sealed cabinet design: Qtc and layout questions

Greetings, as this is my first post in this community.

I'm designing a subwoofer for music and some home theater usage. I miss the twin 18" Bag End subs I got to play with when I was a theater techie in high school, so I've decided to build a large-ish sub for indoor use. My past endeavors have all been for automotive sound systems and nothing really amazing. Mainly, I have been using the prescribed values for sealed airspaces that come with the drivers.

I like the way sealed enclosures sound over the vented subs that I have heard (I can't say I've listened to any very high quality vented enclosures, so I'm probably biased). I've decided that I'll be building a sealed enclosure and that I'll be somewhat emulating a kit that Parts Express sells, but building the cabinet myself partially for fun and partially to save on costs.

Since this is my first stab at designing an enclosure using the thiele-small parameters for the driver, I would feel a lot more confident about things if someone would confirm that I'm going about it correctly. I found an Excel spreadsheet calculator that will do a bunch of useful calculations for sealed enclosures and punched in the values for my driver. Click below for a screenshot of the resulting values and response graphs:

Calculations

Frequency response derived by above calculations

Power response derived by above calculations


Here are the important parts I'll be using:
15" Dayton Titanic mkIII driver Reviews I have found seem positive and it looks like a well-built unit. Any comments or advice on my choice are welcome, naturally. I have already ordered it and it;s on the way, so changing my mind for a different driver is not really an option.

1000w plate amplifier This is the amp used in the kit, which has been reviewed favorably. I figured it would be wise to match the driver with the amp to minimize the chances of unpredicatable results. Same clause as for the driver aplies here.

For the cabinet design, I am planning on something like this. The design shown below is set up for the 3.1 cubic foot result of the calculations above. When I change the desired QTC, the volume naturally changes quite a bit. I plan to extend or retract the design of the cabinet's upper area vertically to achieve the final desired volume.

Cabinet design (not finalized)

The amplifier is probably intended to be stuck right into the airspace, but I would rather have the airspace be sealed tightly with walls of MDF, so I will be building a separate enclosure for the amplifier. I will install the amp with the back tray thing removed for ventilation through the convective ventilation slots I am planning to incorporate. The amplifier claims it doesn't need any cooling and has no fins, but digital electronics can always run cooler, so...


My main questions are thusly:

1) Does it look like I'm doing this correctly? This is my first experiment using calculations based on thiele-small parameters, and I guess I'd like a confirmation that I'm not blatantly doing something wrong.

2) Does the cabinet design I am working on look reasonable? Should I change the layout of the components at all for a better theoretical design? The placement of the hump for the amplifier enclosure is the main variable. right now it is in the lower right rear corner of the enclosure and I figure it can only help reduce standing wave problems by making the interior shape more irregular.

3) Is putting the driver against the floor as opposed to near the top of the cabinet a good idea or a bad idea with regards to sound?

4) Should I use a larger or smaller desired Qtc with a setup like this? I ran the calculations for .71, but I would like advice on what the effects and desirability of moving this up or down are. I have plenty of room, so within reason, cabinet size isn't a factor. This will be used mainly in a roomy living room featuring a half-height wall to a large dinette and kitchen area. Is a higher or lower Qtc a better fit for such a space?

5) Does a sealed enclosure look like the best choice for me based on what I've gone over above? I know that vented can make more noise with less power, but I have intentionally gone a bit overkill on the power of the system with the intention of making a less-efficient but better sounding (less boomy, more responsive) subwoofer. Correct me if this is not the way things work between vented and sealed.

Thanks to you all in advance for getting me moving on this project.
 
Well, Certainly that's a reasonable application for the unit. I might be inclined to go with a somewhat larger box, as I like low-q designs and the transient response that goes with them, but that's just me. If you want the bag end kinda depth, you might want to incorporate a bass boost, easily enough to do with a line-level opamp circuit, but hm... yours will have pretty darned good in-room response... I say go for it! Build away! You can always modify it later if you feel you need more ultra-deep grunt or some such. It should be pretty whompin' as is.
 

