Port tuning below resonance

[dsrviola]

I read on somebody's DIY site about the advantages of (I hope I remember this correctly) tuning the port below resonance to get a response that was more along the lines of a sealed box than a ported alignment. It showed modeling that demonstrated transient response that was almost as good as a sealed box. I believe the roll of was closer to 12 db than 24db and it sacrificed a little in the way of extension compared to an "optimized" ported alignment, but gave better results in terms of transient response.

Does any of this sound familiar? I'm having a hard time locating the site, and I'm interested in investigating this further.

Thanks!

[BHTX]

A -3 dB (or lower) EBS alignment will do this. Basically, peak group delay is actually increased by a massive amount. However, it's pushed down in frequency so that it isn't in the 40+ Hz range where most low frequencies are located, and so that it's not as noticeable, since we're not as sensitive to distortions at such low frequencies. As a result of this, as long as a suitable driver is used and the enclosure is of proper size and tuning, group delay can be significantly lowered to nearly that of a sealed system, at least down until a certain frequency, at which point it'll sky rocket to insane levels... Hence a perceived improvement of transient response. However, there are some major draw backs to this type of design that need to be considered. First, these enclosures can be insanely MASSIVE. Depending on the specific driver being modeled, nearly 30 cubic feet for a -3 dB extended bass shelf isn't out of the ordinary. There are drivers that can be used in this fashion though in very small enclosures, such as the 10" and 12" Dayton Reference HO, or the Peerless XLS (I believe the XLS's would actually require a -6 dB EBS alignment tho)... But at the same time, these drivers also have short comings of their own to consider. Another major thing that needs to be brought to attention besides physical size constraints is tuning of the vent. Basically, more often than not, drivers like those just mentioned above (less efficient that can work in very small vented enclosures) will require a very long port. When you do this, port resonances will usually be lowered, often into the usable frequency range. This is obviously a bad thing. Although it can be partially dealt with by internal damping of the enclosure, I personally refuse to completely rely on such remedies, and would rather just avoid having a first port resonance in or near the frequency range for intended use. On top of all this, don't forget that when you tune a vented enclosure for such a low frequency, you often lose a substantial amount of maximum SPL and power handling capability..

If you haven't already, I suggest downloading WinISD Pro from LinearTeam and spending a long time entering the parameters for various drivers of interest and toying around with it. -3dB and -6dB EBS alignments are already built in, so that you can easily select them when attempting to model a system. However, you'll likely need to make some slight adjustments every time if you want to get optimum results, as it often gives you something a little bit off (too big of a hump at the roll off frequency, a bit too low of a tuning, or often a tuning that could be just a little lower, etc).
http://www.linearteam.dk/default.aspx?pageid=winisdpro

Personally, I now only use very capable woofers with large amounts of power + EQ for sealed systems. If I need more to reach a desired maximum SPL, I'll use multiple drivers. I'll only consider a vented system if the driver is suitable for a large low tuned EBS type alignment. Low FS, suitable QTS and VAS, and the maximum SPL desired from the system must be modeled, as the lower tuning will require a bit more excursion at the very lowest octaves. Therefore, the driver needs to have a good amount of XMAX to reach high SPL's, however, there are drawbacks to these kinds of drivers too, which I'll get to in a minute. Anyway, in my experience, the actual bass shelf must not be above the 30Hz region, or a very boomy response will be perceived around that area, due to room gain.

I've also just recently began to realize the drawbacks of all these massive excursion drivers popping up left and right over the last few years that seem so popular in the DIY community... as well as the disadvantages of using steep lowpass crossovers on such drivers to create subwoofers, rather than just using suitable large drivers crossed a bit higher as a 3-way system. I now have 4 15" woofers that I spent $1200 on and never used. They'll all be sold soon, and I'll lose quite a bit. Oh well. That's what happens when you keep putting off projects... you learn more in the meantime and decide it's not the best thing to do, or something better comes out, or you get better ideas..lol. Learning from my mistakes I guess. Anyway, I've heard many custom systems of the recent popular type... Fairly expensive subwoofer drivers that are inefficient with massive amounts of XMAX on hand, as well as setups that use multiple large drivers that are a bit more efficient, but don't have near as much capability in terms of maximum excursion. Even when both systems being compared are capable of moving very similar amounts of air, the difference between the two is VERY audible. Obviously, the one with larger/more cone area with less excursion always wins. Of course, this is simply because a properly designed system that uses less excursion for a given SPL will almost always yield less distortion and sound noticably better... than a smaller system capable of moving the same amount of air with one large driver flapping harder to do it. As for high order crossovers at very low points... this just raises group delay a substantial amount over the entire usable range.

