Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Raimonds.Skuruls]

Dear Colleagues,

May I try to help to find solution of this topic problem?
First, I would like to introduce you to solution to visualize delays in our loudspeaker systems.
Second, let’s see on hardest case – heavy filtered subwoofer.

Let’s start with 3 way, 2nd order crossover.  (picture 01_CR3w_12.png)
It is creating just 0.9 ms delay on 100 Hz. (picture 02_CR3w_12okTDApl.png)
Let’s tune out delays of crossover bands. The relative delay of each band is easy readable from the graph.
(picture 03_CR3w_12_noTDApl.png)

What about 3 way, 8th order (48dB/oct) crossover. (picture 04_CR3w_48.png)
As you see, it is introducing serious amount of delay. (picture 05_CR3w_48okTDApl.png)
Let’s tune out delays of crossover bands. (picture 06_CR3w_48noTDApl.png)

And now, 2 way system with a heavy filtered subwoofer. (picture 07_CR2wSS_48.png)
Its delay picture  08_CR2wSS_48okTDApl.png
And impulse response of the crossover subwoofer band - 09_IR_LF40_100.png
Let’s introduce some additional delay   -  10_CR2wSS_48noTDApl_1_2.png

And solution.
The best delay tuning result, you can get, will be just the crossover delay response.
For such subwoofer case, as described, it is impossible to read small delay step on crossover frequency point. Therefore, you must introduce significant (30 ... 40...50 ms) additional delay to separate bands in graph. Than measure real delay step by use of data cursors (picture 10_CR2wSS_48noTDApl_1_2.png) and subtract this result from your initially introduced additional delay. You may need final tweaking in range - 90 degrees (-2.5 ms for 100 Hz) from such result by observing AFR at the crossover frequency.

But, why do we need so heavy filtered subwoofer with such huge timing (delay) distortions.
This problem is more serious for band pass subs... You can see even 150 ms of delay.
Of course, it is ok for a loudspeaker that is intended to reproduce just low frequency sound effects.

BR,

Raimonds

[Raimonds.Skuruls]

next pictures

[Timo Beckman]

So i did a blog post a few days ago where i explained a little about how you might eq a system using smaart7.
Since i also measured the output of the processor you can see the effect eq filters have on the live ir.
The measurements on the low driver in that particular system and it's processor output show the ir being stretched more in time with every additional eq filter i put in (as expected) and at the last screen shot on the low driver it moves because i implemented a low pass filter as was also expected.

http://timobeckmangeluid.wordpress.com/2014/02/12/eqin-a-system/

Now would i use the ir for the final alignment : no i would have to phase align the whole thing in the end.
This was a low driver and as you can see the lp filter was set at 1200Hz. If you implement a low pass at say a 100Hz on a sub-woofer the same thing will happen (only the values will be a bit more then at 1200Hz:-).

[Ivan Beaver]

Hi Folks,

Thanks for the great info. I was tying to find delay/propagation times for the different sub-systems I was measuring, hence the use of Impulse Response. I understand now why this is an inappropriate approach, as the energy from the sub is so spread out in time it's very difficult to see what's actually going on in IR. Thanks again for your help.

Which would be more appropriate for measuring phase and slope between these pass bands? Noise/sine/program? Sine wave at the crossover freq in the processor? Or pink noise in the octave 63-125hz, covering the crossover area? Sub heavy music?

Cheers,
Joey

Sine waves will only be one freq at a time-Very little to see there.

Swept sine takes awhile to see results of changes.

Music will not energize all the freq-and it doesn't matter if it is bass heavy or not.

Pink will energize all freq and is the best to use in my opinion.

[Tim McCulloch]

Hi Folks,

I'm hoping someone may help me learn to use Smaart to align  subs to mains systems.

As I understand it, we can’t time align a subwoofer to the mains, as subwoofers are stretched over time per frequency. So we cannot use the Impulse Response function in Smaart 6 to accomplish this. However, when attempting to measure with Impulse Response, the arrival times of the mains and subs at the mix position, I ran into some issues I hope someone can help me understand.

Measuring the arrival time of the mains at the mix position seemed to be very easy. I simply set a level, chose the IR Analysis page, started the Signal Generator using pink noise, and started the IR Analysis. The resulting trace was an easily discernible trace showing arrival time and positive polarity. I utilized the following settings in IR analysis: Type – Linear IR, FFT – 64k, TC (ms) – 1365, Avg – 4.

I then attempted to measure the subs arrival at the mix, with the same settings and method, but failed miserably. I tried increasing the FFT size and averages with no success. The trace that resulted was a nearly flat line, with no way to determine polarity or arrival time.

I hope someone who sees this post will know why I was unsuccessful in my first attempt. I must say I am not completely surprised at the result, as the more I attempt to come to terms with this very powerful tool the more I realize how little I know regarding its use.

Best Regards,
Joey

edited for spelling

Joey, how many threads on how many different sites?

[Ivan Beaver]

Joey, how many threads on how many different sites?

You know that way you get all different kinds of answers-so it makes it confusing to "piece them al together".

[Ryan Hammond]

I use a polarity tester from JL Audio Tools.

[Mac Kerr]

I use a polarity tester from JL Audio Tools.

You might be able to match the polarity of the subs to the mains, but a polarity checker will not help at all in aligning the phase.

Mac

[Ivan Beaver]

You might be able to match the polarity of the subs to the mains, but a polarity checker will not help at all in aligning the phase.

Mac

Phase-polarity-it is sadly the same to most people (((

YES THERE IS A DIFFERENCE-AND it is IMPORTANT!

Think of it this way.  2 cars (lets use NASCAR for example) cross the finish line at the same time.  They are in "polarity".

Which one wins?  The one that is has the most "phase shift" (laps around the track).  The losing car may be laps behind-but they are in the same physical position (polarity).

I guess it "doesn't matter" to some people-but it SURE DOES to the guy who gets the check!!!!

[Cailen Waddell]

Phase-polarity-it is sadly the same to most people (((

YES THERE IS A DIFFERENCE-AND it is IMPORTANT!

Think of it this way.  2 cars (lets use NASCAR for example) cross the finish line at the same time.  They are in "polarity".

Which one wins?  The one that is has the most "phase shift" (laps around the track).  The losing car may be laps behind-but they are in the same physical position (polarity).

I guess it "doesn't matter" to some people-but it SURE DOES to the guy who gets the check!!!!

So Ivan - a somewhat ignorant question.  If both cars cross the finish line at the same time (amplifying each other) but one is a lap behind, how do you see that on measurement software?  On an RtA it would follow that things 'look right'. Yes?  Or am I missing something critical (I probably am).

[ProSoundWeb Community]