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

synergy.....Take #7
synergy.....Take #7
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 22nd May 2020, 08:01 PM   #71
mark100 is offline mark100  United States
diyAudio Member
 
Join Date: Dec 2010
synergy.....Take #7
Quote:
Originally Posted by tmuikku View Post
Mark100, I remember some diy members have critisised the "heavy" FIR processing due to preringing and you have responded it is not much of a concern. I guess the ringing is visible in the impulse response you posted? Have you tried to reduce it by using shallower crossovers or something to optimize between the ringing/overall frequency reponse smoothness and what is the verdict? Great results anyway!
Hi tmuikki, thx

I've also been one of the diy members that criticize over use of FIR processing.
Imo, any high Q, sharp frequency change, over-correction, is to be avoided, whether IIR or FIR.
I'd even go further and say any use of FIR to tune specifically to a spot is over correction, unless you lock your head in a vise
But that's just me...(and it's because I think such tuning always spans xover summation regions which are not minimum phase.)

I do not believe high order linear phase xovers produce any sharp frequency changes.
And high order linear phase don't have the excessive phase rotation that precludes using higher order IIR crossovers.

But the biggie is, I've heard zero evidence of pre-ringing from their use.

The only time I've heard 'pre-ringing' is what I'd call pre-echo...when I was using very long FIR filters across the frequency spectrum...and no doubt over correcting. It was in my early days of learning FIR.
I now relegate pre-ringing to the myth category unless a person uses extreme high Q examples to prove it can be heard.

And I don't think the squiggles you see in the impulse and step i posted are pre-ringing.
I think they are just the digital gack we get in measurements, trying to precisely time all the FFT math needed.
Notice the squiggles are symmetric around t=0. Is that post ringing too? I don't think so, ...
I don't even think it's real, because I've learned if I add a a very high freq 20-22kHz low pass filter to the VHF output, after the FIR filter, the squiggles almost completely go away.
And there is no change in sound, so again, i think they are meaningless inaudible measurement oscillations.
Of course, could be wrong...
  Reply With Quote
Old 22nd May 2020, 08:13 PM   #72
tmuikku is offline tmuikku  Finland
diyAudio Member
 
Join Date: Sep 2004
Location: Savo, Finland
Send a message via ICQ to tmuikku
Cool, because the response after such filter is pretty much flawless and counting in a fast process to implement such filter I don't see any reason not to
  Reply With Quote
Old 22nd May 2020, 10:15 PM   #73
fluid is offline fluid  Australia
diyAudio Member
 
Join Date: Jan 2009
Quote:
Originally Posted by mark100 View Post
I've also been one of the diy members that criticize over use of FIR processing.
Imo, any high Q, sharp frequency change, over-correction, is to be avoided, whether IIR or FIR.
I'd even go further and say any use of FIR to tune specifically to a spot is over correction, unless you lock your head in a vise
But that's just me...(and it's because I think such tuning always spans xover summation regions which are not minimum phase.)
I don't disagree with most of what you wrote in general but sometimes there are occasions where if you are careful you can get a better response at a specific listening position and not have to need to lock your head in a vice.

I target a specific area which covers realistic head and body movement while listening. The first test I do with any correction is move my head around in a circle and if it only sounds good with my head in one position the correction is immediately junked.

How successful any of this can be depends on the directivity of the speakers you are EQing and is another reason why smooth off axis responses are desired.

On first look at your impulse shown I thought the same why is the so much pre ringing, then I saw the shape of it and it clicked, you are right that is a phase delay timing issue not FIR pre ringing.

In a fully symmetric crossover the pre and post ringing on either side will cancel each other out when recombined on axis which was discussed in another thread. The place to look and listen for this will be off axis where they don't combine quite so perfectly.

So there is less of a potential downside for using symmetrical steep slope linear phase filters than there is for a single sided filter such as a steep highpass to protect a woofer.
  Reply With Quote
Old 23rd May 2020, 05:45 PM   #74
mark100 is offline mark100  United States
diyAudio Member
 
Join Date: Dec 2010
synergy.....Take #7
Quote:
Originally Posted by fluid View Post
I don't disagree with most of what you wrote in general but sometimes there are occasions where if you are careful you can get a better response at a specific listening position and not have to need to lock your head in a vice.

I target a specific area which covers realistic head and body movement while listening. The first test I do with any correction is move my head around in a circle and if it only sounds good with my head in one position the correction is immediately junked.

How successful any of this can be depends on the directivity of the speakers you are EQing and is another reason why smooth off axis responses are desired.

On first look at your impulse shown I thought the same why is the so much pre ringing, then I saw the shape of it and it clicked, you are right that is a phase delay timing issue not FIR pre ringing.

In a fully symmetric crossover the pre and post ringing on either side will cancel each other out when recombined on axis which was discussed in another thread. The place to look and listen for this will be off axis where they don't combine quite so perfectly.

So there is less of a potential downside for using symmetrical steep slope linear phase filters than there is for a single sided filter such as a steep highpass to protect a woofer.
Hi fluid, all you say makes sense to me.

I think the idea of an expanded area sweet spot, and smooth polars are kinda kissing cousins so to speak.

And yep regarding pre and post ringing cancelling when fully symmetric and on axis.
Ime/imo, the best way to avoid off-axis issues is by minimizing the frequency range of overlap, hence steep xovers, and hence linear phase.

Below is the same tuning posted earlier with just a mild low pass at 20kHz added. The squiggles clean up and shift their look.
And then the original for easy comparison.

