DiyMobileAudio.com Car Stereo Forum banner

41 - 60 of 73 Posts

·
Registered
Joined
·
4,120 Posts
Discussion Starter #41
Here is a link to Swiss bears from DIy complete write up to do your own room correction using rephase .

This process works great, probably would want a opendrc or a few of them to get it done in a car. But this is the manual way to do a room correction


http://forums.melaudia.net/attachment.php?aid=22240
 

·
Registered
Joined
·
4,120 Posts
Discussion Starter #43
Cool a new forum, that I can read :p
Is it from POS?
Yeah, that guys is complete brainiac and super kind. He helps everyone, and has plenty of patience for noob questions. Super awesome dood.

But yeah I just got caught up on the 115 pages of thread reading I missed
 

·
Registered
Joined
·
4,120 Posts
Discussion Starter #44
one Thing rephase is very useful in a car is getting any unwanted ringing to go away.


If I turn off all eq and processing and listen to a speaker play in the environment it’s in
Via a door or pillar or a box or kick or anything, I’ve noticed any colorations caused by the speakers mounting almost always has phase issues that a simple adjustment might make that location better or enclosure better or pod or door etc


You can definitely do near field measurements the see exactly however
If you have a ringing from a location you can move the phase back a little and heat it go away.


It’s a good idea to unmount your speakers off there mounting and measure the speaker in free air and try to replicate that , it’s pretty easy to hear frequencies when there delayed or advanced after a few go’s at it and after the correction when you fix it and it sounds and measures better,

I would suggest most of the eq done in the 120hz to 400hz range isn’t necessary if you have a flat playing driver and a decent baffle and when the phase is fixed the peaks go down and the speaker plays like it should...


Even the best of installs ever would still need a little rephase in a car to just fix that last little bit.

It works amazingly well.

I’ll try and post some measurements
 

·
Registered
Joined
·
4,120 Posts
Discussion Starter #45 (Edited)
Okay gang here it is

This is the article that explains meaningful
Phase and transfer function and magnitude,

The stuff I’ve been saying for about a year but falls on def ears
It took me a long time of figuring shit out and this kinda explains it all in one read


Pay the fu&$ attention to it when your reading tho it’s easy to let your mind wander. If you listen to what this man is explaining it all makes sense and than you guys might realize that rephase actually is awesome companion with REW and you don’t need no silly time domain software if you have a clue on how it all works together. REW is very good software by itself

This is why transfer function analysis is more important and must be considered when giving advice to someone about there system. The time domain can not be ignored especially in a car. Once you realize what it all means than statements like
“Purposely moving time delay +/- .5ms on a speaker away from tape measure distance can be a good thing” because you don’t know cyclically where the speaker your “lining it up to” is at unless you have a understanding of how to read a transfer function to begin with.

This is the kind of stuff that makes us better


https://www.prosoundtraining.com/2016/11/14/meaningful-loudspeaker-phase-response/
 

·
Registered
Joined
·
4,120 Posts
Discussion Starter #47
Thanks for sharing that. I've never read it despite being a member. Lol.
Definitely!

And thanks for all the thanks everyone. It took some diggin to find that one.



I remember I posted a pair of transfer functions about a year ago (I think it was a Dirac live correction for left and right taken in REW) it had a three wraps in the phase but was all beautiful downhill. And left and right looked very similar. Anyway , I remember someone said something like ~”what exactly am I looking at” and “I thought the phase was supposed to be flat”

Getting flat phase on a graph isn’t exactly what we’re trying to achieve exactly,
It’s more about getting left and right to have the same phase and magnitude shape from the listener spot. The transfer function. Which placement and aiming of the speaker strongly makes it achievable. Sometimes even with massive amounts of dsp some parts just can’t be fixed and still either 1. Sound good or 2. Get very loud.