Ron E

Member
2002-06-27 10:41 pm
USA, MN
kstoerz said:
Greetings, as this is my first post in this community.

My main questions are thusly:

1) Does it look like I'm doing this correctly? This is my first experiment using calculations based on thiele-small parameters, and I guess I'd like a confirmation that I'm not blatantly doing something wrong.

2) Does the cabinet design I am working on look reasonable? Should I change the layout of the components at all for a better theoretical design? The placement of the hump for the amplifier enclosure is the main variable. right now it is in the lower right rear corner of the enclosure and I figure it can only help reduce standing wave problems by making the interior shape more irregular.

3) Is putting the driver against the floor as opposed to near the top of the cabinet a good idea or a bad idea with regards to sound?

4) Should I use a larger or smaller desired Qtc with a setup like this? I ran the calculations for .71, but I would like advice on what the effects and desirability of moving this up or down are.

5) Does a sealed enclosure look like the best choice for me based on what I've gone over above?

1) your design is fine.

2) it looks OK, you will probably want to consider some bracing though, and WRT standing waves, you won't really have any so placement of the amp doesn't matter except for how it affects how you place bracing.

3) nearer the floor is usually best - you get a 3dB boost from each boundary you place it near to. - placed in a corner you get a theoretical 9dB boost.

4) Qtc is fine. There are people who think that lower Qtc is tighter, based mostly upon theory that Qtc=0.5 is "critically damped". John Kreskovsky has a paper on his site that confirms that higher Qtc woofers track signals better than lower Qtc ones. IMO, Qtc is not an appreciable factor in sound as long as it is kept in the range of around 0.6 to 0.9. Also IMO, Qtc of 0.5 gives you an unnecessarily large box - it may give more impact in the room, but it is not sonic impact, it is visual impact.

5) Try it - make sure you take it easy on the volume control till you know your limits, since you could probably exceed excursion by a fair amount below ~35Hz or so with the amount of power you have available. It is nice to have the headroom though.

Good luck - from your twin city..
 
This is true. I suppose I could always build another enclosure or modify the existing one if I decide that it's not quite what I want.

If you recommend larger airspaces for transient response, how large is too large? Should I aim for a Qtc of something like .6? .6 would require five and change cubic feet of airspace. I redesigned for about 5.25 cubic feet (which doesn't take into account any internal bracing or the displacement of the driver) and the new cabinet looks more like this:

Cabinet revised to make 5 cubic feet

Would it be better to move the amplifier's partition up along the back of the cabinet to get it away from the rear of the driver or leave it where it is to create a more irregular airspace for less standing waves and whatnot?

Should I try to stuff this cabinet full of something, or should I just line the surfaces with some eggcrate material?
 
The more I read here and there, the more it sounds like I may not be so happy with Qts down around the .6 realm. It seems that .707 is the "magical happy spot" for sealed enclosures, so I'm tempted to back down closer to that number, possibly compromising around .65 or .68 to push a bit firther away from the boomy side of things.

Ron E, can you elaborate a bit on the excursion issue? I understand the physics of it WRT the driver hammering against the stops, but what exactly dictates where this can happen? Is it tied strongly to the resonant frequency of the system?

Speaking of resonant frequencies, I am wondering what happens when a drivers' resonant frequency is right in the middle of the range you are trying to reproduce with it. Wouldn't this create a spike in the frequency response around that point that might need to be corrected for?

Is this just a subwoofer problem, or do most drivers (for other ranges) have a resonant frequency within the desired range of reproduction?

how much of a problem might this be and what could I do to prevent having to deal with it? Would an equalizer on the signal help, or is it just the nature of the way the driver ends up moving to create sound at these certain frequency areas when the gain is cranked up?
 

