Car Audio | DiyMobileAudio.com | Car Stereo Forum

Car Audio | DiyMobileAudio.com | Car Stereo Forum (https://www.diymobileaudio.com/forum/)
-   Member Reviews & Product Comparisons (https://www.diymobileaudio.com/forum/member-reviews-product-comparisons/)
-   -   TDA (Time-domain analysis) - Measurement Software - Review (https://www.diymobileaudio.com/forum/member-reviews-product-comparisons/171172-tda-time-domain-analysis-measurement-software-review.html)

brumledb 04-17-2016 06:23 PM

Re: TDA (Time-domain analysis) - Measurement Software - Review
 
Something isn't right. It will let you see the 3D and 2D graphs in the demo mode.


Sent from my iPhone using Tapatalk

Hanatsu 04-17-2016 06:23 PM

Re: TDA (Time-domain analysis) - Measurement Software - Review
 
Something is wrong... Looks like the install failed or something. Did you install the Matlab compiler?


Sent from my iPhone 6 using Tapatalk.

Babs 04-17-2016 06:50 PM

Re: TDA (Time-domain analysis) - Measurement Software - Review
 
Yep. Figured it out. Had to redownload. Now to do some measuring when I can get a chance in a little bit. First plot I discovered my mic had fallen. Yay. Now to get kids to bed and try again.


Sent from iPhone using Tapatalk

strohw 04-17-2016 08:07 PM

Re: TDA (Time-domain analysis) - Measurement Software - Review
 
Finally, tried this app but I'm not getting consistent results with it. Using a uca202 sound card and get consistent repeatable results in rew and holmimpulse. However, with this app one second it's showing +/- 1ms on my 20-80hz and the next it's showing a +5ms smear over a portion of the same range.

brumledb 04-17-2016 08:56 PM

Re: TDA (Time-domain analysis) - Measurement Software - Review
 
Quote:

Originally Posted by strohw (Post 3675002)
Finally, tried this app but I'm not getting consistent results with it. Using a uca202 sound card and get consistent repeatable results in rew and holmimpulse. However, with this app one second it's showing +/- 1ms on my 20-80hz and the next it's showing a +5ms smear over a portion of the same range.

Are you staying in the car for the measurement? If so, try sticking mic into the headrest and take some measurements without being in the car. Hanatsu has mentioned that measuring while sitting in the car will cause inconsistent measurements.

strohw 04-17-2016 09:30 PM

Re: TDA (Time-domain analysis) - Measurement Software - Review
 
I have the mic clamped in the headrest roughly where the center of my head would be. I am sitting in the back seat of my crew cab. That's mainly because I don't have a long enough cable to run the laptop outside the truck.

Babs 04-17-2016 10:34 PM

Re: TDA (Time-domain analysis) - Measurement Software - Review
 
Remember I believe this tool assumes your tweets and mids are darn close to level and EQ matched. This may have an affect I'm sure.

So I messed with it quite some time and not so positive I have it mastered but was able to achieve this.

Two way
SB29/XSPTW 24db @ 1907hz LR
SB17 at 80-2khz roughly (split)
JL 12TW1 @ LR4 75hz

http://uploads.tapatalk-cdn.com/2016...ddae2b96eb.jpg

http://uploads.tapatalk-cdn.com/2016...4210830be0.jpg

So I'll have to say I'm impressed with the stage and coherence. From history with the tune I expected the sub to be all kinds of crazy out of phase and didn't really know what to expect. I'd have to say I'm a believer. Though was unable to get any better than the above plot. I think a lot of the troubles are simply modal and reflection related. And it does really come down to mic placement.

Hanatsu and others I'm wide open for suggestions.


Sent from iPhone using Tapatalk

brumledb 04-17-2016 10:44 PM

Re: TDA (Time-domain analysis) - Measurement Software - Review
 
Quote:

Originally Posted by Babs (Post 3675530)
Remember I believe this tool assumes your tweets and mids are darn close to level and EQ matched. This may have an affect I'm sure.

So I messed with it quite some time and not so positive I have it mastered but was able to achieve this.

Two way
SB29/XSPTW 24db @ 1907hz LR
SB17 at 80-2khz roughly (split)
JL 12TW1 @ LR4 75hz

Looking good. I would be curious to see how your sub delay would change by using a 2nd order butterworth.