A lot has been written that a speaker radiates below beaming at all directions and it’s dosent matter how it’s aimed. Well that kind of logic is very easy to prove true but unfortunately isn’t how things pan out. The aiming of a speaker absolutely contributes to its transfer function. Getting symmetrical Impulses before any dsp put a car leagues ahead of the game


The electromechanical nature of a transducer and the amount of air it needs to displace per given input signal and the motion of the speaker is sorta in my mind how the time domain starts at the acoustical level.

If a speaker is moving 1mm at 1k (depending on a whole lot of electromechanical variables) to get to zero degrees at 1k (based on the input signal at that specific magnitude) if more power is given to that speaker at 1k it moves positive in phase (because it’s getting louder and going uphill) when you look at it from right to its peak , however when you look at it from its peak down the phase becomes negative. That is the angular part of the phase that gives a minimum phase peak have those properties in its phase response. So if the amplitude is raised the speaker has to move faster to stay vibrating at 1000hz because it has to move farther from peak to peak because it has to move the suspension farther to make the amplitude greater. Longer travel makes the walls (or axis) of each angle longer as well. Imagine a triangular shape that gets bigger and bigger the louder you make a frequency compared to the other frequencies around it. So even though the speaker is moving 1000x a sec it’s amount of travel is increased to make up for the higher amplitude. We all know a speaker is not a perfect piston at all amplitudes and frequencies however usually does a very good job in its operating range. It’s the non linearities in the speaker and it’s suspension that affects the time domain and it suggests that those differences is where a loudspeaker gets parts of its timbre or sound and subsequent phase response when measured closely other than its own natural reverbance from its build materials. So arbitrarily saying a speaker radiates the same as long as it’s not beaming is false. The cone itself size dictates beaming and the cone itself is one of the axises on the angles of which it’s displacement can produce either faithfully or not.
 

·
Registered
Joined
·
715 Posts
This is the best thread I've read in the past few months.
 

·
Registered
Joined
·
618 Posts
Apologies if this has been answered but can I use this with the C DSP 6to8? Would it need some additional hardware?
 

·
Registered
Joined
·
4,120 Posts
Discussion Starter #51
I am officially off Dirac live
And have migrated fully to opendrc and ran my 1st full tune via rephase room correction. So far , better. I can officially beat Dirac live now. (Finally)

And it takes hundreds of measurements and time and patients
But godam. It’s very nice to choose which path to take to my liking.

I think this thread will become very active in future days.

My friend on DIY rephase thread gave me the missing link to extreme sq on my own tunes

Cascaded firs. Everything has the exact same polars as if nothing had a crossover turned on at all.


I will go into depth soon. I need to finalize a few things first



Out with Dirac sharc hello opendrc @12,288 taps





upload pics
 

·
Premium Member
Joined
·
676 Posts
...it takes hundreds of measurements and time and patience..

My friend on DIY rephase thread gave me the missing link to extreme sq on my own tunes

Cascaded firs. Everything has the exact same polars as if nothing had a crossover turned on...

Out with Dirac sharc hello opendrc @12,288 taps
Sounds interesting! Looking forward to this.
:snacks:

But wanted to check...
Does the process require hundreds of measurements? :eek:
or
Have you mastered the process after hundreds of measurements? :scholar:
 

·
Registered
Joined
·
4,120 Posts
Discussion Starter #53
Sounds interesting! Looking forward to this.
:snacks:

But wanted to check...
Does the process require hundreds of measurements? :eek:
or
Have you mastered the process after hundreds of measurements? :scholar:

Thanks ,

So the cascade wasn’t working as well as I thought it would, it works in therory perfect from his pics of xsim it’s awesome. But it seems my HDs lack the power to pull it off and the way the cascade is blocked out I don’t have anywhere near enough fir. I would need a opendrc on each pair. So 5 opendrc boxes to pull it off
In my 10ch system.

So I’m working on doing the room correction now. As of now the mids and highs sound exactly like Dirac. It’s very very close.

The sub integration is a little tricky I’m getting help on it tho from some smart fellas.


I dare you to say smart fella as fast as you can. Say it faster !
 