Ron E

Member
2002-06-27 10:41 pm
USA, MN
kstoerz said:
Ron E, can you elaborate a bit on the excursion issue? I understand the physics of it WRT the driver hammering against the stops, but what exactly dictates where this can happen? Is it tied strongly to the resonant frequency of the system?

Speaking of resonant frequencies, I am wondering what happens when a drivers' resonant frequency is right in the middle of the range you are trying to reproduce with it. Wouldn't this create a spike in the frequency response around that point that might need to be corrected for?

Well, it is tied to the resonant frequency only in the sense that a woofer is mass controlled above resonance and stiffness controled below resonance - the stiffness of the box+suspension combination determines teh amount of power that can be applied below resonance before the cone reaches its excursion limits. Making a smaller box means a tighter spring and more powerhandling at low frequencies. If you really want to pound on this woofer, making a smaller box might be a good idea, but I think the Q=0.7 box will probably work fine. THe difference between Q=0.65 and Q=0.75 is not going to be audible in any way, shape or form. Sealed box design is not that futzy.

The resonant frequency impedance spike would only matter if you had a very high output impedance amp, which yours is not. Frequency response will be as predicted, except for the effects of driver inductance, which Brian's spreadsheet does not calculate, but which should be minor anyway. You have a built in parametric EQ, so you have nothing to worry about.

As far as damping material, it is not going to be necessary but you could put an inch or two of foam on the walls if it makes you feel better. With the driver and power you have, I'd worry more about solid box construction or you will have a box rattle problem.
 
kstoerz said:


Speaking of resonant frequencies, I am wondering what happens when a drivers' resonant frequency is right in the middle of the range you are trying to reproduce with it. Wouldn't this create a spike in the frequency response around that point that might need to be corrected for?


It creates a spike in the impedance of the driver, and can also result in a hump in the frequency responce depending on enclosure size and other factors, fill, bracing ect.

Fill in sub enclosures tends to be a contentious issue, I don't know why, it's effect is easily measureable (and audiable)

The role fill plays in a sealed sub enclosure; (assuming totally filled)

*Damping effect on the osciliation of the driver, primarially at resonance.

*reduction of the impedance rise at resonance (generally by around half depending on fill used)

*reduction in amplitude of the hump or knee at resonance (varies with enclosure size but typical effect on a "hump" is a 3db reduction, flattening effect)

*some absorption of higher frequency noise/distortion caused by the driver exciting the walls of the enclosure (helped more by bracing)

*subjective improvements in overall SQ, if you prefer sealed over vented, then you will most likley prefer filled over non-filled.
 
Volenti said:

It creates a spike in the impedance of the driver, and can also result in a hump in the frequency responce depending on enclosure size and other factors, fill, bracing ect.

Fill in sub enclosures tends to be a contentious issue, I don't know why, it's effect is easily measureable (and audiable)

The role fill plays in a sealed sub enclosure; (assuming totally filled)

*Damping effect on the osciliation of the driver, primarially at resonance.

*reduction of the impedance rise at resonance (generally by around half depending on fill used)

*reduction in amplitude of the hump or knee at resonance (varies with enclosure size but typical effect on a "hump" is a 3db reduction, flattening effect)

*some absorption of higher frequency noise/distortion caused by the driver exciting the walls of the enclosure (helped more by bracing)

*subjective improvements in overall SQ, if you prefer sealed over vented, then you will most likley prefer filled over non-filled.

Interesting. Sounds like stuffed might be the way to go. I imagine that I don't want to stuff it too tightly, but rather fill the brunt of the airspace with a somewhat loose but substantial packing.

What can I use? I've done similar things with pink fiberglass home insulation in the past, but I can't imagine that's terribly healthy for the driver in the long run. Would a few bags of synthetic pillow stuffing from Wally World be an acceptable substitute?

What will this do to my effective airspace? Doesn't stuffing make the driver "see" three times or so the airspace? If this is so, should I make my cabinet one third the size, or should I use the airspace I calculated for Qts .68 of 3.25 cubic feet and just stuff it anyways?

