Home Theater Forum and Systems banner

1 - 9 of 9 Posts

·
Registered
Joined
·
17 Posts
Discussion Starter · #1 ·
Hello HTS and company,

I have long been a fan of DIY but other than build a box and throw a woofer in it and use a half baked off the shelf XO solution, I have been afraid to try to measure speakers and learn what the data means and how to use it... until now! :surprise:

I have built a 2 way system for the theater L C R. They consist of a 3X JBL AS2225STD (2X2226 and 2446/2380a per cab) with the Ti diaphragm removed and a Truextent Be installed in it's place. Amps are Crown DCi 8|600N's (my DSP).

I have had a relatively easy time using REW because of all of the fantastic documentation and just the fact that REW is a really great piece of software, Thanks John! :)

I have measured my center speaker at 1 meter at approximately the height of the listening position if you were to string a string from where the horn meets the woofer on the baffle to the listening position. So now the mic is at about 39" X 50". I have 3 full freq. sweeps with the timing reference as the left channel. One of both drivers, one of the HF and one with the LF. The center delay on both driven is 1.48ms as is the HF only driven sweep delay. The LF delay is 1.65ms. I am still unsure if I am doing it right though.

So my question is: Do I just subtract 1.48 from 1.65 to arrive at .17ms as a delay for these drivers? The XO recommended for the system is 18db Bessel for both HP and LP so mine is at 800hz 18db Bessel.

The weird part to me is that the horn sits all the way to the back od the box and the 2226's have about 5-6 inches of clearance to the back of the box so the delay's I am seeing look backwards to me. Maybe it is the XO's that are providing to current time alignment?

Any help or reading that I can be pointed at would be greatly appreciated.

Thank you in advance! :nerd2:
 

·
Premium Member
Joined
·
1,562 Posts
The XO will delay the woofers. The steeper the XO the more the delay. Possibly the Bes18 is chosen as that is the approximately the ideal delay needed for the driver offsets? The measurements can tell you. You have already taken the correct set of measurements. Those are the 3 needed for close analysis.

Just looking at the SPL overly chart of the 3 traces will give some idea of the timing. If there is good SPL fill in the XO range then the timing is pretty good. The H+W trace should fill the XO range dip shown by the individual H and W traces. A delay can be added to one or the other drivers to fine tune the SPL fill as needed. Increments of 0.1 ms would work pretty well for the 800 Hz XO.

For a more detailed analysis we can look at the phase overlay chart. We want close phase tracking of the 2 drivers throughout the XO range. That process is more complicated to explain. You can check out the following threads to get an idea of what is involved. If you want me to look at your data to confirm the timing and/or make a timing recommendation just post the mdat here.

http://www.hometheatershack.com/forums/rew-forum/104529-phasing-acoustic-xover-point.html#post1024433

http://www.hometheatershack.com/forums/rew-forum/99577-trying-get-my-crossover-setup-2-way.html#post979737

http://www.hometheatershack.com/forums/rew-forum/149385-time-alignment-fir-filter-linear-phase.html#post1449465

http://www.hometheatershack.com/forums/rew-forum/88561-time-alignment-3.html#post913057

http://www.hometheatershack.com/forums/rew-forum/157649-measurement-acoustic-time-reference-2-1-a.html#post1509697
 

·
Registered
Joined
·
17 Posts
Discussion Starter · #3 ·
Thank you very much for the reply jtalden! :)

I will review the directions you provided and read some more at the links as well. I will post back after a few days when I have given some of this a try. I would prefer to understand why I am doing what i'm doing (at least somewhat!) so that your information is not in vain! :smile:

Thanks again!
 

·
Premium Member
Joined
·
1,562 Posts
My understanding is that the purpose of setting the timing to create close phase tracking of the 2 drivers is so that:

> The direct sound of each frequency in the XO range arrives at the same time from each of the 2 drivers. This provides the maximum SPL support of the direct sound at the LP for each frequency in the XO range.
> A stable frontal lobe faced directly at the LP is achieved. That means the that the maximum SPL of the direct sound is centered on the LP for all the XO frequencies. This provides the most uniform sound field possible with the largest possible sweet spot at the LP. If the timing is not correct the frontal lobe points up and down at the frequencies above and below the XO point.
> The horizontal off-axis SPL response will be as smooth as possible. [It is also necessary to choose drivers that have similar dispersion patterns at the XO frequency in order to minimize the SPL irregularity in the off-axis sound pattern. Smooth horizontal off-axis SPL response is now considered a relatively important characteristic.]

The importance of this to sound quality varies by builders/reviewers of speakers. All agree that creating in-phase timing at the XO point is advantageous for SPL support and sound quality. The importance of close phase tracking away from the XO point is a preferred condition, but is not as important to many. Most all commercial speakers do not provide close phase tracking. Possibly most of the newer the newer active systems are doing this as they include a DSP XO. Some of the more expensive passive designs have included this feature for years.
 

·
Premium Member
Joined
·
1,400 Posts
Re: 100% NooB! ( Component ) Time Alignment

<<<<<SNIP LOTS>>>>So my question is: Do I just subtract 1.48 from 1.65 to arrive at .17ms as a delay for these drivers? The XO recommended for the system is 18db Bessel for both HP and LP so mine is at 800hz 18db Bessel.<<<<SNIP>>>>
Hi,

Since you have REW what matters most is that there's a smooth frequency hand-off between the LF and the HF.
- John has nicely summarized it all within the preceding post.
- You can dial in some delay ( on either component ) & then measure/remeasure until you get the smoothest hand-off ( as well as off-axis performance ).


I'll suggest that you repeat your 3 sweeps, but this time ( instead of a full range sweep ) limit each sweep to 1.5 octaves below to 1.5 octaves above your 800hz crossover point ( so sweep, 200hz to 2000hz ).

I'll be interested to see if the measured LF delay changes ( lessens ).
- BTW, which visual indicator ( within REW ) are you using to determine these 2 delay times?

Two ( Apparent ) Delay Paradoxes;

(i) There's such a thing as propagation delay ( large heavy objects take a bit more time to overcome inertia ).
- The woofer is slower ( but even that is not absolute & is frequency dependent ).

(ii) Also, every voice coil ( in every driver ) has a minute amount of delay attributable to its own inductive ( in mH ) value.
- The woofer has a much larger coil / so its delay ( measured in uSec ) is greater than the compression driver.


:)

PS; Posting your original ( plus any newly made ) .mdat files is always useful
 

·
Registered
Joined
·
2 Posts
Hi,
First, scuse me for my poor english :-/
You have a great jbl set !!
It is major to take time for the measure and the filtering with this kind of speaker if you want to pull of them the best! I have almost the same speakers but with only one 2226. After a lot of measure, i have found a good compromise with a Xo 650Hz : LP LR24 and HP LR12, drivers in the same polarity and 0.55ms delay for the 2226.
Another possibilty to analyse the phase at the Xo is the GD around the Xo range . It must be the more flat as possible .
The difficult with your set will be maybe the 2 woofers configuration with a different distance of each with the horn/compression ( phase shift at the Xo ) : more higher is the Xo more problematic it was
If it was the case : a 2.5 way is possible ( lower woofer : 20-150hz / higher : 20-650hz ( for example ) ... only the woofer near the horn do the cross )

Best regards
SpiOn
 

·
Registered
Joined
·
17 Posts
Discussion Starter · #7 ·
Hi everyone,

Thanks for the replies and help thus far! :)

Sorry for the delay, I have to get back to work this week so hopefully I can get some more measurements by this weekend.
 
1 - 9 of 9 Posts
Top