Group Delay, Phase, Electronic EQ?

[KennyG]

Hi all you brain twisted DIY'ers, I'm just sitting here with my brain twisted and frying, and figured I'd try and pass alittle of it off! I've been looking at group delay and phase shift.
Would it be safe to assume two of the reasons sealed subs have a cleaner sound are, group delay and phase characteristics?
Two of the sub designs I've modeled using the WinISD program that Thomas sent me show a large delay centered around the vented units port tuning, it's about a 30ms difference, which I would think creates a blurring effect.?
Also group delay at 20hz is 132 deg for sealed and 175 deg for the vented design, is the 43 deg difference noticable? and if so, how does this extra shifting of phase manifest itself?
Now the main question, does anyone know of an EQ'ing system that can take the natural roll off of a seal cabinet, and EQ it so it's flat to 20hz?
The sealed unit I've modeled has an F3 @ 35.4hz and drops to -10.5db @ 20hz.
I realize that for the amp to provide that much gain at 20hz, it will need to be about 8 times larger than what I would "normally" use, maybe more, so my thought is a used pro amp.
Any ideas on the EQ system???

[ThomasW]

Kenny,

aaaaaakkkkkk!

You're going to drive both of us crazy if you keep reading the "Cookbook", and try to create the "perfect" sub. As an abstract concept.

Sorry, I'm coming down with a cold, (and feel like crap), so here are quick and dirty answers to the "easy" questions.

1)if you're looking at roll-off from software plots, don't forget the concept of "room gain". This boosts the bottom end and it doesn't roll-off like the software model shows. So you don't even consider EQ, until the system is measured in the room where it's going to be used. If needed EQ should only be used in the "cut" mode to drop spikes. It can not fill nulls. The best types of EQ are parametric.

If you want to see the real effects of room gain, look at Bing's DIY subwoofer thread. Put his T/S numbers in WinISD, and do a plot. Then look at his measured in room response.

2)GD and the audible effects of phase shift are and will be debated until the end of time. The only way to answer this for yourself, is to build identical sealed and ported systems for your comparison. Yes, sealed systems "show" much lower GD than ported. But whether this is audible, is questionable for subs. Also I'd question the concept of audible "blurring" at the tuning frequency, we're taking 20Hz or lower here.

For subs, buy the biggest amp you can afford. Remember that most lower priced pro amps are fan cooled. So you need to live with the fan noise or be able to put the amp away from the listening area




theAudioWorx
Klone-Audio

[KennyG]

Thomas, you've created this monster. I never thought anything about the above mentioned items until you sent me that winISD program. 8)
After I received the program, I started looking at the phase plots and group delays of each sub design. The cookbook is not to blame for this one.
I have used parameteric EQ's in the past, my first seperates were old SAE components, their pre-amps had parameteric EQ's...plus their amps had great a great LED system that danced with the amp's output!
Anyway, aren't there some subs that are using electronic EQ'ing not just to remove freq humps caused by room interaction, but to also flatten out response where gain is needed?

[ThomasW]

aren't there some subs that are using electronic EQ'ing not just to remove freq humps caused by room interaction, but to also flatten out response where gain is needed?

I don't know. :? The "servo" subs use the servo to control cone excursion and limit distortion.

Something you might consider is the time honored tradition of making "prototype" boxes. These are quick and dirty test boxes to try various ideas.

Email Dan Wiggins (dwiggins@adireaudio.com) He has a method of creating low Qtc ported boxes. Tell him you want to compare a Qtc 0.5 sealed vs ported box. I bet he'll be glad to "run the numbers" and give you all the necessary data to create these. Low Qtc ported designs can't be "modeled" using of the software programs. Dan uses the filter formulas for the electrical circuit of a box and inserts the "Q". I don't know exactially what he's doing

Also GD and phase angles are most important for designing and building regular speakers, not subs




theAudioWorx
Klone-Audio

[Bing Fung]

Kenny, I asked the same questions about Group delay when I was first gathering info fo my project.

I think the key is that yes, GD can affect the sound but it is critical only at the higher frequencies above where your sub will be operating.

Sealed vs ported should be determined based on your goals. Unless you have alot of power, apperently the ported is the only way to reach the low end of the 20Hz spectrum with out active processing.

If you want to build a sealed, but still want the low extension, you may want to look at this . Personally I don't think it's the way you should go for I feel you should design the response into the box, rather than have to manipulate the signal to be flat.

But then, I really don't know much about this stuff either....




Bing

[ThomasW]

Guys

I've got a Marchand Bassis. Really great unit, but pretty specialized. Although Marchand states it's for "sealed" systems, actually it will work anywhere you want to boost the bass to compensate for the natural roll-off of the system. Mine we "modified" (actually by just doubled the value of 2 caps), to lower the Fs range to 1/2 that of the stock units.




theAudioWorx
Klone-Audio

[KennyG]

Bing, thanks alot for that site, that's exactly what I'm looking for! That's it!!!
I don't think it matters how you reach a flat response...or even IF you reach a flat response, all that really matters is that each person is satisfied with the end product. I don't think electronic EQ'ing is any worse than using the back wave of a driver, and most of the sealed cabinets I've heard really do have a tight, articulate sound that is different from vented units. Of course this EQ could screw that up, but with the right amp...well I'll never know unless I go for it, and that's the fun of this whole DIY thing, after researching, I've come up with what I think will be a good design...as long as I execute it right. Again thanks for that site, it's exactly what I'm taking about.
Gotta run and fix the electrical wiring in my favorite Mexican food restuarant.
Thomas, what was gained in the modification of your unit?

