SVS SB16-Ultra powered subwoofer Measurements

Sidebar 3: Measurements

I took these measurements while setting up the SVS SB16-Ultra, to optimize its match with my Quad speakers by adjusting the sub's position. Although I used the Fast Fourier Transform (FFT) module of Studio Six Digital's AudioTools, my measurements should not be compared with the in-room responses taken by John Atkinson that accompany all loudspeaker reviews published in Stereophile. Mine were taken in a room very different from his, and JA uses a better microphone, and bases his averaged results on more data points.

I took my measurements with Studio Six Digital's iTestMic (footnote 1), a professional-grade test and measurement microphone for the iPhone 4 and iPad. The mike's connector required an Apple Lightning adapter to plug into my iPhone 6. That done, the mike was auto-calibrated by Studio Six's AudioTools app, v.10.5.6, which stores the microphone's output, then analyzes and plots the results. I drove the SB16-Ultra with a digital file of uncorrelated pink noise supplied by Kevin Voecks, of Revel Speakers.

I took eight measurements at slightly different points around the ear-level position at the back of my listening room chair, and averaged them. I set AudioTools to an 8kHz, 16-bit reading, 1/6-octave, with a 1-second graph decay. The FFT resolution was set to 0.5Hz. I took readings for the low-frequency range of AudioTools (5Hz–2kHz). The SPL readout was C-weighted.

I measured the Quad ESL-989s run full-range, with no external crossover filter (fig.2). Note that the in-room frequency response peaks at 80Hz, begins to fall off at 55Hz, and is down 20dB by 31Hz. There are room-related peaks at 40 and 200Hz. I then added the JL Audio CR-1 external electronic crossover, with its high-pass filter set to 63Hz, 12dB/octave and its low-pass filter set to 43Hz, 24dB/octave. The SB16-Ultra's internal parametric equalizer was used to linearize the response at 43 and 160Hz. The FFT analysis plotted the Quad ESL-989s' room response with the high-pass filter, measured with the sub turned off (fig.3, green line); the SB16-Ultra alone with the low-pass filter (orange); and the Quads and SVS combined (light blue). The Quads plus SVS sub showed a room frequency response of ±3dB, 20Hz–1kHz. A dip in frequency response at 38Hz was somewhat smoothed by pulling the subwoofer out from the corner a few inches into the room, just behind the right ESL-989.


Fig.2 Quad ESL-989, room response around ear level at listening chair at 96", averaged across 24" horizontal window (orange trace).


Fig.3 Spatially averaged room responses at 96" of: Quad ESL-989 alone (green), SVS SB16-Ultra alone (orange), and both together (light blue), all with the electronic crossover's high- and low-pass filters on.

The use of an extended FFT measurement down to 5Hz revealed a sound (blue line in both figs. 2 and 3) outside my awareness. It was 12dB below the Quad's room response at 31Hz, and peaked at 14Hz. I suspect that this was produced by the radon-mitigation exhaust system in the basement of our house, under the listening room.—Larry Greenhill

Footnote 1: Studio Six Digital's AudioTools, v.10.6.6, is available for download at the Apple App Store for $19.99. The company's iTestMic2 costs $199.99 and can be ordered at
SVS Inc.
260 Victoria Road
Youngstown,, OH 44515
(877) 626-5623

avanti1960's picture

Lacks high level inputs which I have found are critical for audiophile applications.
Driving with high level input (parallel with speaker terminals) sounds better and integrates significantly better when the speakers respond to the amplifier's signal- not the preamplifier signal.
Doubly important when tube amplification is used.

Also of critical importance is a flat, musical frequency response through its useable range. Unfortunately no measurements were provided.

John Atkinson's picture
avanti1960 wrote:
Unfortunately no measurements were provided.

Not sure I understand your point. This review does include in-room measurements.

John Atkinson
Editor, Stereophile

arve's picture

Lacks high level inputs which I have found are critical for audiophile applications.

Using high-level inputs for subwoofers is one of the more fatal mistakes of audio applications:

1. You will most definitely want to place the subwoofer so that speaker-boundary interference is not an issue. This typically involves placing the subwoofer close to the front wall of the room
2. Ideally, you will also want to kill sidewall reflections, which means placing them in the front corner of the room

This means that a subwoofer should typically not be placed at the same distance from the listener as the (front) pair of speakers. For the highest fidelity, it is therefore absolutely critical to have proper bass management, and the simple phase switch/dial on a subwoofer's controls is not nearly enough. You need to be able to apply appropriate delays between the subwoofers and main speakers.

Beyond the obvious issue of time alignment, a high-level input is also completely unable to provide adequate control over crossover frequency: Yes, you may be able to set a low-pass for the sub, but without any control of the high-pass, a subwoofer is a complete and total waste.

avanti1960's picture

to place the sub in the optimum location for best response with high level inputs.
in my experience with critical 2-channel audio the sub integrates better when it sees the amplifier's signal. same signal that the main speakers see. it sounds better and can literally disappear, which is why some of the subwoofer brands that audiophiles love have and recommend the use of high level signal input. connecting to the preamp out tends to make the subwoofer more likely to call attention to itself. in addition, i would never feed my main speakers through yet another crossover or DSP loop. I prefer the setting the sub crossover to the natural acoustic crossover (roll-off point) of the main speakers.
this is an area where home theater and 2-channel audio clearly have different needs and expectations.

