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

rePhase, a loudspeaker phase linearization, EQ and FIR filtering tool
rePhase, a loudspeaker phase linearization, EQ and FIR filtering tool
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 8th January 2017, 12:28 PM   #1481
wesayso is offline wesayso  Netherlands
diyAudio Member
 
wesayso's Avatar
 
Join Date: Jul 2007
One might wonder where the ringing is worst in the equipment.
Load that equipment IR into REW and look at the (very) early waterfall plot. I bet it will only show ringing at very high frequencies. Close to it's cut off frequency.

Weather it shows in actual measurements would still be a question, Though your high frequencies are extended enough.

To add another thought, the IR you show is DAC or soundcard out to Mic/Line in, right? Did you apply any calibration to it?
The reason I ask is I had my mic calibration still on (as for regular measurements), but the mic isn't inline. To show my real IR I had to remove that Cal file in REW.
__________________
Use Science to design your speakers and they will sound like a piece of Art...

Last edited by wesayso; 8th January 2017 at 12:45 PM.
  Reply With Quote
Old 8th January 2017, 01:39 PM   #1482
3ll3d00d is offline 3ll3d00d  United Kingdom
diyAudio Member
 
3ll3d00d's Avatar
 
Join Date: Jan 2014
Quote:
Originally Posted by wesayso View Post
The reason I ask is I had my mic calibration still on (as for regular measurements), but the mic isn't inline. To show my real IR I had to remove that Cal file in REW.
How do you mean? As far as I recall, the cal file is only applied to the FR.
  Reply With Quote
Old 8th January 2017, 01:39 PM   #1483
jtalden is offline jtalden  United States
diyAudio Member
 
Join Date: May 2009
Location: Michigan
I spent some time creating a 44.1k measuring system calibration IR (soundcard, Pre-Pro, DCX2496, P-amp, and mic). I had many difficult issues, but did get the expected results on 1 or 2 trials; enough to see that the expected improvement in the appearance of the step response occurred and that SPL and phase were still correct. I never got it refined as a simple reliable process.

For me, the other charts tell the whole story anyway so there was no practical need to get it working reliably. It did make the step response look much better, but once the visual impact on the ringing on the charts was confirmed it could just be ignored. This was my approach for some time.

More recently I shifted my sample rate for the measuring system to 96k so the ringing is gone now. My music is all 44.1k so I just make the rePhase FIR filter at both sample rates and load the one that's needed. The measurements then do not take into account the impact of the 44.1k reconstruction filter on the sound, but that is likely to be a minor concession in terms of impact on sound quality.
  Reply With Quote
Old 8th January 2017, 01:55 PM   #1484
3ll3d00d is offline 3ll3d00d  United Kingdom
diyAudio Member
 
3ll3d00d's Avatar
 
Join Date: Jan 2014
Quote:
Originally Posted by mark100 View Post
Hi, yes..."Estimate IR delay" has become a very routine step ...

Trying to learn more about arithmetic operations, A vs B......do you have any recommended links?

Going to 3ll3d00d's link now....
The Holm impulse manual has a description of each function in lay person terms. Basically A+B is like playing 2 drivers at the same time and measuring the result, A-B is same but 1 driver inverted, A*B is like applying a filter, A/B is like applying a mic cal file (aka like applying the inverse of a filter), (A+B)/2 is averaging but can also be used to copy a single measurement. Bear in mind the window and smoothing from trace A will be applied to the result.

The link was just a description of a process, you'd need to write the code to do this yourself (can be done in python for example)
  Reply With Quote
Old 8th January 2017, 02:08 PM   #1485
mark100 is offline mark100  United States
diyAudio Member
 
Join Date: Dec 2010
Quote:
Originally Posted by wesayso View Post
One might wonder where the ringing is worst in the equipment.
Load that equipment IR into REW and look at the (very) early waterfall plot. I bet it will only show ringing at very high frequencies. Close to it's cut off frequency.

Weather it shows in actual measurements would still be a question, Though your high frequencies are extended enough.

To add another thought, the IR you show is DAC or soundcard out to Mic/Line in, right? Did you apply any calibration to it?
The reason I ask is I had my mic calibration still on (as for regular measurements), but the mic isn't inline. To show my real IR I had to remove that Cal file in REW.
Here's a water fall of the x-32. I'm not sure how to set time settings for best look, so they are shown for redirection....
And I don't know what to look for in terms of HF ringing...sorry to be such a noob here...
The mic cal file was not in play.

I'm confused on how the calibration files fit in with the impulse response, and everything REW generates from the impulse response.

It seems like an impulse response has to be a system impulse...that includes the combined response of all components in the chain.

I know the calibration files, be they soundcard or mic, adjust SPL and phase.
But do they effect anything else?
Attached Images
File Type: jpg x32 waterfall.jpg (113.3 KB, 167 views)
  Reply With Quote
Old 8th January 2017, 02:16 PM   #1486
mark100 is offline mark100  United States
diyAudio Member
 
Join Date: Dec 2010
Quote:
Originally Posted by 3ll3d00d View Post
The Holm impulse manual has a description of each function in lay person terms. Basically A+B is like playing 2 drivers at the same time and measuring the result, A-B is same but 1 driver inverted, A*B is like applying a filter, A/B is like applying a mic cal file (aka like applying the inverse of a filter), (A+B)/2 is averaging but can also be used to copy a single measurement. Bear in mind the window and smoothing from trace A will be applied to the result.

