Join Date: Mar 2009
Location: Heusden-Zolder
And what about an EL4? Targets in first plot for a xo at 500 and 2500Hz. Compared with EL3 and LR4 at same xo points in other plots.
You can see the EL3, I designed before, is more brickwall around the xo points. EL4 has a more LR4 behavior.
Anyway I will test the EL4 on my own speaker with miniDSP, how it sounds. First time I will design such one.
EL4 450 - 2500 Hz is also possible of course...
Attached Images
File Type: jpg Targets EL4 500-2500.JPG (136.7 KB, 115 views)
File Type: jpg Targets EL4 versus EL3 500-2500.JPG (140.5 KB, 117 views)
File Type: jpg Targets EL4 versus LR4 500-2500.JPG (139.7 KB, 115 views)
  Reply With Quote
Old 13th January 2019, 10:35 PM   #649
Qts is offline Qts
diyAudio Member
 
Join Date: May 2016
Preparation of response measurements for crossover simulation with VituixCAD

https://kimmosaunisto.net/Software/V...eparations.pdf
  Reply With Quote
Old 14th January 2019, 01:19 PM   #650
mbrennwa is offline mbrennwa  Switzerland
diyAudio Member
 
mbrennwa's Avatar
 
Join Date: Mar 2005
Location: Zurich, Switzerland
Quote:
Originally Posted by Qts View Post
Preparation of response measurements for crossover simulation with VituixCAD

https://kimmosaunisto.net/Software/V...eparations.pdf
The question is not how to get data for x-over simulation it in Vituix (or any other software package). The question is which way to go in terms of x-over concept (type of filters, complexity vs. simplicity, etc.).

In my opinion, the circuit of the current EL3 concept reflects about the limit of parts count / complexity that we should consider for this project. Yes, and EL4 or EL5 would have higher stop-band attenuation, but I'd say their circuits are too complex.

I do like how the EL3 filter allows rather narrow overlaps between the drivers. That's why I am hoping to keep the EL3 concept, but with a better stop-band attenuation that I did in my earlier design version. If this does not work, I guess the next step might be to change to a Butterworth-like filter. A Bessel-type also has some appeal due to its smooth phase transitions, but it would involve some larger overlaps.

I am not sure, but I believe it is not wise to mix different filter types within a single x-over (for example Butterworth for the woofer low pass and elliptic for midrange high pass). The transition from one driver to the other should work out in a consistent way for both amplitude and phase in order to get a clean summed output in the transition / overlap region.
  Reply With Quote

Reply


Open Source Monkey BoxHide 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
[Open source] TINA The total open source Audiophile Player based on Beaglebone Black DragonWar PC Based 7 18th December 2016 03:31 PM
Ultimate Open Source XMOS USB-I2S: The source to end all sources. krfkeith Digital Source 4 14th June 2013 04:54 PM
Volume / Source selector - open source project ? AuroraB Analog Line Level 22 22nd September 2012 03:21 PM
Open call for suggestions on Open Source DIY Audio Design gfergy Everything Else 1 15th April 2007 08:33 AM


New To Site? Need Help?

All times are GMT. The time now is 08:52 AM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 14.29%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Copyright ©1999-2019 diyAudio
Wiki