Product Review - SVS CS-Ultra Subwoofers - Part 2 - May, 2002
John E. Johnson, Jr.
The sound quality of the SVS CS-Ultra is really quite spectacular. I was presented with very powerful, deep, and magnificently clean sound effects as well as music. I used them as stereo subs, one in each front corner, Samson amp set to stereo, and my Theta Casablanca II crossed over at 50 Hz (no Rane in the circuit, as it is only for one channel, or two subs run in mono). I reconfirmed to myself that a good, in this case, great subwoofer doubles the fun with movies. All my favorite bass buster DVDs came through loud and clear. The studios put tens of millions of dollars into making movies, so the least we can do is play the DVDs on systems that deliver everything there is on the disc. I don't think there was anything I could not hear or feel that was coming off my DVDs into the SVS. Some consumers like the boomy harmonics that are produced by many subwoofers, but they drive me nuts. I like powerful bass, but if it is boomy, I turn it down until the boom is not there anymore. It might be a low level SPL in the deep frequencies that results, but I prefer that over loud and boomy. The SVS is clean all the way down in frequency, and all the way up in loudness. You cannot ask for more than that.
Music is wonderful through the SVS as well. Of course, you need a CD, SACD, or DVD-A that has some low frequencies to make use of it, but a piano can comply with that since its lowest key is about 28 Hz. Pipe organ is another instrument that needs deep and clean to work, and the SVS could handle all of it. I listen to a lot of 24/96 music on DVDs these days, and in particular, I like some of the old 1970s analog recordings that have been remastered for 24/96 digital. (I am not particularly interested in any 16/44.1 recordings that have been remastered as 24/96.) Anyway, the low strings on the guitars in the 24/96 jazz albums I obtained need clean bass. Again, if it is boomy, the sound is very unnatural. Kick drum is also something that just won't sound right if the subwoofer is boomy. I ought to know, as jazz drums is another hobby of mine.
You don't need to turn a subwoofer up very loud to reap its advantages. In fact, most of us (me included) tend to turn it up too high, just because it feels so good. That is fine once in awhile, but turn it back down now and then to listen in a real world. It does not take much deep output to add that final touch to the music, and that is what I found with the SVS. So clean, it just fits right in with the rest of a high-performance audio system. I can't get enough of music when it sounds this good.
Now, to the Bench . . .
All of the graphs shown below were gathered using a single subwoofer. Test cables, including computer-to-amplifier, and microphone-to-computer, were shielded XLR balanced, custom made for us by BetterCables.com. Nordost Red Dawn was used from amplifier-to-subwoofer.
First, let's talk about the room response, graph shown below. The red line at the top is with the calibrated microphone placed 8" from the subwoofer output region (bottom of the subwoofer). The remaining three lines were gathered from microphone placements in other parts of the room. Notice that the red line is the flattest. This is because having the microphone very close to the sub overpowers the room effects. The different peaks and valleys are where the signal is augmented or cancelled at the various frequencies. Consistent, though, is the decline beginning at about 35 Hz regardless of where the microphone was placed.
I then put the Rane PE-17 in the circuit and took a reading at the listening position with the Rane bypassed. Then, I used the various band units on the Rane at different frequencies to see how much I could flatten the response. The results are shown below. The red line is with the Rane bypassed, and the yellow line is with as much EQ adjustment as I could make when the Rane was active. You can see that the response is flatter, but certainly not entirely flat. This is because EQ can only help, rather than completely eliminate room peaks and dips. In particular, some dips are nulls, where the frequency cancels quite a bit, and no amount of increasing the output in that region will get rid of the valley, since they continue to cancel. So, in fact, you should not try to increase the signal in a dip. Rather, try to decrease the signal in peaks.