·
Registered
Joined
·
4,120 Posts
Discussion Starter #54 (Edited)
Got the sub dialed in better than Dirac ......only tiny bit better but still better for me


Here’s what’s up.

I have to thank BTRTT and FLUID on the DIY side of home audio for the help super great guys!


I basically asked a few questions and pretended to be more noob than i am to try to get them to talk to me in a scaled down way, as these guys are way smarter than I am and say some things that are still a bit Chinese to my book of terms.


So after a few friendly posts I learned how to get my sub completely nailed down to blend with the midbass in a way that is just unparalleled.


Here’s how

1. The room gain adds a tremendous amount of boost in the subsonic range all the way down to DC. In fact there’s a chance that your “room mode” at 50-120hx (wherever it is) isn’t a room mode at all, it very well could just be you have so much room gain below that spot. For example a 70hz dip could be side to side cancellations from doors and not necessarily the room. Not always. There still is room modes that are related to the size of the car , however the big one sometimes may not be as intimidating as it looks once the proper sub eq is performed.

I had to use a 6db BW @ 20hz to straighten the responce to DC
This subsonic range being eq really makes the bass get in time with the music
It has something to do with a transfer. So this subsonic must be gloabal and also all sub eq work must be done gloabally.

2. The phase

You basically take a single measurement at listen position
Remove the time of flight. REW impulse centering works great.
Generate minimum phase
Checkbox off the minimum phase version so only excess phase and measurement is on the screen
Than export the measurement as txt
Than import into rephase
Make corrections
And move excess phase to match the minimum phase plot.

Generate the fir and load into DSP.

You can see in this measurement the phase at 80hz
That is a reflection or boundary. (The car shape and size and it’s loading characteristics) below that the phase moves up due to the subsonic rolloff.
The natural rolloff should be left. There is no reason to linearize natural roll off.
The excess phase can be manipulated to better correspond with the minim phase and the response

 

·
Registered
Joined
·
1,320 Posts
You say better than Dirac, which version?

Looks like you need to make a video! :D

Thanks for sharing your learnings. You are persistent and we’re very fortunate you are.
 

·
Registered
Joined
·
4,120 Posts
Discussion Starter #57 (Edited)
You say better than Dirac, which version?

Looks like you need to make a video! :D

Thanks for sharing your learnings. You are persistent and we’re very fortunate you are.

Both.

So I need to go slow on this because we all know I get ahead of myself.

So I’m starting with the sub and working my way up and we’ll all see how the transfer function will sum beautifully on all combined in a 4 way.


So better, yes. And this is why. Dirac is excellent. It does a downright amazing job for an auto tune. It’s so very very good at what it can do. It’s only shortcoming is it has rules that are fixed.

We all know in tuning there’s more than one way to achieve the same measured outcome. One way may have better sonic attributes and are more musical than technical. Like weather to boost or cut, which route to take to achieve a given result. Weather to treat a reflection with phase or not, weather to leave the high Q dips alone or not, or even work on them a little. The software works within its framework of rules period. Although laid out extremely well, arguably the best room correction software and easiest to use for a complete auto tune experience that gets you tuned extremely well in a tiny amount of time.

I’m okay with taking the time to take all the measurements required and I like that I have the say in what is corrected or not.


I’ve spent 4hrs on the mids and highs and I haven’t even scratched the surface and honestly it sounds the same as Dirac, but it sounds the same as my very best Dirac tune I’ve ever had. And that’s without really spending as much time as I know I could have.

So I stopped there and started with the sub, that was the hardest to get right and now I got it. That subsonic really massively made a big difference.


I’ll try and quote his post from another forum so we can understand what he was saying about it. Very interesting and I can’t wait till I can tune as good as this guy. Frikin brilliant guys on this forum.
 

