Going back to sealed from LLT

[thxgoon]

Well guys, as much fun as the LLT's have been, they are just too big in my new room. You have to turn sideways to fit between the subs and the couch to get into the room, literally. No good. So I thought I'd throw some ideas out and see what you all thought.

TC Sounds 15" TC1000's X2
EP 2500
BFD 1124

Originally, I'd noped to build a smaller vented enclosure similar to avaserfi's Infinity build, but with these drivers the box would have to be too big to get decent results. In a sealed design with both drivers mounted in 8 cu ft gives a Qtc of about .7. Power handling is right at 800 watts per driver before exceeding xmax (power compression not withstanding) so the design appears to be amp limited.

I took some REW sweeps of current locations vs new location and compared it with WinISD models. It appears that I should be able to get to within 1db of max ouput with a sealed design in a new location at the front of the room as compared to the LLT's at 20hz. This is thanks to the ability to co-locate the two drivers as well as better response from the new location.

Has anyone used an LT circuit? I'd like to reduce the box size to 6 cubes and this seems a good way to make up for lost performance. (max spl stays within 1 db of 8 cubes) Good idea? Bad idea? Is there any difference between an LT circuit and using external eq?

What do you guys think?

[Mike P.]

Have you tried modeling the 15" TC1000 in a sealed and also with a LT and compared them?

[thxgoon]

Yes. When I get home later I'll try to post some plots, but on the work puter that's not doable

Max SPL between 6 and 8 cubes is the same and this all I'm really considering given that any response irregularities can be tuned via eq.

There are 2 things I don't know:

1 - Whether LT circuit and external eq behave exactly the same
2 - How much power compression will become a factor in the design. Smaller box needing more power.
3 - How accurate WinISD is at predicting air spring related roll off. Oddly, response doesn't seem to be affected much until it gets below 1 cu ft per driver. And at infinitely large boxes, response drops to zero.

Okay, so that's 3

[Mike P.]

How can the SPL between the two be the same if a LT boosts the low end?

[thxgoon]

Mike P. wrote:

How can the SPL between the two be the same if a LT boosts the low end?

The thought is that I can apply the same boost with the BFD.

Max SPL is a function of xmax or power as I understand it. So eq aside, the max spl is dependent on excursion, power handling, and available power. More power is needed in a smaller box for the same output. Between 6 and 8 cubes (according to WinISD) there is no substantial difference.

[Mike P.]

The thought is that I can apply the same boost with the BFD.

Ok, gotcha.

[avaserfi]

thxgoon wrote:

1 - Whether LT circuit and external eq behave exactly the same

If given a choice between passive (LT) and active equalization (BFD) I would recommend active equalization virtually everytime. Ideally, you could just build a circuit and you would have your desired response, but because of room interaction this will never be the case at these frequencies.

The LT will already be using valuable headroom from the sub, couple this with the equalization required to get the desired response and you are potentially wasting headroom between the two. Now, if you have no LT, you can still achieve the same low end response while allowing for room interaction. Furthermore when you attempt to equalize you will have no issues with trying to defeat passive method electronically (wasting headroom). If you really want to try the LT out do it externally such that you can remove it at will and measure which method allows more headroom.

2 - How much power compression will become a factor in the design. Smaller box needing more power.

This will be largely driver dependant. You do have extremely high quality drivers at your disposal so I would expect less compression than typical.

3 - How accurate WinISD is at predicting air spring related roll off. Oddly, response doesn't seem to be affected much until it gets below 1 cu ft per driver. And at infinitely large boxes, response drops to zero.

Low frequency modeling is fairly accurate, but not 100% so. I have never seen or bothered trying to verify the full accuracy of air-spring roll off as modeled before so I cannot sufficiently comment on this statement.

If you post the T/S for the driver and your goals for size, output and reach I would be more than willing to model it for you with for maximizing these aspects while minimizing size. This method would allow for maximized performance above tuning when compared to a sealed model.

-Andrew

[thxgoon]

avaserfi wrote:

If given a choice between passive (LT) and active equalization (BFD) I would recommend active equalization virtually everytime.

Excellent! That's what I was thinking. WinISD must not take power into account for the filters when it calculates response and excursion for an input signal. It was telling me that only 350 watts were needed for a given output at 20hz with LT vs 1600 watts without. Some kind of voodoo magic...

Here are the T/S parameters:

Qes .381
Qms 3.820
Fs 15.6
Vas 393 l
Mms 238 g
Re 3.66
Xmax 27 mm

I'd like to maintain the performance I currently have if possible. I took room sweeps using REW of the 2 locations of the LLT's currently and the new location. I did this by placing one sub at the listening position and measuring all of the sub locations. The purple line is the new location and the yellow and green lines are the present locations of the subs.



From the looks of things, I gain 3 db down low from the location and should gain another 3 db by co-locating the subs. This should offset my losses of removing the ports. I'm not opposed to a vented design, as long as I don't lose 20hz extension.

Ideally I'd like to maintain over 110 db at 20hz from the sub before room gain and if possible keep as much sub 20hz output as possible. I would not be opposed to losing the sub-sonic stuff for lower distortion or higher output. The box needs to house both drivers, and be in the neighborhood of 24" by 28" by 30". At the moment I'm thinking 8 cubs sealed is the best bet but I'm open to any ideas.

Thanks for your help!

[Mike P.]

It sounds like active equalization together with REW will produce a sealed sub optimized for a specific location which should out perform a sealed sub with a LT applied and then being equalized for the best performance. Looking forward to seeing how this develops.

[avaserfi]

The size you gave is only about 7 cubic feet, but since you are thinking about 8 I went with that for extra performance .

Below is the model of a 2x drivers in 8 cubes sealed (yellow) and 2x drivers in 7 cubes with a 12" x 3" x 50" port (16Hz tune total port volume is about 1 cube) (gray) both are with 1400 watts input assuming 4 ohms nominal being powered by an EP2500, 1 sub per channel:


Max SPL:


Cone excursion:


Port airspeed:



As you can see the ported enclosure will have about 3-4dB more output. Both units easily clear your 110dB request before room gain (it is safe to presume these drivers will be linear even at high output/power due to their extreme design). The main reason to go ported is the huge headroom you will have if you ever decide to get some more power. It isn't even close to reaching its limits yet.

With the recommended port there will be no issues with compression or chuffing. Due to its length port resonance will occur at about 130Hz, this is not even a remote issue if a 4th order crossover is used at 80Hz.

I would go ported without question.

-Andrew