Better cleanup to near textbook is when low pass is set above 20kHz, but q-sys doesn't let me do that.
Both step and impulse improve using 22kHz or above.... then no small dip in step, and virtually no squiggles in impulse.
Like said, I've come to the conclusion it's simply graphical minutia that's immaterial to sound.

imp sand step with 20k lp.jpg
syn7 imp and step.jpg
  Reply With Quote
Old 23rd May 2020, 10:11 PM   #75
fluid is offline fluid  Australia
diyAudio Member
 
Join Date: Jan 2009
Another instance where the impulse is dominated by high frequencies.

Most DACs that use a linear phase reconstruction filter show some of this in a loopback measurement and it is often at or near the Nyquist frequency so almost certainly inaudible.

To prove it to myself I made a synthetic combination of Linear Phase Linkwitz Riley Crossovers 96dB (LR16) at 1000Hz LP and HP recombined using a 44100 sample rate.

I won't show the Frequency and phase because they are flat lines

This is the impulse

synergy.....Take #7-lr16-combined-jpg

The ringing is there but it is periodic the period of the oscillation is 45 microseconds give or take my ability to get the cursor on it in REW.

And 45 microseconds is the period of 22,222Hz. So if I was more accurate with my positioning I would probably have got closer to 22,050. Getting it exact is hard because you are then at two decimal places or more in microseconds to get it bang on.

In this case close enough to show the ringing is at the half sample rate frequency.
Attached Images
File Type: jpg LR16 Combined.jpg (93.1 KB, 196 views)
  Reply With Quote
Old Yesterday, 05:53 PM   #76
MFagains is offline MFagains  United States
diyAudio Member
 
Join Date: Jul 2017
I have a Synergy noob question..... been searching for it having a hard time finding the answer... Should the upper end midrange freq notch be above or below the XO freq [2k] passing off to the CD? Everything else I get....



Mike
  Reply With Quote
Old Yesterday, 07:26 PM   #77
Soldermizer is offline Soldermizer  United States
diyAudio Member
 
Soldermizer's Avatar
 
Join Date: Feb 2012
Location: Tamper, FL, USA
Even I am competent to answer that one!


All unity/synergy horns will have the mid's 1/2 wave cancellation notch. A rule of thumb for the x-over then, is one octave below that, but you have some wiggle room, moving up towards the notch. For example, I have a pair of slightly modded Yorkville Unity U15. My measured 1/2 wave notch is about 1.9KHz. I chose an x-over of 1.0 KHz (a bit lower than stock, I think is 1.5KHz). So my mids will cross at 1.0 Hz, probably could push it to 1.5 KHz. Yes, you want the mid's upper range well below the notch because at the notch frequency you will have reduced output no matter what the power of your drivers
__________________
Joint Probability = P(rolling papers)*P(bag of weed)

Last edited by Soldermizer; Yesterday at 07:30 PM.
  Reply With Quote
Old Yesterday, 07:41 PM   #78
MFagains is offline MFagains  United States
diyAudio Member
 
Join Date: Jul 2017
Quote:
Originally Posted by Soldermizer View Post
Even I am competent to answer that one!


All unity/synergy horns will have the mid's 1/2 wave cancellation notch. A rule of thumb for the x-over then, is one octave below that, but you have some wiggle room, moving up towards the notch. For example, I have a pair of slightly modded Yorkville Unity U15. My measured 1/2 wave notch is about 1.9KHz. I chose an x-over of 1.0 KHz (a bit lower than stock, I think is 1.5KHz). So my mids will cross at 1.0 Hz, probably could push it to 1.5 KHz. Yes, you want the mid's upper range well below the notch because at the notch frequency you will have reduced output no matter what the power of your drivers
I was thinking that was the case just from an output perspective..... thank you for your reply.


BTW.... love your sig..... does that formula also apply to bongs?


Mike
  Reply With Quote
Old Today, 01:45 AM   #79
Soldermizer is offline Soldermizer  United States
diyAudio Member
 
Soldermizer's Avatar
 
Join Date: Feb 2012
Location: Tamper, FL, USA
Not sure. Just a lousy statistics joke. I prefer edibles or liquid product myself
__________________
Joint Probability = P(rolling papers)*P(bag of weed)
  Reply With Quote
Old Today, 01:38 PM   #80
mark100 is offline mark100  United States
diyAudio Member
 
Join Date: Dec 2010
synergy.....Take #7
Like Soldermizer says, "you want the mid's upper range well below the notch because at the notch frequency you will have reduced output no matter what the power of your drivers".

fwiw, I've found with active processing, that as long as the notch is above xover, it works fine.
My builds go straight from a CD to 12"'s or 10"s, in the 500-650Hz xover range, and often don't allow even a half-octave of distance between xover and notch.
  Reply With Quote

Reply


synergy.....Take #7Hide 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
Synergy DIY ellisr63 Multi-Way 62 19th May 2020 07:39 PM
REW tweaking of synergy Droco Software Tools 0 22nd March 2020 11:53 AM
Two Way Synergy??? Patrick Bateman Multi-Way 28 2nd December 2019 11:47 PM
Klipsch synergy sub-10 skwrangler Subwoofers 1 5th March 2014 04:24 PM
Synergy dipole ? bzfcocon Multi-Way 12 24th November 2013 09:10 PM


New To Site? Need Help?

All times are GMT. The time now is 02:05 PM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2020 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 14.29%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2020 DragonByte Technologies Ltd.
Copyright ©1999-2020 diyAudio
Wiki