[Bing Fung]

That's it in a nutshell Kenny, your right, there is no right or wrong and it dosen't matter how you get there, only that you enjoy the journey!




Bing

[Patrick Sun]

But don't come crying to us when you hose it all up.




PatCave; HT Pix;Gear;DIY Projects;DVDs; LDs

[ThomasW]

Kenny

As I described in my above post, we doubled the value of the caps that control the Fs "range". The stock range on the pot is 65-130Hz. By doubling the value of the caps, the range is cut in 1/2, so 32.5-65Hz. A little more reasonable for the newer generation of subs.

But before getting one of these build the sub. Because with room interaction you may need something more flexible like a parametric.

We got the Bassis to EQ the open baffle "free-air" mounted woofers in the Whispers, and for another dipole woofer project Jon is working on. It's just a little cheaper than the $1200 processor sold by Legacy




theAudioWorx
Klone-Audio

[KennyG]

Yes, I will build the sub first, good point, I can't figure out where I'm going if I don't know where I am.
Thomas, I have decided on a starting cabinet that uses a Qtc of .72 with a Vb of 62.3 Liters.(seems very small?)
This is a Sealed Isobarik-push/pull design. Do you see these as viable design perimeters?

If this cabinet sounds good I will then add a second layer of 3/4 MDF glued and screwed so that all joints overlap. (plus internal bracing)

[ThomasW]

Kenny

Here's where I'd start my experiments if I were you, a design from "Dan the man" himself

>>>snip

3.25 cubic feet
Isobaric pair of Tempests
Tune to 14 Hz
Stuff with 48 ounces of polyfill

What will you get:

A box that has a second order roll-off to 12 Hz (sealed box rate)
A box that has a Qtc of 0.5 (critically damped; tightest bass you can get)
A box with an Fc of 20 Hz

The response is often considered the "ideal" audiophile response: critically damped response with a corner frequency of 20 Hz.

Why do the vented versus a pure sealed? More headroom. As compared to a sealed Tempest, you'll have between 3 and 7.2 dB MORE SPL capability with the vented alignment, from 22 Hz down to a bone-rumbling 11 Hz. That means you get the sealed sound, with the benefit of the vented SPL capability.

Additionally, because the box is SO small, it takes a FULL 750W linearly. And, in a typical room, you should have a final system F3 around 16 Hz. F8 would be down around 11 Hz. DEEP bass output!

Worth considering, IMHO.

Dan Wiggins
Adire Audio


>>snip




theAudioWorx
Klone-Audio

[KennyG]

O.K.

[ThomasW]

See above post for design details.

As I read his design he talks about the benefits of ported vs sealed. So I'm confused, he talks about "tuning to 14Hz? This maybe one of his "famous" low "Q" ported boxes with an Isobarik driver set-up. That's really cool! But he's over my head with the specifics of the tuning

Copy this info and email him, and ask if the design is sealed or ported? If it's ported I don't know how he calculates the port.

This quote is from a post he did here

http://www.hometheatertalk.com/ubb/Forum7/HTML/000175.html

Actually I think I figured it out. Ported box, Fb 14Hz, drivers mounted push/pull. This calculates a 2" port 12.12" long. This is to small a port, so I must be doing something wrong.

Anyway, email him he's great to deal with, and loves this stuff. It's better to get the info straight from the source, than have me try to second guess his design.

Sorry for rambling, I'm a little spacey from taking "cold meds"




theAudioWorx
Klone-Audio

[KennyG]

Ramble on my friend, with specs like these I suppose I'd better talk to him!

[ThomasW]

Ok Kenny

I emailed Dan and here's the data,

>>>>snip
Your 2" ID by 10"L vent is fine; however, I'd look at a 4" ID by 40"L vent. How to do this vent? Why, flexible piping, of course... 4" flexi pipe would work well here; just coil it up in the bottom of the cabinet, and you're set.
>>>snip

Now this seems like too weird for a port so I think that a PR would be in order. If you download the demo version of LspCAD from the Adire site, I can email you the Tempest isobaric data file Dan sent to. Then you can run the LspCAD demo, importing the data, and use the built-in PR modeling tool.

How's that sound?




theAudioWorx
Klone-Audio

[KennyG]

Thanks Thomas, I'll let you know when I get it downloaded.
Got an E-mail from Dan today myself. Talk to ya later Kenny

[KennyG]

GEEZ...well I got it downloaded, but I think I need something else to be able to read a zip file, I just don't know what it is. When I open the file all I get is program language...I think.
Computers drive me nuts, I can build'em, and if it's got a manual I can learn any program, but if I don't have a manual I'm lost...and I'm lost now

[ThomasW]

Yep

You need something like

http://www.winzip.com/ddchomea.htm

I looked at sending you the Adire LspCAD unzipped as an email attachment but it's really too big, 1.75 megs.




theAudioWorx
Klone-Audio

[KennyG]

Alrighty then, I'm ready to go, tonight I will learn how to use this program! Once again thanks Thomas, I'm sure I'll have questions later!!!
KennyG