By the way, I'm guessing you probably came across all the "LLT" talk over on AVS forum. If not, I wouldn't be the least bit surprised if what you saw had originally come from there and was spread around to another place by someone else from that forum, whether you realized it or not. Seen all the people over there with big tall tubes for subwoofers, made out of Sonotube concrete forms? That's what these are... large EBS alignments. They're commonly referred to as Sonosubs, self explanitory.

[BHTX]

A good example below. These woofers were originally recommended by the manufacturer for a smallish 2.2-2.5 cubic foot sealed enclosure, although they worked very well in vented systems. This is probably just because it was a small mobile audio manufacturer, and therefore focused primarily for the car audio market.

If you're going to jump on the high excursion bandwagon, this particular subwoofer driver was well-known around AVS to be one of the best for this type of large-low-tuned EBS system, although the amount of space required was pretty darn big at over 10 cubic feet per sub. Unfortunately, this particular small family owned mobile audio business quit making them like 3 years or so ago when they decided to part with Adire's XBL^2 (R.I.P), and I'm stuck with 4 of these that I haven't ever used. People seemed to go crazy over them for a while, but now no one seems as interested. Just my luck, I guess. Anyway..

15" Ascendant Audio Avalanche AVA15 -

T/S parameters and specs:

Qes .33
Qms 3.5
Qts .302
Fs 15.7
Vas 300.6 liters
Re 3.2 ohm
Le 2.40 mH
Sd 750 cm^2
Xmax 27 mm
Pe 800 watts
87.51 dB/W
Dual 2 ohm VC's
EBP: 47.6

Estimated frequency response of the Avalanche 15 in the following systems with a 4th order Linkwitz-Riley crossover at 120 Hz:

-3 dB EBS: 10.35 ft^3 @ 14.7 Hz
BB4/SBB4: 4.35 ft^3 @ 15.7 Hz
QB3: 4.075 ft^3 @ 20 Hz
.703 Qtc sealed: 2.5 ft^3 (Fsc: 37.95 Hz)
.703 Qtc sealed: 2.5 ft^3 + Linkwitz Transform EQ - fp 20 Hz, Qp .50

[BHTX]

Group delay (in milliseconds) for the above.

As you can see, the -3dB EBS alignment is comparable in transient response to that of the critically damped sealed system with EQ for Qtc of .5 and -3dB at 20 Hz, until the response reaches much below 30 Hz, at which point it skyrockets to insane levels and goes off the chart around the tuning frequency. Of course, the sealed system with the Linkwitz Transform for a critically damped system has the best transient response of them all, all the way down to the lowest frequencies. After those two alignments, we have the SBB4, tuned to the woofer's free-air resonant frequency (Fs) of 15.7 Hz. This enclosure is MUCH smaller than the massive 10.35 cubic foot EBS at just 4.350 cubic feet, but still quite a bit larger than the 2.5 cubic foot sealed box with EQ that beats them all. The group delay of this SBB4 alignment has a fairly steep but steady rise all the way down to the lowest frequency. That crappiest transient response would be that of the modeled Quasi-Butterworth alignment, due to its highest tuning frequency of 20 Hz. It is only slightly smaller than the SBB4 box at 4.075 cubic feet. Notice that the group delay of the large EBS system is actually a tad bit LOWER than that of the sealed box without EQ, until you get to below 30 Hz when it goes thru the roof. Also notice how the EQ'd sealed box's group delay is a fair amount lower than that of the same sealed box without EQ, until frequency reaches down to just above 20 Hz, at which point it increases very slowly upward, and remains extremely linear. Notice how group delay of the EBS system is quite a bit lower than that of the SBB4 box, until you get to frequencies below 20 Hz, at which point it goes WAY higher than all of them, up to a sharp peak at the tuning frequency...

[BHTX]

Excursion of the above:

Here's the predicted behavior of the cones movement for the previous alignments, with the maximum amount of power that could be fed to the driver until it reached xmax at whatever frequency in the usable range. For the 2 sealed systems, I raised the power until Xmax was reached at 30 Hz.

Apologies for not mentioning before now that the vented systems are the gray lines, and the two sealed alignments are the yellow lines, incase you didn't get that.

Basically, the lower the tuning of the system, the lower its power handling. Therefore, it should be noted that power handling/maximum SPL and low frequency extension is a natural trade-off due to physics. Low frequency response versus maximum SPL...decisions have to be made.

From left to right.. The EBS reaches maximum excursion at around 20 Hz with 773 watts. The BB4 at around 25 Hz with 1,315 watts. The QB3 at around 30 Hz with 1,750 watts. The sealed system with the Linkwitz-Transform EQ reached Xmax at 30 Hz with 950 watts. The sealed box without EQ reached Xmax at 30 Hz with 1,625 watts.

