Driver specs vs application questions (1 Viewer)

[Kimmo Jaskari]

After seeing the choices available in subs and the costs, I started thinking about building my own. I pretty much have figured out what I'm going to build:

• a sealed sub
• 12 or 15 inch driver (probably 15... I want it to dig deep!)
• accurate bass rather than "boomy", ie better for music than HT
• fairly high power plate amp (300 to 500 W is what I'm aiming for)
• classic, square construction from MDF rather than sonotube

To simplify (probably a lot) that would mean I have to build the enclosure, buy a plate amp and the proper driver for that, any crossovers etc are already in the amp... right?
My problem right now is that I'm not sure what I am looking for when it comes to the driver. What characteristics should a good driver for a closed box like this have? I'm still at the point where I'm not 100% up to speed what all the specs on a data sheet for a driver means, but I'm studying everything I can get my hands on.
------------------
/Kimmo

"Don't tell me; we're about to go over a huge waterfall?"
"Yep."
"Sharp rocks at the bottom?"
"Most likely."
"Bring it on."

 

[Greg Monfort]

If you're following the other Sealed thread, then you should be getting some idea where you need to be WRT F3/Qtc, which helps with driver selection.
BTW, build it out of no-void 3/4" plywood such as Baltic Birch, Appleply, or marine grade, if best performance is desired in a ~acceptable weight cab, and a multiple driver IB isn't an option.
GM
------------------
Loud is beautiful, if it's clean

 

[Kimmo Jaskari]

Well, I'm reading what I can find that makes sense, but this is mostly greek to me (which isn't easy since I'm from Finland

If I understand correctly, for a sealed box, I want a driver with long Xmax, but that is still about all I have managed to figure out with any certainty.
What, in so many words, are the parameters to look for, that determines that a driver is better for a vented box or a sealed one?
If possible, please use short words in the explanation, since this stuff has me believing my IQ is in the double digits. Low double digits.

------------------
/Kimmo

"Don't tell me; we're about to go over a huge waterfall?"
"Yep."
"Sharp rocks at the bottom?"
"Most likely."
"Bring it on."

 

[Greg Monfort]

>Well, I'm reading what I can find that makes sense, but this is mostly greek to me (which isn't easy since I'm from Finland
If I understand correctly, for a sealed box, I want a driver with long Xmax, but that is still about all I have managed to figure out with any certainty.
=====
Correct, excursion increases 4x/octave, so as we Americans like to say: there's no replacement for displacement.

The trade-off is that the greater the displacement, the less useable BW it has due to amplitude induced distortion. Not that big a deal for a HT LFE channel IMO, but most definitely is for music.
=====
>What, in so many words, are the parameters to look for, that determines that a driver is better for a vented box or a sealed one?
>If possible, please use short words in the explanation, since this stuff has me believing my IQ is in the double digits. Low double digits.
=====
The simplistic answer is: EBP

 

[Kimmo Jaskari]

Thanks. That does help. (Slightly)
Tried some calculations in WinISD on a driver I found that looks interesting but either WinISD is insane or I am, cause it says I need a 12.1 liter box for a 12 inch driver to get 0.707 Qtc!
I mean... I can't even fit that driver onto the front panel of a sub that is 12 liters!
Oh well... I'm still missing something no doubt, have to keep trying.
------------------
/Kimmo

"Don't tell me; we're about to go over a huge waterfall?"
"Yep."
"Sharp rocks at the bottom?"
"Most likely."
"Bring it on."

 

[Greg Monfort]

That's a reasonable volume for a HE driver when only driven with

 

[Kimmo Jaskari]

Well, no other way to put it.. you lost me completely there.
The program asks for the data on the driver, and I enter the driver data... then I'm supposed to alter the data "by feel" to compensate for something or other?
And divide what by 1.2, the enclosure size I wind up with? That will make it even smaller, it's already too small for the driver to fit onto the front panel!

How do you calculate the volume of the driver itself - buy two and toss one of them into a brimming full bathtub and weigh the water that comes out?
As for how the room will interact with the sub... why is that a factor in building one? I mean, SVS can't come to my home and measure my room to build me a sub, yet they apparently manage to build pretty darn excellent ones anyhow. Is there some mathematical formula you use for this or is it more voodoo?
I didn't use to think I was retarded, but perhaps I was wrong, cause I'm definitely not getting this.

------------------
/Kimmo

"Don't tell me; we're about to go over a huge waterfall?"
"Yep."
"Sharp rocks at the bottom?"
"Most likely."
"Bring it on."