Alberello's picture

"Ideally, you will also want to kill sidewall reflections, which means placing them in the front corner of the room"
- Why you want place a woofer in the worst acoustic part of a room?

"Beyond the obvious issue of time alignment..."
- If you use stereo configuration with two subs, you can place them to the side of the speaker so the woofer are at the same distance of the listener compared with main speakers.

"but without any control of the high-pass, a subwoofer is a complete and total waste."
- If you have parametric equalizer to reduce a couple of critical point in your room response, you can use full range response from your main speakers and use the subs with 30Hz crossover configuration just for improve the low base response.

JRT's picture

By adding a high pass filter to the mains, you not only reduce excursion at low frequencies (especially important where a woofer unloads to high excursion below the tuning frequency of a bass reflex Helmholtz resonance), but also that high pass on the mains provides poles that affect phase in the sum for a flatter summed response well above the subwoofer's low pass corner frequency, into lower midrange. ...if the crossover is properly executed.

Note that the effected spectrum is into the important telephonic frequency range where hearing is more sensitive to voice quality, and above the Schroeder frequency where Eigentones related to the room's modal response do not dominate the problem set.

At the link below, take a look at Siegfried Linkwitz's write-up, "Issues in loudspeaker design, page 5, section V, Crossover topology issues", and within that look at the graphic labeled "IMPROVED crossover topology" and the associated improvement made to the summed response in the lower midrange and upper bass, above the subwoofer's low pass corner frequency.

avanti1960's picture

filtering in the precious signal chain before the main speakers, but i will not be.
many speakers already include a low pass crossover within the design to help with such issues, the ones that do not usually have a very steep, natural roll-off at the F3 frequency, especially if it is a ported design.
measure with RTA software to find that frequency and adjust your subwoofer crossover to correspond with it.

ednazarko's picture

Love that phrase. Consider it stolen.

However, I've found that sometimes pressuring the room can turn musical bass into what I've heard colorfully described as low frequency flatulence. I've struggled with finding that boundary between just pressure or just polite bass with subs. I've had to mess with the subs in the two systems where I have them to find that sweet spot, where the bass is corporeal, but still has pitch. One of my subs has a very sophisticated (for 10 years ago) interface - the UI is very "pong" feeling - that's made it easy to get things right in several different rooms, with several different speakers. And enough different adjustments that you can get it exactly right, or horribly wrong if you don't know what you're doing.

Now I'm very interested in the SVS sub, because the other sub I have, while I've been able to get things right... it took WEEKS of twitchy tweaky adjustments in distances and angles and volumes (with hardly any controls over things like crossover characteristics.) The range of adjustments and app capability is very attractive. I need to find a physical seller I can convince to let me set up an SVS sub with a new set of speakers, so I can get a feel for the process and results.

You mention changing the sub profile for different types of music. The setup process for my older subs is painful enough that I don't even try. But in my AV system, which uses Goldenear Reference towers, I've got several different profiles set up in the AV pre-pro, and it's wonderful. With my headphone setup, ditto - I've got a profile that I call ECM (record label) and another I call "live orchestra recordings," among others. It'd be nice to have that capability in my other systems.

Doctor Fine's picture

One sub is OK I suppose but for deeper separation you really should use a minimum of two.
I use four.
Two SVS 12 inch sealed box subs and two Velodyne 15s.
Setup is to first get my mains sounding their best, then add the 12s and get it to sound its best and then add the 15s and tune them into the array.
Location of the sub voicecoils is best when physically time aligned with the main mid bass/midrange/tweet voice coils BUT in installation where it looks too bulky the big 15s MAY be hidden at the edges of the soundstage and by using a variable phase control they can be integrated nicely.
No idea what the earlier guy was on about as all my phase parametric crossover and other controls are still operable when using speaker level inputs on both my SVS and Velodynes
SVS delivers a high quality product in my experience and I have used the largest Rels, Velodynes and others for professional installs for years.
Sealed box subs are tight sounding and quick.
Ported can get more bass extension.
Yet another reason I use BOTH kinds.
Sub control is mandatory in any case.
Crossover, continuously variable phase control, parametric EQ and of course volume should be considered the minimum you should expect to see on a sub used in a musical system as opposed to a simple theater boomer.
Nice review.

karlosTT's picture

Interesting discussion here. I can't really accept the theory that High level works best, even for audiophile applications, though in practice I'm sure there are cases where it works well.
But 2 issues:-
1) High level means chaining 2 power amps in sequence after the pre, which could theoretically cause timing discrepancies between the sub and main speakers
2) It is generally accepted that filters and crossovers etc are best placed as far upstream as possible, to lessen the harm they do to the signal. This is the concept behind active speakers. So that would typically mean between the pre and the power, or actually in the digital domain, such as a DSP within a pre/pro. Using a sub's LFE input in this way also means there is no overlap of bass frequencies between the sub and main speakers, which can create dual sonic signature and reduced bass 'focus'.
Leastways, that's the theory as best I understand it.....