·
Registered
Joined
·
4,120 Posts
Discussion Starter #58
Here’s the important parts of that conversation....super awesome stuff





so here is some measurements of just the sub at listening position





















host photos online





so I did this in rephrase







First will say remember room is global so don't count its corrections dedicated for the lowest pass bands but as global for the summed system passband, and yes room gain is of minimum phase except the reflections that arrive late at listening position, one can repair late reflections but think forget about it because correction only works at that particular spot in space where microphone was positioned. Reflection free is no room but we can get superior sound with a perfect dialed in speaker system corrected for room gain and a house curve that is tweaked to suit the speaker/room profile.


Below is sim model your system HP stopband with the red curve, you can see the numbers in first row, next i set the program WinSpeakerz model for auto cabin response with the numbers in second row. Now look the blue curve it is hot and worthless in you get a flat boost all the way down to DC, this will sound bloated with DC thumbs and mud sound signature way up into higher frq.




Cure for that if WinSpeakerz model is spot on is a linkwitz transform as the first two rows below, first it linearize your HP stopband to flat in row 1 then in row 2 the new 66Hz HP stopband is set, now we are flat down to DC because remember your system is at 2nd order 66Hz corner and auto cabin response boost a 2nd order 66Hz flat down to DC, therefor set some HP filter in area 15-20Hz (15Hz 2nd order used below). Overall correction is the blue curve and red curve is your new speaker plus room HP response, hope it makes sense ...:)



No box linearization for the Q0,83 @45Hz stopband only linerize subs LP filter : )

Here is a block diagram example of 4-way system where i used numbers for your sub because you told those numbers, the rest is for example. As we were around pages back to hinder destructive ripple calculate final target slope per transducer to be a cascade that includes the side band transducers slopes plus stopbands exactly as in block diagram. As you see stopbands are of minimum phase and not linearized so we end up a true system sum minimum phase curve as was it a single point wide band transducer 45Hz(Q0,83)-20kHz(Q0,707).



Curves for above block diagram, lows stopband will extend enormous into your car so find out that probably bit ragged curve and smooth it best you can, probably also you need set a highpass in 15-20Hz area to get it musical or use a linkwitz transform to transfer Q0,83 @45Hz up or down to whatever better stopband that blend smoother with the car cabin, all the room gain corrections should be global and of minimum phase.



Hi Oabeieo,

Know it can be hard to grasp therefor suggest you take some virtual study to educate your self, it is not that hard copy my example pass bands in Rephase export them as frd files import to REW where you can sum them on "ALL SPL" tab using the wonder math it can offer in "Controls" dialog. You will then see example sum perfect as was it one pointsource transducer, and when you do the same exercise but omit side and stop bands you get varius errors in amplitude and phase domains no matter you use FIR or IIR XO slopes.

Understand its a textbook example to get four way sum perfect flat when looked isolated from a real indoor enviroment, at the same we place that system into a indoor enviroment we get tons of extension in low end and ugly ripple from bounderys, but it doesn't change on that the optimal relation between pass bands is as in the example, to count for room gain plus ripple plus dial in the right house curve for system installation shall use global correction to form a perfect integration sum of the optimal textbook multi way with the room enviroment gain signature.

For example say you room enviroment gain signature was a friendly curve looking exactly as was it a linktwitz transform from Q0,83 @45Hz to Q0,7 20Hz, then we was set and happy in system pass band is now a flat minimum phase domain covering 20Hz-20kHz audio band.

Another note is i noticed my box simulation program in good old WinSpeakerz model auto cabin response as a linkwitz transfer from Q0,7071 @66Hz flat down to DC point (0Hz), if that is right think you should try transfer your system stopband in global EQ from Q0,83 @45Hz to Q0,7 @66Hz and set a 2nd order highpass in area 15-20Hz.
That was pretty much the discussion that was had above. If you linearize the sealed box you will change the phase away from minimum and end up with a phase that does not follow the amplitude response.

Measure the result you have and use REW with a FDW to clean up the phase, generate a minimum phase version in REW and compare to what you have. If there's excess phase you can export that and use rephase to try and correct for it and see if you like that any better.
 
41 - 60 of 73 Posts
Top