Babs 04-17-2016 10:47 PM

TDA (Time-domain analysis) - Measurement Software - Review
 
Quote:

Originally Posted by brumledb (Post 3675562)
Looking good. I would be curious to see how your sub delay would change by using a 2nd order butterworth.


Hmm interesting. I'll try it when I get a chance. Yeah it's hard to tell what "good" should look like since we know there's so much destructive phase craziness in car. I could only tweak, measure, tweak and see if it improves or not. I guess it's a completely trial and error iterative thing.

And isn't it strange my temporary sub is literally in the back seat, though by all accounts measuring it acts like it's in the car behind me. Is it the crossover doing this?

thebookfreak58 04-17-2016 11:05 PM

Re: TDA (Time-domain analysis) - Measurement Software - Review
 
So with this tool, measure the entire system all channels playing? Or one side at a time?

Basically adjust T/A on each driver till it locks in?

Babs 04-18-2016 06:18 AM

Re: TDA (Time-domain analysis) - Measurement Software - Review
 
Quote:

Originally Posted by thebookfreak58 (Post 3675586)
So with this tool, measure the entire system all channels playing? Or one side at a time?

Basically adjust T/A on each driver till it locks in?


I ran all drivers. You can tell through the frequency region if tweets are out or mids are out etc. So if pairs (left/right) or sides (tweet/mid/midbass/sub) are out.

I guess the goal is to get all to one perfect vertical line at 0ms.

Alextaastrup 04-18-2016 07:17 AM

Re: TDA (Time-domain analysis) - Measurement Software - Review
 
From the practical point of view and based on psychoacoustics: 1-2ms should be just fine. For tweeters the magnitude (volume of the sound) is more important, but keeping them also within the abovementioned band will not spoil the sound stage :).

It is more critical with the low end, where sub enclosures have their own generic delays: Sealed is the best, 6th bandpass - worst. Typical delay of a vented port box - 20-30ms - just to mention one. TA left/right channels might not feet 100% with the subwoofer, so it is a compromise, as always.

strohw 04-18-2016 07:33 AM

Re: TDA (Time-domain analysis) - Measurement Software - Review
 
Using that idea for the sub visualized in this thread helped me get much better integration. In my setup the sub is only 40" away which is roughly the same length as the left mid.

I was always testing 0+ ms. After seeing the results here I tested the other way and started at 0 and delayed everything else. I had 12ms left that I could use and after doing a lot of testing I found -8.5ms measured the best in REW. I sat there and measured 1ms at a time and each measurement showed a pretty significant change in the 70-100hz range where the subs/mids integrate.

Babs 04-18-2016 07:40 AM

TDA (Time-domain analysis) - Measurement Software - Review
 
Quote:

Originally Posted by strohw (Post 3676010)
Using that idea for the sub visualized in this thread helped me get much better integration. In my setup the sub is only 40" away which is roughly the same length as the left mid.

I was always testing 0+ ms. After seeing the results here I tested the other way and started at 0 and delayed everything else. I had 12ms left that I could use and after doing a lot of testing I found -8.5ms measured the best in REW. I sat there and measured 1ms at a time and each measurement showed a pretty significant change in the 70-100hz range where the subs/mids integrate.

Quote:

Originally Posted by Alextaastrup (Post 3675962)
From the practical point of view and based on psychoacoustics: 1-2ms should be just fine. For tweeters the magnitude (volume of the sound) is more important, but keeping them also within the abovementioned band will not spoil the sound stage :).

It is more critical with the low end, where sub enclosures have their own generic delays: Sealed is the best, 6th bandpass - worst. Typical delay of a vented port box - 20-30ms - just to mention one. TA left/right channels might not feet 100% with the subwoofer, so it is a compromise, as always.

Folks will say TA in bass region isn't very important, and phase is everything. I dissagree.. I believe doing this little exercise did something quite special in the bass region. I guess "coherence" is the best term I can come up with, but a kick drum is awesome. It's not bloated or undefined. I should clarify in-phase AND time-aligned = very cool!

Can't wait to do this with the two IB12AU's.

Strangely though, I couldn't do much at all about that 50hz and below so do I assume that's modal and un-fixable or rather group delay related with the sub itself?

Hanatsu 04-18-2016 08:58 AM

TDA (Time-domain analysis) - Measurement Software - Review
 
T/A is phase shift...

Linear phase shift vs freq ;)

But you're right, delay in LF IS important


Sent from my iPhone 6 using Tapatalk.

