DiyMobileAudio.com Car Stereo Forum banner

1 - 20 of 55 Posts

·
Registered
Joined
·
1,559 Posts
Discussion Starter #1
When modeling different subs in WinISD, I see that group delay goes up dramatically for a ported alignment. Is there a method to improve on this aspect and get it closer to a sealed enclosure, and how much does this matter?
 

·
Registered
Joined
·
1,559 Posts
Discussion Starter #4
The reason I asked is because I have a new 10" Dayton Titanic MKIII laying around...I have a small sealed box of .8 ft³ (net), but I was thinking of trying it out ported as well.
 

·
Premium Member
Joined
·
5,992 Posts
It doesn't matter.
You're killing me Andy :p

Of course it matters, I can heard the difference between a ported sub, a front loaded horn, a tapped horn, and a sealed box. The difference is impossible to miss once you've listened to the various box types for a few years. I can hear the difference in literally five seconds.

As I see it, here are the differences:

1) Ported
Relatively easy to build, medium sized box, high power handling. To my ears, the worst sonically. The phase rotates 180 degrees, which means that the harmonics and the fundamental are not in sync. For instance, if you have a ported box tuned to 30hz and a 30hz note is played, the 2nd and 3rd harmonic are about 180 degrees out of phase.

2) Sealed
Easiest to build, very forgiving. Small box, high power handling. To my ears, the 3rd best sonically, but can be the best if you use a lot of them. Well behaved phase response. It's achilles heel is distortion; due to the relatively low efficiency the next two types will generally have lower distortion. IE, sealed boxes don't have high distortion, but they have low efficiency, which means that the next two types can play with lower distortion because they get louder with less power. But this can be addressed by using very large sealed boxes with multiple small drivers or one large driver. (IE, four eights, one fifteen, two twelves, etc.)

3) tapped horn
Easy to build. Larger box than vented. It's main advantage over ported is more output and the phase rotation is half as much, just 90 degrees. This can give you the best of both worlds, a lot of output for a single driver, while retaining a phase curve that rivals a sealed box.

4) front loaded horn.
Difficult to build. Largest box of all. It has the following advantage over the other types:
a) Very high output, which can exceed all other types, depending on box size. (IE, Hoffman's Iron Law dictates that output is determined by box size, and FLHs tend to have the largest box size. If box size is the same for all four, output will be about the same.)
b) Excellent phase response. A FLH has a phase curve that is nearly as good as sealed, better than ported, and similar to a TH
c) The "icing on the cake" is that a FLH dramatically reduces distortion. This is what sets it apart from an array of sealed boxes; the chamber in front of the woofer drops distortion a significant amount. There is no way to replicate this electronically, a coupling chamber and horn are a physical filter that reduces distortion.

 

·
Registered
Joined
·
4,941 Posts


1) Ported
Relatively easy to build, medium sized box, high power handling. To my ears, the worst sonically. The phase rotates 180 degrees, which means that the harmonics and the fundamental are not in sync. For instance, if you have a ported box tuned to 30hz and a 30hz note is played, the 2nd and 3rd harmonic are about 180 degrees out of phase.

But around the tuning frequency the non-linear distortion is generally very very low making this a small issue. If Fb lies around 25-35Hz the phase response should be very similar to that of a sealed box a half octave above that. The main issue with horns is no low frequency output without building huuge enclosures.
 

·
Registered
Joined
·
4,941 Posts
Here's a simplistic model;





Chapter 5

The results from this thesis research imply that phase distortion is of secondary importance as compared to frequency response irregularities, which is in agreement with previous research results.
How a typical group delay measurement looks;

Case Study: Sub Alignment using REW v5 - Home Theater Forum and Systems - HomeTheaterShack.com



The issue isn't absolute group delay, it's the non-flat delay over a given frequency range as the fundamentals and harmonics are delayed towards each other as you said. Any lowpass function as we always use in cars will increase GD as well. The non minimal phase areas will screw up GD too... I don't think GD is that audible unless it jumps really high, 1-1,5 cycles or so. The important thing to get right is the frequency response first off, imo.
 

·
Banned
Joined
·
4,394 Posts
GRAPHS!

Hanatsu, killing it...

group delay of 20 milliseconds, in a wave 30 feet long, or 1/30th of a second, doesn't account to much smear.

it's the change in linear response as the musical information travels through the beat, that does make a difference, we're responding to the sound of the speaker handing off the wave to the vent, at different times.
 

