High-Q driver + vented enclosure + active EQ... does it work well?

[Brian Steele]

Just wondering if anyone tried anything like the above (using a high-Qts driver in a small vented box, along with active EQ to flatten the passband). If you did, were you happy with the results?

Here's what I'm considering:

Driver parameters:
Vas: 2.05 cu.ft.
Qts: 0.610
Fs: 28.3 Hz

Box parameters:
Vb: 2.51 cu.ft.
Fb: 35 Hz

Active EQ:
-7dB @ 42 Hz, Q=0.7

[johnnyrichards]

Browsing the service manuals for several of the Infinity powered subwoofers I noticed a lot of them feature a "friend" circuit, which is in fact a notch filter like you are proposing. They use these on the Primus subs and some of the older ones as well, basically anything that uses the driver I used in my sealed bedroom sub.

Looking at the parameters of the CMMD 12" driver I used, it has a Q near 0.6, an Fs of 23 hz and a Vas around 3.5 cubic feet. Good for a sealed sub with just a little boost down low. Infinity went the route you are proposing, it will give higher output for a given enclosure size than the sealed alignment, but requires a more complex network to get it to work right.

I debated doing an alignment like that, but decided against it for two reasons:

1: I lack the electronics to generate the notch filter actively and
2: Doing it passively would require some expensive components

If you have the capacity to do this, I think it will result in a very nice subwoofer.

[Brian Steele]

Lucky for me, I should be getting my hands shortly on a cheap car audio parametric EQ that can create the type of notch filter I'm suggesting.

[johnnyrichards]

That would work great.

[billfitzmaurice]

Just wondering if anyone tried anything like the above (using a high-Qts driver in a small vented box, along with active EQ to flatten the passband).

High Qts drivers really dislike small boxes. To get what you're looking for you need to go in the opposite direction, with a low Qts (under .20) high BL (above 20) driver. It's comfortable in a small box, vented or sealed. You can use either a simple Linkwitz transform circuit in a sealed box or full DSP in a VB to flatten response. The downside is that you need a lot of Pe and Xmax, and amp power, all of which add to the cost. But if you've got more money than available space it's a viable option.

[Brian Steele]

High Qts drivers really dislike small boxes. To get what you're looking for you need to go in the opposite direction, with a low Qts (under .20) high BL (above 20) driver. It's comfortable in a small box, vented or sealed. You can use either a simple Linkwitz transform circuit in a sealed box or full DSP in a VB to flatten response. The downside is that you need a lot of Pe and Xmax, and amp power, all of which add to the cost. But if you've got more money than available space it's a viable option.

Thanks Bill. Usually I like to start with a low-Q driver, but in this case I actually have the driver in question in-hand, so I thought I'd experiment with it a little.

Its t/s parameters suggest that the unassisted alignment I'm considering would have a frequency response that has a peak of several dBs above Fb that can be removed by the notch filter. What intrigues me is that, not only does this result in a flat passband, amplification requirements and excursion are also reduced in the passband (compare with LR assisted alignments, where the exact opposite occurs).

[billfitzmaurice]

Thanks Bill. Usually I like to start with a low-Q driver, but in this case I actually have the driver in question in-hand, so I thought I'd experiment with it a little.

Doing what you propose will work, but the 2.5 cu ft box size isn't all that small, so for one who didn't already have the driver in hand I wouldn't recommend the added cost of a processor in lieu of a more suitable driver. The high Bl driver scenario is intended for a very small box, on the order of less than 1 cu ft with a twelve. There have been a few JAES articles in the past few years on the subject, the downside being the 1kW Pe, 15mm+ xmax and +1kW amp required to make it work well. But it can work.