Go Back   Home > Forums > Loudspeakers > Multi-Way

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 19th February 2013, 11:15 AM   #11
Dissi is offline Dissi  Switzerland
diyAudio Member
 
Join Date: Sep 2012
Location: Winterthur
Quote:
Originally Posted by AllenB View Post
If you can measure the driver impedance (or find a *.zma for it that someone else has done), you can simulate the passive crossover and there you have the transfer curve. You don't need any driver frequency response measurements.
Doing so gives the following transfer functions:

DriverVoltagesOvernightSensationsRev2.jpg
DriverVoltagesCT248.jpg
  Reply With Quote
Old 19th February 2013, 12:25 PM   #12
soren5 is offline soren5  Denmark
diyAudio Member
 
Join Date: Jun 2011
Location: Copenhagen
Quote:
Originally Posted by Dissi View Post
Doing so gives the following transfer functions:

Attachment 331407
Attachment 331408
Pardon my ignorance, but what exactly has been done here? Where have you found the driver impedances? And how did you make these transfer functions? Also, how should I go about designing an active filter based on the transfer functions? Sorry for all the questions, but I'm trying to learn

Thanks!
  Reply With Quote
Old 19th February 2013, 01:52 PM   #13
Dissi is offline Dissi  Switzerland
diyAudio Member
 
Join Date: Sep 2012
Location: Winterthur
It's just the output of a speaker simulation program. Accidentally I had a simulation of both speakers you mentioned, so it was no big deal. That means the program knows cabinet, passive crossover and driver data of both speakers and is able to simulate some results. The curves show the transfer function of the passive crossover. If you manage to get the same transfer function using a digital crossover, then you have reached the goal of converting a passive loudspeaker to an active equivalent.
  Reply With Quote
Old 19th February 2013, 02:13 PM   #14
soren5 is offline soren5  Denmark
diyAudio Member
 
Join Date: Jun 2011
Location: Copenhagen
Quote:
Originally Posted by Dissi View Post
It's just the output of a speaker simulation program. Accidentally I had a simulation of both speakers you mentioned, so it was no big deal. That means the program knows cabinet, passive crossover and driver data of both speakers and is able to simulate some results. The curves show the transfer function of the passive crossover. If you manage to get the same transfer function using a digital crossover, then you have reached the goal of converting a passive loudspeaker to an active equivalent.
I see. Thanks for the explanation. What simulation software have you used here?
  Reply With Quote
Old 19th February 2013, 02:31 PM   #15
Dissi is offline Dissi  Switzerland
diyAudio Member
 
Join Date: Sep 2012
Location: Winterthur
It's my own program, unfortunately still unreleased and not available.

MZ SpeakerDesigner
  Reply With Quote
Old 19th February 2013, 05:36 PM   #16
Jay is offline Jay  Indonesia
diyAudio Member
 
Join Date: Feb 2003
Location: Jakarta
Quote:
Originally Posted by soren5 View Post
I've never made a passive crossover before, but I've read hundreds of times how you experts advice us noobs to begin with a proven design, that's why I'd like to "copy" instead of trying to "improve" as you suggest.
There is no active OS. The OS is passive and an active equivalent is a different design, except in general parameters such as choice of crossover frequency.

Quote:
Originally Posted by soren5 View Post
I was hoping there was a general way of doing this without having to measure each driver. Is it possible to design a digital crossover simply by looking at passive design?
I have no experience with DSP but I think I know how things work, so...

1) From passive XO you can see the L-PAD. This tells you how much you want to pad down the tweeter because it is more sensitive than the woofer. Tuxedocivic gave a hint that it is -6dB (the level of the tweeter). But actually you can use ears to find the level you are comfortable with.

2) Any popular speaker project will tell you what is the XO frequency (here I think 4 kHz) and what is the acoustical filter (here I think LR4). You can copy this setup then later compare with what Tuxedocivic has done, because the benefit of active XO over passive XO is that you can bring the XO lower by steepening the slope (Here Tuxedocivic did it with BW5) without ill effect of high order passive filters.

3) To flatten the response you can use the parametric equalizer. Tuxedocivic already gave you a hint about frequency points where you need to boost or cut and the Q. You can see the response measurement of the drivers to visualize the dips and the peaks.

Do this step by step, from the most critical peak. You can bypass your setting to hear whether your EQ really hit on the right target or not.

4) Phase and delay. This is the most critical imo. The good thing about LR4 is that there is "no" issue with phase. I don't know whether a novice can set the delay based on ears. But you can try. If you switch the speaker cable polarity, a deep null will be heard as the signal being greatly attenuated. This is a good condition of being in phase at XO frequency.

5) If you know how things work and especially if you have good ears, you can try various setup relatively quickly and then can just trust your ears. This is the benefit of such active XO.

ADD:

In the passive XO for the woofer you can see a capacitor across the inductor. This is to build a deep notch at high frequency so as to steepen the filter to meet LR4. This is not an easy thing to do with your DSP. I don't know how the PEQ in the MiniDSP work but you can try. Simply listen to the woofer, does it have intolerable peak/distortion. If so, pay attention to the woofer PEQ setup above XO frequency. If you still cannot bear the peak/distortion then you should consider steeper crossover. Or just compare to what Tuxedocivic has done.