·
Premium Member
Joined
·
5,992 Posts
In this thread I demonstrate that ported boxes are just beyond the limit of audible group delay:

http://www.diymobileaudio.com/forum/2052295-post105.html

I will concede the following:

1) Sealed works quite well
2) Ported works well, if you don't mind screwed up transients
3) Ported can work nearly as well as sealed boxes and horns, but only if the tuning is very low. Basically the group delay gets worse as you get closer to the tuning frequency. So one way to offset the crummy group delay of ported boxes is to simply tune them to a frequency that's lower than the musical content. For instance, 25hz.

But a ported box tuned to 30 or 40hz? Oh hell no. We've all heard these boxes; it's how most people built their ported boxes.

Horns still have an efficiency advantage if you can't afford a lot of drivers, and a FLH can easily have the lowest distortion.

 

·
Registered
Joined
·
11,959 Posts
It doesn't matter.
x100! you cant hear it unless it gets to obscene levels, much larger than you will ever see.

If you want to play with it, port area is what will control GD most.
 

·
Registered
Joined
·
4,941 Posts
In this thread I demonstrate that ported boxes are just beyond the limit of audible group delay:

http://www.diymobileaudio.com/forum/2052295-post105.html

I will concede the following:

1) Sealed works quite well
2) Ported works well, if you don't mind screwed up transients
3) Ported can work nearly as well as sealed boxes and horns, but only if the tuning is very low. Basically the group delay gets worse as you get closer to the tuning frequency. So one way to offset the crummy group delay of ported boxes is to simply tune them to a frequency that's lower than the musical content. For instance, 25hz.

But a ported box tuned to 30 or 40hz? Oh hell no. We've all heard these boxes; it's how most people built their ported boxes.

Horns still have an efficiency advantage if you can't afford a lot of drivers, and a FLH can easily have the lowest distortion.

But still. There's no low frequency output from horns. I want one that can do 30Hz with authority ;)

From your standpoint and mine a 4th order BP would be the best compromise. The graph in that post looks like it's nearfield outside the car. I argue that it's the space outside the enclosure that dominate the FR and phase response. It won't look that good as soon it's in the car.
I believe that the main reason some alignments sound better than others simply can be described by non-linear distortion and SPL vs Freq (FR). I had major success eliminating the ringing modes by delayed active "anti-phase" subs. Removing bass note lingering for 500-1500ms does so much more than reducing GD imo.

Tapaaatalk!!
 

·
Registered
Joined
·
4,941 Posts
I think vented generally sound best among the "normal" enclosure types. Jugding from some of the other threads you made, it seem you like run subs much higher in frequency than I do. I tend to cross mine in the 50-60Hz area normally. Horns might generally sound better 60Hz+, idk. Not much experience with that.

Tapaaatalk!!
 

·
Registered
Joined
·
1,842 Posts
I had major success eliminating the ringing modes by delayed active "anti-phase" subs. Removing bass note lingering for 500-1500ms does so much more than reducing GD imo.

Tapaaatalk!!
That's pretty interesting. Is that for all frequencies that the sub is reproducing?




Sent from my SAMSUNG-SM-N900A using Tapatalk
 

·
Registered
Joined
·
4,941 Posts
That's pretty interesting. Is that for all frequencies that the sub is reproducing?

Sent from my SAMSUNG-SM-N900A using Tapatalk
Only required at modal peaks so the lingering frequencies are even over the entire frequency range the main sub is reproducing. I only used mine at the first peak at ~45Hz, I haven't experimented much with this yet. It's one of my upcoming experiments in my project thread...
 

·
Registered
Joined
·
4,941 Posts
Ok. Did some measurements. I can conclude that the phase discussion is indeed silly. Look at this:

An "anechoic" measurement of a sealed enclosure-subwoofer.



Same enclosure inside the car;



Same driver in a vented box with Fb 33Hz (inside car with EQ);



*The vented box only have a 100deg acoustic phase shift in its limited passband. Not 180deg.

*Compare the two first measurements, see how much the phase changes and how dependent it is on the frequency response. The EQed vented box looks more behaved.
 

·
Premium Member
Joined
·
4,235 Posts
Sorry. My answer was too short.

As Patrick has suggested, tuning a vented box low is a really good idea. Avoiding a big PEAK in the group delay is good. (See the red curve in one of the graphs above).

As far as fundamentals and harmonics being in phase, I can't say this is something I've ever paid attention to. I've always to tried to minimize distortion in subwoofers I've had a hand in designing. If it's harmonics of the fundamental that are included in the recording, then those are most likely going to be reproduced by another speaker anyway.

I've never built a TL, tapped horn or front loaded horn for use in a car.
 
1 - 20 of 55 Posts
Top