subwoofer advice for very small room needed

[mlamp]

i recently bought the RSS315-HF when it went on sale at parts express, and need some box/tuning advice for using it in a very small room. the room is approx. 8.5' wide, 10' long, with 8' ceilings. (before it became my theater room, we called it 'the big closet') the floor is carpet over concrete slab, the walls drywall. the rest of the system consists of cjd's rs150mtm for the front left, right, and center, and the matching mt for the rears. i'm guessing on a crossover frequency around 80 hz to the sub. could be lower as this setup sounds very full in such a small space.

most of the designs i've seen for this driver are either sealed around 2 cu. ft., or ported around 4 cu. ft., and tuned to 16-18 hz. i could do either size, or even smaller if it made sense, but probably wouldn't want to go too much bigger than 4 cu. ft. My first thought was to do the sealed version, and let the room gain fill in the bottom end a bit, but i'm starting to wonder if even that might be overpowering. i realize that the 'right' answer would be to get a BFD, do measurements, room treatments, etc. to combat room gain and room modes, and i'd like to eventually get to that. That will probably be a while, however, and in the meantime i'd like to start with something that would at least be pleasant to listen to, and make the eq and room treatment tasks that come later easier, if possible.

i haven't bought an amp yet, but was leaning toward something external, maybe a pro sound amp.

[owdi]

2 cu ft sealed is right on the money for you, given the HT application. You'll have the same bottom end as a 4 cu ft ported box in a large room. Even 1.5 cu ft sealed would work.

If you were to build the 4 cu ft vented enclosure, you would have to EQ the bottom end down to get neutral response.

Dan

[rc white]

The smallest subwoofer with this driver is a Linkwitz transformed 1cubic foot box.

http://sound.westhost.com/project71.htm

In typical Northern hemisphere situations the room gain seems to be about half the theoretical amount, around 6db. per octave, if this is true of your room then an optimum system uses an additional 1st. order filter, i.e. a smaller input coupling capacitor to give the characteristic shown in this screen shot..



This is with 250Watt input, the cone excursion does not exceed 14mm. with this..



Given the above conditions this will be flat to 20Hz. from a 1 cubic foot box.
rcw

[brent_s]

Since HT is in the equation, I'd personally recommend 3 ft^3 net tuned to 20hz. It's -4dB at 20hz on its own. If you use something like the BASH-300, you'll have rumble filter at 18-20hz that'll pull that -4dB down a bit more. Should compliment the expected room gain nicely. Even if you end up wit a bit of positive gain below 30hz, it'll be more of a mild house curve that lots of folks seem to prefer.

"overpowering", to me, is a function of SPL, which you control with judicisous use of the trim and volume settings. ;-)

-Brent

[brent_s]

This is a 315-HF in 3.5 ft^3 net, slot ported to 19hz. Corner loaded in a 24x16x9 room...bonus room over garage so typical wood joist floor, stud walls, dry wall, etc. REQW, no smoothing, sub only, 12' measurement distance.

-Brent

[mlamp]

thanks everyone for the suggestions. brent, the room you took the measurements in is almost 5 times the volume of mine. are there any ways to figure/guess what that curve would look like in a different space?

[brent_s]

I vaguely recall that theoretical room gain should start at the frequency with a half wavelength equivalent to the longest dimension of the room. For you, that would be roughly 54-55hz, if I did the math correctly. Theoretically, you could pick up 12dB/octave below that point...typically it's more like 6-8dB.

Being wildly theoretical, that would give you something like +8dB @ 25hz added to the predicted -2dB from WinISD (assuming no rumble filter) for a net of +6dB @ 25Hz relative to the 55Hz level. Even with room gain, response should start to fall off below that as the ported alignment's response is dropping at a faster rate than room gain *should* be supplementing it.

Food for thought. The lowest note on a standard 4 string electric bass is 42Hz. Unless you listen to music with pipe organ, synthesized bass, or maybe orchestral pieces with really large kettle drums, the theoretical room gain won't really be touching the musical range. For movies that plumb the depths of the 20-40Hz octave, lots of folks would see any "free" bass boost or house curve as a good thing.

Think about it another way. If you build the 3 ft^3/20hz alignment and the bass is too fat for your taste, you could stuff the port and have a low Q sealed alignment. WinISD says Q would be .642 with an f3=37hz and an f10=20hz. My purely theoretical room gain numbers above would bring that back to flat down to 20hz.

As MDF is fairly cheap, you could go ahead and build both a sealed and vented alignment to see which one works better for you. If you're going to that trouble, I'd probably go with a 2ft^3 for the sealed alignment.

Another disclaimer. My 315HF measurement didn't include any sort of rumble filter. My sub amp is a full range NAD with a low end rolloff spec'd at 3Hz. The dip at 17hz is a room null do to measurement/listener location. It appears in every sub I've measured in that room, both sealed and ported with a couple of different sub locations.

-Brent

[rc white]

If you look at the frequency response plot I posted you will see that it has a first order roll off that begins in the 60Hz. region this exactly compliments what you will find in a typical Northern hemisphere room and will be flat to 20Hz. in such a room.
Generally you can make the subwoofer smallest by using the L-T type of scheme, but they need a more expensive higher powered power amplifier, and of course the L-T electronics.
Vented systems are cheaper because they have smaller power needs and can use in built rumble filters etc., but they are also bigger.
In a small room most people go for L-T since it is the least intrusive, but it is also more expensive, its your choice.
rcw