The link was just a description of a process, you'd need to write the code to do this yourself (can be done in python for example)
Thank you,
yes i realized the link was a process...one I couldn't follow further

Your A to B info is very clear. I have Holm and will revisit the help files.
(I quit playing with HOLM due to crashes and inability to load impulses..Win10. Anybody OS versions know to work well?)
  Reply With Quote
Old 8th January 2017, 04:06 PM   #1487
3ll3d00d is offline 3ll3d00d  United Kingdom
diyAudio Member
 
3ll3d00d's Avatar
 
Join Date: Jan 2014
Personally I don't think Holm has anything rew doesn't these days. I have also found it very much prone to crashing.
  Reply With Quote
Old 8th January 2017, 05:36 PM   #1488
wesayso is offline wesayso  Netherlands
diyAudio Member
 
wesayso's Avatar
 
Join Date: Jul 2007
This is what I think I am seeing as the ringing in your IR:
rePhase, a loudspeaker phase linearization, EQ and FIR filtering tool-ringing-jpg
Peak of ringing at 24 KHz.

Just load an EQ tab in REW, have the IR plot show in bottom window and add a low pass filter at 22 KHz, it will cut the amount of ringing in half. This isn't a proposed solution, just trying to show this is the actual cause of those ripples.
The low pass only removed some ringing and created lesser new ones at a lower frequency (the new cut-off). So it's the filter at the top end as BYRTT said. Better filters will show less ringing.

Meaning jtalden's way may be an excellent trick to get rid of it in measurements. Set it at 96 KHz, and the ringing will move up. As long as your speaker doesn't have output all the way up there the measurement will be free from the ringing coming from the sound card. At least I'd expect it to be.
Attached Images
File Type: jpg ringing.jpg (92.9 KB, 299 views)
__________________
Use Science to design your speakers and they will sound like a piece of Art...
  Reply With Quote
Old 8th January 2017, 07:15 PM   #1489
BYRTT is offline BYRTT  Denmark
diyAudio Member
 
Join Date: Dec 2009
rePhase, a loudspeaker phase linearization, EQ and FIR filtering tool
Quote:
Originally Posted by mark100 View Post
...I'm confused on how the calibration files fit in with the impulse response, and everything REW generates from the impulse response.

It seems like an impulse response has to be a system impulse...that includes the combined response of all components in the chain.

I know the calibration files, be they soundcard or mic, adjust SPL and phase.
But do they effect anything else?
Probably i'm a noob too but interested to learn, especially when a system band-pass acoustic gets as high in quality as your latest build then think its interesting to interpret in higher resolution to squeeze the last bit and truth out of system.

Have two of those Cirrus Logic equipped ringing bells myself in UMC204HD and XONAR-U7 but they live on a shelve because of the bell and because AP192 and ESSENCE STX II have far better looking filters and distortion profiles so STX II run my JRiver steered speaker system locked at 192kHz feeded analog by AP192 with JRiver run as player on another computer playing tracks native sample rate (reconstruction filter).


Think below is interesting loop tests run with AP192 at 48kHz (for info phase shift at low end is approximate a 2,4Hz 2. order HP).

1. is raw soundcard without any calibration.
2. is same as point 1 with IR centered commanding "Estimate IR delay".
3. is soundcard calibration and looks exactly same as point 2 except amplitude is shifted down to zero dB.
4. is with calibration file loaded.
5. is same as point 4 with IR centered commanding "Estimate IR delay".

Haha with full calibration and centered IR at HF stopband waterfall gets worst tail of them all probably because the higher order a filter gets the more it rings and with calibration everything is pure flat up to kind of brickwall cut.

Comments ?

Attachments is same (50% and full resolution)
Attached Images
File Type: png 200c.png (251.4 KB, 150 views)
File Type: png 200a.png (482.2 KB, 145 views)
  Reply With Quote
Old 8th January 2017, 07:30 PM   #1490
wesayso is offline wesayso  Netherlands
diyAudio Member
 
wesayso's Avatar
 
Join Date: Jul 2007
Why look at such large window settings if you want to see the ringing?
Set rise time to 0.1 and window to ~1.5 and move a bit up and down in that window time.
Time range can be 3 ms. See it ring if you move trough time of the window 1,5 to 1,6 etc? Move down in time to 0.5 ms too, You'll notice it will stop ringing and stay there at an early point.
Now start the EQ tab, set a low pass on one of the worst looking pulses. Cut it below the ringing top end, Does the IR in the preview window below look better?

In your setup, try to run a test up to 24 KHz (limit the output up high), while leaving the setup at this resolution. (your graphs show 30 KHz). Is there still a ringing visible in the IR?
__________________
Use Science to design your speakers and they will sound like a piece of Art...

Last edited by wesayso; 8th January 2017 at 07:37 PM.
  Reply With Quote

Reply


rePhase, a loudspeaker phase linearization, EQ and FIR filtering toolHide 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
FIR linear phase plugin for MiniDSP? diyjb01 miniDSP 17 9th June 2016 01:35 PM
FIR filter design tool for Loudspeaker magnitude equalization ttmusic Software Tools 3 24th May 2013 08:30 PM
FIR Filtering experiences Olombo PC Based 8 10th February 2013 03:45 PM
AVX based FIR VST, crossover / EQ / DRC and delay KOON3876 PC Based 97 26th November 2012 07:18 AM
Phase EQ using FIR filters Grasso Multi-Way 2 2nd July 2003 10:37 PM


New To Site? Need Help?

All times are GMT. The time now is 01:47 AM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 14.29%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Copyright ©1999-2018 diyAudio
Wiki