 

[Greg Monfort]

>Well, no other way to put it.. you lost me completely there.
The program asks for the data on the driver, and I enter the driver data... then I'm supposed to alter the data "by feel" to compensate for something or other?
====
Correct, if you want a design closer to real world use. Like I said, basic programs assume a max power input of And divide what by 1.2, the enclosure size I wind up with? That will make it even smaller, it's already too small for the driver to fit onto the front panel!
====
The net box volume. Yes, 100% stuffing makes it acoustically larger due to adiabatic to isothermal conversion of the trapped air. This allows you to make the box ~20% smaller for a given Qtc.
Like I said, even if 12L was the needed volume (which it's not, per my other comments), the box will be bigger than the driver.
====
>How do you calculate the volume of the driver itself - buy two and toss one of them into a brimming full bathtub and weigh the water that comes out?
====
Actually, folks have stuck them in plastic bags and placed them in buckets of water to measure the amount of water displaced. The better manufacturers list the info now though.
====
>As for how the room will interact with the sub... why is that a factor in building one? I mean, SVS can't come to my home and measure my room to build me a sub, yet they apparently manage to build pretty darn excellent ones anyhow.
====
I imagine they do, but it has nothing to do with in-room response. I assume they use an EBS alignment, since it blends reasonably well in the typical corner location of most tube subs. I bet they need quite a bit of EQ for best in-room response though. It's not mandatory, but if you're going to DIY, designing for your room will minimize the amount of EQ needed.
====
>Is there some mathematical formula you use for this or is it more voodoo?
====
No Voodoo, just mass quantities of applied physics. Or you could just build any old ***, turn it ***, get *****/******, and think it was the best ****-*** sound ever. (edited to meet PG rating)

====
>I didn't use to think I was retarded, but perhaps I was wrong, cause I'm definitely not getting this.
====
The retarded folks I've known didn't realize they were (unless repeatedly told so), so I guess you're just smart enough to know how ignorant (as opposed to dumb) you are, like me.

Anyway, AFAIK, very few folks seem to 'get it' quickly, I know I sure didn't. Bob's site lists some good books on the various subjects, so curl up with a few if you're serious about this stuff. If not, copy somebody else's design and just enjoy.

GM
------------------
Loud is beautiful, if it's clean

 

[Brian Bunge]

Greg,
Could you define "100% stuffing" for me? Some sort of example such as for internal volume X, use Y stuffing. I know I've heard to use approx. 1lb. stuffing per cubic foot as a general rule of thumb.
Thanks,
Brian
------------------

 

[Greg Monfort]

1.5lbs/ft^3 of cab volume for R19 fiberglass, and 1.0lb/ft^3 for polyester fiberfill (polyfil), so a 3ft^3 net cab needs either 4.5lbs of fiberglass, or 3.0lbs of polyfil.
Note that fiberglass does a much better job of attenuating mid/HF content if it's present, so is preferred in wide BW designs.
GM
------------------
Loud is beautiful, if it's clean

 

[Kimmo Jaskari]

Ok, after more reading and manual calculation I have realized that neither WinISD nor myself (I hope) is insane. The driver I had in mind actually has a 12.1 liter volume requirement (using only the data from the speaker manufacturer). Whether or not the driver in question is a good one to use for a sealed sub is another matter entirely at this point, I just want to figure out how to do these calculations correctly.
Increasing Qes seems to have no effect whatsoever on the enclosure volume, and according to the equations I found at this page: http://www.emunix.emich.edu/~zirk/audio/audioboxtype.html

"Don't tell me; we're about to go over a huge waterfall?"
"Yep."
"Sharp rocks at the bottom?"
"Most likely."
"Bring it on."

 

[Greg Monfort]

>Increasing Qes seems to have no effect whatsoever on the enclosure volume, and according to the equations I found at this page: http://www.emunix.emich.edu/~zirk/au...ioboxtype.html
... Qes never enters the size equation??
====
If changing Qes doesn't have any effect, then there's either a program problem, or you have to manually change Qts too. Either way, the program is worthless IMO.
Qts is a combination of Qes and Qms, so without it no calculations can be made. The formulas on that page are the much condensed versions utilizing only Qts. You would have to recalculate Qts using the revised Qes to make it work.
====
>But ok, with WinISD I can get a recommended size of the enclosure. How do I then go about adapting that to a real-world enclosure size, if the 12.1 liters isn't it?
====
Since WINISD doesn't seem to recalc Qts when Qes/Qms is changed, then if you can do spreadsheets, download Brian's from his site I previously listed.
If not, then you can do the calcs using the formulas here: http://www.hi-fi.com/speaker/boxes/ or buy a decent program.
GM
------------------
Loud is beautiful, if it's clean

 

[Kimmo Jaskari]

Sorry, guess that's what I get for a; trying to do maths at 2 am with a fever and b; not paying attention to what the software was or wasn't doing.

As you already realized, the software only accepts the data as you feed it to it, it doesn't check that Qts = (Qes*Qms)/(Qes+Qms) so obviously nothing happened when I altered Qes only.
------------------
/Kimmo

"Don't tell me; we're about to go over a huge waterfall?"
"Yep."
"Sharp rocks at the bottom?"
"Most likely."
"Bring it on."