* Remember, this woofer was only rated at 800 watts RMS. After my toying around with 1 or 2 of them once or twice over the past couple of years that I've had them, I'd say it's probably pretty accurate, if not less. So, with that said, what's that massive 27 millimeters of Xmax for?? Exactly..

During a long low frequency signwave at these levels, it's very likely that the voice coils would have melted long before Xmax was ever reached, especially with the higher tuned/less extended systems..

[BHTX]

SPL of these systems with the input power listed above for each:

Notice how the QB3 alignment, which had the slightly higher tuning and worst transient response of all, has the greatest SPL capability. The EBS has the lowest, even lower max SPL above 30 Hz than the sealed + EQ system! Of course, with the sealed system without any EQ having the least low frequency extension, its maximum SPL is almost as high as the QB3 vented enclosure..

[BHTX]

..Hope this helps.

Hope my posts were pertaining to your questions, and that I didn't somehow misunderstand you, too! (happens sometimes)

[PeteMcK]

hi BHTX,
re: "I'll only consider a vented system if the driver is suitable for a large low tuned EBS type alignment. Low FS, suitable QTS and VAS"...
As a matter of interest, what do you consider suitable Qts & Vas values? (I think traditionally, EBS has been used with low Q High Vas drivers, although Low Q is probably more important?)
Cheers,
Pete McK

[hermanv]

Perhaps I'm missing the point. Bass at 105 dB SPL will cause things to wander around on your coffee table of their own volition, windows may rattle and long exposure would probably cause hearing damage.

While 110, 115 or even 120 dB SPL is academically interesting it serves little use in an average house. Why not design for low distortion or good bandwidth? This significantly expands your driver choice, lowers cost and increases driver bandwidth and most likely increases overall driver quality.

I own a powerful and expensive amplifier (400W RMS @ 4 Ohms), I think it would be hard pressed to deliver those kinds of volumes without compression or god forbid, clipping at some other part of the spectrum.

My experience says design for your ears, not the spec sheet.

[BHTX]

Originally posted by PeteMcK
As a matter of interest, what do you consider suitable Qts & Vas values? (I think traditionally, EBS has been used with low Q High Vas drivers, although Low Q is probably more important?)
Cheers,
Pete McK

I guess I should have said that a little differently. A very low Fs will usually tend to 'work' better for this type of design, since you normally want the bass shelf to be as low as possible. As for Qts and Vas, these are just a couple of the T/S parameters, along with Fs, that I usually tend to glance at with my eyes first when looking at any woofer to get an idea of its potential for a particular design or application. Those, along with all the other parameters, can usually kinda give you a 'feel' of the woofer and what it 'wants'. A higher Qts is probably going to require a larger enclosure, although certainly not always. That's partially what Vas is for. A low OR highish Qts can have a very small Vas, and therefore probably has a low Cms and probably a very stiff suspension (common with all these inefficient drivers that are willing to work in tiny enclosures, with oversized stiff foam surrounds and unecessary amounts of Xmax). A large Vas will normally require a much larger enclosure for this type of design. These 3 parameters can usually give somewhat of a quick idea of how low the woofer is willing to go, and how large or small of an enclosure it's willing to do it in.

Originally posted by hermanv
Perhaps I'm missing the point. Bass at 105 dB SPL will cause things to wander around on your coffee table of their own volition, windows may rattle and long exposure would probably cause hearing damage.

While 110, 115 or even 120 dB SPL is academically interesting it serves little use in an average house. Why not design for low distortion or good bandwidth? This significantly expands your driver choice, lowers cost and increases driver bandwidth and most likely increases overall driver quality.

I own a powerful and expensive amplifier (400W RMS @ 4 Ohms), I think it would be hard pressed to deliver those kinds of volumes without compression or god forbid, clipping at some other part of the spectrum.

My experience says design for your ears, not the spec sheet.

I agree with you 110%, which is why (as previously mentioned) these four 15" drivers that I have (the ones modeled above) are now for sale, without me ever using them, and with me taking a huge loss of $$$$ in the process. A few years ago when I started collecting them, I had gotten my misinformed self caught up in the current bandwagon of high SPL at low frequencies with ridiculous amounts of Xmax. Now, a few years later, I strive for the exact opposite. Now, I simply enjoy what sounds good, not what looks good or loud on paper (exactly as you mention). And as far as bass is concerned, I'll personally take a larger Sd with less excursion any day of the week, over an insane Xmax driver with ridiculously low efficiency.

I mentioned this more than once in all my rambling above, but it's clearly understandable if you didn't read all of that.