What about partially stuffing the cabinet?

Jeez; I'm just loaded with questions. Thanks for helping a noob out .:)
 
kstoerz said:


Interesting. Sounds like stuffed might be the way to go. I imagine that I don't want to stuff it too tightly, but rather fill the brunt of the airspace with a somewhat loose but substantial packing.

Exactly, extreme over filling can reduce the effective volume of the enclosure.


What can I use? I've done similar things with pink fiberglass home insulation in the past, but I can't imagine that's terribly healthy for the driver in the long run. Would a few bags of synthetic pillow stuffing from Wally World be an acceptable substitute?

Yea polyester pillow stuffing is what most people use (me included) fibreglass insulation should be better acoustically (finer fibres) but I havn't done any testing to compare.

What will this do to my effective airspace? Doesn't stuffing make the driver "see" three times or so the airspace? If this is so, should I make my cabinet one third the size, or should I use the airspace I calculated for Qts .68 of 3.25 cubic feet and just stuff it anyways?

This is where the controversy comes in:D, from my measurements the resonant frequency of a sub in a filled enclosure drops by around 2-4hz, generally considered too small a difference to heard.

The driver doesn't "see" a larger airspace, the fill affects the damping of the cone around resonance, lowering the QTS.

Think of fill more as a way to make a driver behave nicely in a smaller than "optimal" enclosure rather than making the driver "see" a larger enclosure volume. (you get the smoother/flatter frequency response of a larger enclosure with fill, but not the extension and efficiency of a larger enclosure (you get nothing for free))


What about partially stuffing the cabinet?

Partial fill simply gives a partial effect.

Jeez; I'm just loaded with questions. Thanks for helping a noob out .:)

no problem :angel:
 
The sealed layout looks good to me- a box (147l) with a Q=.6 will be essentially flat down to below 10hz, when room gain is added.

As far as woofer placement, you have chosen a good location. The closer to the floor, the better the coupling (and more bass output). Check this link:

http://www.nutshellhifi.com/ME2txt.html

The only problem with producing bass at that low of a frequency (like the Bag End ELF) is the built-in rumble (subsonic) filter on the plate amp. It starts to roll off at ~20hz. The plus side is that you'll have no worries about exceeding xmax!
note: all of the plate amps that PE offers has this feature.

-Hans
 
Hey Volenti; you don't hang out on ProCooling by any chance, do you?

I'm feeling pretty confident about this project, and my driver and amplifier will be arriving on Tuesday. I made a run to the Home Desperate last night and had them lop up a sheet of MDF into tablesawable pieces for me. The guy there only smashed one corner on the ground, so they're doing pretty good from what I've seen in the past.

I decided on a final design centered around a QTC of .68 and an airspace of about 3.5 cubic feet. The box is more or less assembled now (I was up until sunrise this morning and then slept until 2 or so) and now I need to disassemble it, do all the joints up with liquid nails, and screw it back together again. I'm going all out with the bracing, too. I like the designs I've seen here and there incorporating entire dividing shelfes that have been cut apart like a 4-light window to rigidify the middles of all four sides while still allowing plenty of room for air to move. I made the roof of the amplifier enclosure into one of these with an oddly shaped piece to better center the support on the weaker areas of the sides and another one is going above this piece vertically to secure the roof of the enclosure. Down behind the woofer will be another piece of similar function but designed to fit in around the rear of the magnet and basket.

I don't know exactly how deep the driver is from the mounting surface to the highest area of the surround, so I am not securing the driver mounting piece yet. It can slide in and out to fit where it is needed in the end.

My grille is going to be a piece of expanded steel carefully welded into a small frame.
 
Here are some pictures of the cabinet under construction. I'm waiting to mount the driver panel until I can measure some dimensions. More bracing is to follow shortly.

09t.jpg


01t.jpg


http://www.basementfreaks.com/members/karl/titanic/cabinet_building/