Babs 04-18-2016 09:10 AM

Re: TDA (Time-domain analysis) - Measurement Software - Review
 
Quote:

Originally Posted by brumledb (Post 3646329)
This is the best I have been able to achieve. But as you can see my sub is 20+ ms out and I am out of delay on my tweeters. The Helix only goes to 15.6ms. So any suggestions on what I should try?
Right now I using LR 24's for all crossovers. Should I try using a lower order crossover and maybe Bessel or Butterworth instead of LR?

[IMG]http://i284.photobucket.com/albums/l...psyrugpgvt.png[/IMG]

Quote:

Originally Posted by Hanatsu (Post 3646377)
Acoustic crossovers are what matters, see the crossovers in your DSP as response shaping. There seem to be sub/mid integration issues as well as T/A and/or L/R FR inconsistancies. I would do individual measurements on all driver by RTA/noise so you get as much averaging as possible. Then compare all the individual measurements to a full left side+sub and a full right side +sub, check if all drivers sum at crossovers. Check actual slopes. Do a sweep of left side and one of the right side in RoomEQ for example, check excess group delay for large peaks which indicates destructive interference.

If the L/R EQ is set correctly you should be able to set T/A by ear very easy with correlated pink noise. I can measure my car tomorrow and show you why some of these issues occur.


Sent from my iPhone 6 using Tapatalk.

Quote:

Originally Posted by brumledb (Post 3646473)
Ok, sounds like I just need to start back at step one. I want to change my house curve anyways. I will go back and make sure I have everything dialed with with EQ before trying to utilize TDA.

brumledb, How'd this work out? You mention "back to formula".. Definitely I'd agree individual driver plots to assess "balance", and I'd encourage don't be afraid to split up your crossover points in the Helix between left/right sides if need be. My mid low-pass's are 1734 and 2244hz for example. Strikingly different electrical xo points, but they match acoustically. Two goals.. 1. Have their up-slopes and downslopes matching. 2. Individual driver EQ and levels matching between left/right sides. You want left and right individual plots approaching the impossible, being both looking like one line on top of each other.

:D Let's see the success!

Babs 04-18-2016 09:13 AM

Re: TDA (Time-domain analysis) - Measurement Software - Review
 
Quote:

Originally Posted by brumledb (Post 3646329)
This is the best I have been able to achieve. But as you can see my sub is 20+ ms out and I am out of delay on my tweeters. The Helix only goes to 15.6ms. So any suggestions on what I should try?
Right now I using LR 24's for all crossovers. Should I try using a lower order crossover and maybe Bessel or Butterworth instead of LR?

[IMG]http://i284.photobucket.com/albums/l...psyrugpgvt.png[/IMG]

Quote:

Originally Posted by Hanatsu (Post 3646377)
Acoustic crossovers are what matters, see the crossovers in your DSP as response shaping. There seem to be sub/mid integration issues as well as T/A and/or L/R FR inconsistancies. I would do individual measurements on all driver by RTA/noise so you get as much averaging as possible. Then compare all the individual measurements to a full left side+sub and a full right side +sub, check if all drivers sum at crossovers. Check actual slopes. Do a sweep of left side and one of the right side in RoomEQ for example, check excess group delay for large peaks which indicates destructive interference.

If the L/R EQ is set correctly you should be able to set T/A by ear very easy with correlated pink noise. I can measure my car tomorrow and show you why some of these issues occur.


Sent from my iPhone 6 using Tapatalk.

Quote:

Originally Posted by brumledb (Post 3646473)
Ok, sounds like I just need to start back at step one. I want to change my house curve anyways. I will go back and make sure I have everything dialed with with EQ before trying to utilize TDA.

brumledb, How'd this work out? You mention "back to formula".. Definitely I'd agree individual driver plots to assess "balance", and I'd encourage don't be afraid to split up your crossover points in the Helix between left/right sides if need be. My mid low-pass's are 1734 and 2244hz for example. Strikingly different electrical xo points, but they match acoustically. Two goals.. 1. Have their up-slopes and downslopes matching. 2. Individual driver EQ and levels matching between left/right sides. You want left and right individual plots approaching the impossible, being both looking like one line on top of each other. Looks like most of your issues is in the region of those 10F's. Mute the sub and forget about it for a while, until you make the scans and GZ front-stage line up like little ducks.

Caveat.. I really barely know enough about this to be a danger to myself and everyone around me. LOL!