Last edited by Jay; 19th February 2013 at 05:50 PM.
  Reply With Quote
Old 19th February 2013, 06:15 PM   #17
soren5 is offline soren5  Denmark
diyAudio Member
 
Join Date: Jun 2011
Location: Copenhagen
Thanks for the input Jay. Very appreciated!

Just one thing I'm not sure about:

Quote:
Originally Posted by Jay
Any popular speaker project will tell you what is the XO frequency (here I think 4 kHz) and what is the acoustical filter (here I think LR4).
Is the crossover always the same frequency for the two drivers crossing? So if it's at 4kHz then it's at 4kHz for both the woofer and the tweeter? The reason I ask is that in Tuxedocivic's active filter design it says:

Woofer: BW5 @ 2400Hz
Tweeter: BW5 @ 3400hz

Perhaps a typo?
  Reply With Quote
Old 19th February 2013, 06:44 PM   #18
diyAudio Member
 
Join Date: Jan 2008
Quote:
Originally Posted by soren5 View Post
Just one thing I'm not sure about:

Is the crossover always the same frequency for the two drivers crossing? So if it's at 4kHz then it's at 4kHz for both the woofer and the tweeter? The reason I ask is that in Tuxedocivic's active filter design it says:

Woofer: BW5 @ 2400Hz
Tweeter: BW5 @ 3400hz

Perhaps a typo?
BW5 appears to be a typo for BW6, unless MiniDsp has some filter I have never heard of before.

Underlapping ( leaving an electrical "hole") crossover points is often required to get the desired acoustical crossover response.

You should note that Tuxedocivic later posted new settings that are quite different than the ones posted previously after actually testing the acoustic response. I have posted them below.

Whether they match the acoustic response of the original crossover is unknown without seeing the individual response of the low and high driver output of using the passive crossover.

Duplicating the electrical response of a passive crossover ignores the speaker parameter interaction, one really needs to measure the acoustical response to duplicate the acoustical response curves.

That said, since time alignment with passive components is difficult to implement, but easy with DSP, one can generally improve response using DSP compared to passive.

Without good test gear and knowledge of how to use it, one never knows how far off simulations are.

Art
Attached Images
File Type: png Revised settings.png (26.0 KB, 75 views)
  Reply With Quote
Old 19th February 2013, 10:38 PM   #19
soren5 is offline soren5  Denmark
diyAudio Member
 
Join Date: Jun 2011
Location: Copenhagen
Thanks for the comments. I learned something today

Quote:
Originally Posted by weltersys View Post
Duplicating the electrical response of a passive crossover ignores the speaker parameter interaction, one really needs to measure the acoustical response to duplicate the acoustical response curves.
I begin to understand the problem with the idea I had. You can't make a digital filter simply based on a passive crossover design that you see on the computer screen.

Quote:
Originally Posted by weltersys View Post
That said, since time alignment with passive components is difficult to implement, but easy with DSP, one can generally improve response using DSP compared to passive.

Without good test gear and knowledge of how to use it, one never knows how far off simulations are.
I need a microphone and I need to read more about the miniDSP possibilities I think:

PROTOTYPING A 4-WAY OPEN-BAFFLE SPEAKER WITH THE MINIDSP 24

REFINING A 4-WAY OPEN-BAFFLE SPEAKER WITH THE MINIDSP 24
  Reply With Quote
Old 20th February 2013, 12:45 AM   #20
Jay is offline Jay  Indonesia
diyAudio Member
 
Join Date: Feb 2003
Location: Jakarta
Quote:
Originally Posted by soren5 View Post
Is the crossover always the same frequency for the two drivers crossing? So if it's at 4kHz then it's at 4kHz for both the woofer and the tweeter? The reason I ask is that in Tuxedocivic's active filter design it says:

Woofer: BW5 @ 2400Hz
Tweeter: BW5 @ 3400hz

Perhaps a typo?
If the frequency is different then it is not crossover but passover

There is standard terminology for crossover frequency, especially for standard filter (LR, Bessel, Butterworth, etc) such as the -3dB position of the acoustical roll off.

If you apply the same slope type for both tweeter and woofer (symmetry) then they should cross somewhere around -3dB to -6dB (acoustically) but not too much. Especially for steep slope such as BW6 it is not a good thing to cross at 2K4 for woofer and 3K4 for tweeter. I'm here assuming that MiniDSP uses the standard filter terminologies in the software.

BTW, +3dB or -6dB at XO frequency is not unusual. But for standard LR4 it should be theoretically 0dB (flat).

From the passive XO you can see that the woofer has issues around the XO, which is why the capacitor across the inductor. I don't know how the inside of the MiniDSP work, but it seems similar to analog. In this case you cannot do tricks like in passive XO. So the 2K4 is possible as a active XO trick but is not a good approach imo.

So you see, DSP is not as easy and simple as people might think. Yes, it makes life simpler for those who doesn't know much about passive design. For those who knows much, may be not.
  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
DAC output filter...passive or active? soundnovice miniDSP 4 14th June 2012 10:05 AM
active filter from rod elliot and sub filter questions wuffer Parts 0 2nd October 2005 01:12 PM
dB loss by using passive crossovers? Active vs Passive and 1st vs 4th order Hybrid fourdoor Multi-Way 3 11th July 2004 09:16 PM
passive filter for sub moi Multi-Way 34 28th July 2003 02:12 PM


New To Site? Need Help?

All times are GMT. The time now is 02:36 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2