Since we don't have an anechoic chamber, we used MLS filters, which simulate anechoic response. I placed the microphone 8" from the subwoofer, and used white noise with MLS filters, and a 65 ms window, which means that everything which the microphone receives 65 ms after it starts receiving sound is eliminated from the measurement. MLS (Maximum Length Sequence) is not an exact science, and only an approximation, but it is all we have when no anechoic chamber is available. Most hi-fi magazines I know of, use the MLS technique, if they measure the frequency response at all. So, with that in mind, here are the graphs, shown below. The first one is the Impulse Response and the 65 ms window indicated by the purple lines. The window of 65 ms starts at 13.500 ms and ends at 75.500 ms. The 0 to 13.500 ms region represents the delay between when the signal is sent to the subwoofer by the software and when the sound actually starts being received by the software back at the computer. The second graph is the frequency response, calculated using the impulse response and a 65 ms window of that impulse response. It indicates that the SVS has maximum response at 50 Hz and drops by about 10 dB at 20 Hz. This is really quite good. Again, it is a computer calculation of what the subwoofer would do in an anechoic chamber. We should note here that speakers and subwoofers are designed to perform in rooms, not in anechoic chambers, but the MLS response is one way of standardizing measurements.
Next, I tested the THD of the SVS at several frequencies. Results are shown below. The microphone was 8" from the subwoofer output area again, and I input single sine waves (Rane PE-17 not in the circuit) until I recorded 100 dB SPL at the same spot where the microphone was placed (we use 100 dB as a reference point because it is a level that most subwoofers can reach, but it is nowhere near the maximum output capability of the SVS). The first graph shows the THD spectrum for a 20 Hz sine wave. The THD is less than 5%, which is acceptable. The bottom part of each graph has the right channel loop through with the same input signal to show what the input signal looks like by itself. If you are not familiar with reading THD spectra, it might appear that all those harmonics must be more than 5%. However, the dB scale (Y axis) is logarithmic, not linear. The 20 Hz peak is at - 30 dB. The second harmonic, at 40 Hz, is at - 56 dB, which is 26 dB less than the 20 Hz peak. For each 10 dB, the signal is 10 x different (higher or lower). So, the 40 Hz peak is less than 1/100th (less than 1%) the intensity of the 20 Hz peak. If you add up all the harmonics, it comes to 4.93% of the total signal.
At 31.5 Hz, the THD dropped to 1.42% (first graph below), and at 40 Hz, it was 1.39% (second graph below). These figures are quite good, and show how the THD drops off significantly as the frequency goes up.
The spectrum shown below is the maximum output I could get from a single SVS, with one channel of the Samson power amplifier, and the microphone 1 meter from the subwoofer. At 20 Hz, it was 112 dB. This is really an incredible performance. Of course, the THD jumps dramatically at this output, but there are lot of subwoofers out there that cannot even begin to output this kind of SPL at 20 Hz, and their THD could approach 20%. Again, this is with just one of the subs. Two would give you 118 dB. Obviously, you would not be watching movies at a constant 118 dB of 20 Hz output all through the movie, but what it means is that the SVS can really deliver when those nuclear explosions take place. Most of the time, the sub would be coasting along at about 80 dB - 90 dB. A piece of cake for the SVS.
The last spectrum, shown below, is the IMD when the input signals were 20 Hz and 30 Hz (red arrows). Harmonics are indicated by the green arrows. This is a torture test for speakers and is very revealing. Lots of harmonics are there, but all subwoofers produce them. The real test is what it sounds like, and the SVS was still very clean-sounding during this test. If the harmonics were overpowering, it would have sounded boomy, which it did not. In fact, the SVS is as close to a servo-feedback sub, without actually being one, as I have ever heard. Ah, the wonder of computer aided design.
As a comparison, shown below are some spectra from Subwoofer "X". It is a 10" model with a very powerful amplifier in a small enclosure. The measurements were taken in the same position as the SVS. Note that the maximum output with a 20 Hz sine wave is 98.5 dB, and the THD is very high. The third harmonic at 60 Hz is only about 7 dB below the fundamental. This makes a subwoofer sound boomy. It can also make a subwoofer sound "louder" because the harmonics are much easer to hear than the 20 Hz fundamental. Some consumers like this effect, and I have no argument with their choices.
Below are shown the Subwoofer "X" Impulse Response and Frequency Response using a 65 ms window. The response declines from a maximum at about 60 Hz by 15 dB at 20 Hz. Subwoofer "X" is about the same price (MSRP) as one SVS CS-Ultra and the accompanying amplifier.
If you are in the market for getting a subwoofer that represents one of the all-time great bangs for the buck, with astonishing performance, give the guys at SVS a ring . . . I mean an e-mail. I can't imagine anyone being disappointed in having one or two of them.
- John E. Johnson, Jr. -
© Copyright 2002 Secrets of Home Theater & High Fidelity
Return to Table of Contents for this Issue.