:D Let's see the success!

brumledb 04-18-2016 09:24 AM

Re: TDA (Time-domain analysis) - Measurement Software - Review
 
Unfortunately I haven't had time to do any tuning since posting that. Mainly because I can't tune a little at a time, usually when I start I spend hours on it. I did do a little by ear just to get the stage centered. It doesn't sound horrible right now. The bass has a good impact up-front.
About the cross-overs, until recently I thought just set it and forget it with LR4. I didn't consider that different slopes or roll-off points should be used to electrically alter the speakers response to acoustically match whatever cross-over slope you want to attain. I have been doing a good amount of reading about cross-overs lately so now I understand this a lot better. I should be able to get some tuning in later this week, so hopefully I'll have some new and better graphs to post.


Sent from my iPhone using Tapatalk

Babs 04-18-2016 09:53 AM

Re: TDA (Time-domain analysis) - Measurement Software - Review
 
Quote:

Originally Posted by brumledb (Post 3676530)
Unfortunately I haven't had time to do any tuning since posting that. Mainly because I can't tune a little at a time, usually when I start I spend hours on it. I did do a little by ear just to get the stage centered. It doesn't sound horrible right now. The bass has a good impact up-front.
About the cross-overs, until recently I thought just set it and forget it with LR4. I didn't consider that different slopes or roll-off points should be used to electrically alter the speakers response to acoustically match whatever cross-over slope you want to attain. I have been doing a good amount of reading about cross-overs lately so now I understand this a lot better. I should be able to get some tuning in later this week, so hopefully I'll have some new and better graphs to post.


Sent from my iPhone using Tapatalk


First thing I'd do is forget everything else and just do individual plots on your mids and make those corrections to get their individual plots looking more identical in REW. Then go at TDA again with their timing in mind and an idea of their distance differences.

Alextaastrup 04-18-2016 10:01 AM

Re: TDA (Time-domain analysis) - Measurement Software - Review
 
After adjusting L/R separatelly i tried to make a combined test with both channels playing. And got some nulls and cancellations. So it became nesessary to make further improvements. It should work for you also with the help of TDA.

brumledb 04-18-2016 10:43 AM

Re: TDA (Time-domain analysis) - Measurement Software - Review
 
Quote:

Originally Posted by Babs (Post 3676634)
First thing I'd do is forget everything else and just do individual plots on your mids and make those corrections to get their individual plots looking more identical in REW. Then go at TDA again with their timing in mind and an idea of their distance differences.

Yeah, I am starting back at square one. I am going to focus on making sure I have everything dialed in with REW before trying to utilize TDA again.

I'll probably post over in the REW dump thread once I get to work on it. Try to get some opinions and make sure everything is copacetic before proceeding further. I may be going to the big SQ show at College Station in June, so I have to get a good tune in, lest I be embarrassed to demo my truck.

And my new Cross Spectrum UMIK just arrived, time to get to tuning!

strohw 04-18-2016 12:01 PM

Re: TDA (Time-domain analysis) - Measurement Software - Review
 
I'll have to check a full system rta when I get a chance. Usually, I can get 1.5k-20k for both channels to overlap with very nice accuracy. But after I go through and balance for center imaging stuff doesn't look so nice anymore. I'm curious how they're combining overall. I hate not having a peq for the mids though.

Alextaastrup 04-21-2016 03:33 PM

Re: TDA (Time-domain analysis) - Measurement Software - Review
 
A new version of TDA no.11 came out from the Raimonds's table. Now it can also show phase graphs.

Hanatsu 04-21-2016 11:40 PM

Re: TDA (Time-domain analysis) - Measurement Software - Review
 
I'll try the new version out later ;)


Sent from my iPhone 6 using Tapatalk.

Alextaastrup 04-22-2016 02:29 AM

Re: TDA (Time-domain analysis) - Measurement Software - Review
 
If one has a capacity limit in TA, what is the best solution? MiniDSP? Others?

I am looking for DSP or simply cheap time machine with more than 20ms.

Has anybody tried to ask Raimonds about limitation in the APL1 soft (less than 5ms)? Is it possible to increase this figure or not? Are there any soft upgrade for the APL with better TA possibilities?


All times are GMT -6. The time now is 12:11 PM.

Powered by vBulletin®
Copyright ©2000 - 2019, Jelsoft Enterprises Ltd.
Search Engine Optimization by vBSEO 3.6.0
vBulletin Security provided by vBSecurity v2.2.2 (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.

Ad Management by RedTyger