Audio Research LS5 preamplifier & BL2 input controller
- May 28, 2020
- No Comments
From the front, the LS5 looks identical to Audio Research’s popular LS2: two knobs on either side of the Audio Research nameplate, and a row of toggle switches along the bottom. But that’s where the similarities end; the LS5 is a completely different animal from the LS2, or even the balanced LS2B.
For starters, the LS5 is fully balanced from input to output. To understand the difference between a preamp with balanced inputs and outputs and a truly balanced preamp like the LS5, let’s look at how most balanced input and output preamps process a balanced input, using the LS2B as an example.
Although the LS2B has balanced inputs and outputs, the input signal is converted to single-ended by a differential amplifier at the input. The preamp then handles the internal signal as single-ended, with one gain stage per channel and one volume-control element per channel. The single-ended signal is then converted back to balanced with a phase splitter just before the output jacks. This technique adds two active stages to the signal path.
By contrast, a fully balanced preamplifier like the LS5 handles the signal differentially from input to output—no differential amplifier or phase splitter. This method requires double the circuitry: a gain path for each phase of the balanced signal, and a four-element volume control (± left channel, ± right channel). Moreover, the noise, distortion, and gain must be identical between the two phases. Any difference between the phases will become part of the signal. The advantages of a preamplifier being truly balanced are the elimination of two active stages (the differential amplifier at the input and the phase splitter at the output) and the fact that any noise or distortion common to both phases of the balanced signal will cancel when combined.
The LS5 has six pairs of balanced inputs, two pairs of balanced main output jacks, and one balanced tape loop. The front panel has an input selector switch, volume control, and four toggle switches. A green LED indicates when the unit is turned on, dimming when the LS5 is muted by a front-panel toggle switch, or is in automatic mute when the unit is first turned on.
One of the toggle switches selects the amount of gain provided by the LS5—either 12dB or 30dB. If you’re used to using the LS2, you could be in for a surprise when using the LS5; the LS5’s gain selector switch is in the same location as the LS2’s mute switch. It’s easy to throw the LS5’s gain switch thinking you’re muting the signal. Instead, the switch adds another 18dB of gain. Yeeow.
My initial auditioning suggested that the LS5 sounded better in the 30dB position, so that’s what I used for subsequent listening. This was a little too much gain for high-output source components. For example, the Sonic Frontiers SFD-2 digital processor’s maximum output level of 6.65V (RMS with a full-scale digital signal, balanced) forced me to use the LS5’s volume control around the 9 o’clock position.
Unlike all Audio Research preamps made in the past ten years, the LS5 is a pure tubed design. Though there are transistors and op-amps inside the LS5, these are not in the main audio signal path, apart from one FET in series with the audio signal, used as part of the gain switching.
The input stage consists of both halves of a Sovtek 6922 (a 6DJ8 dual triode), followed by both halves of a 12BH7A dual triode. The output driver is half of a 12BH7A, capacitively coupled (with high-quality MIT caps) to the output jack. This circuit is repeated for the other half of the balanced signal. In all, the LS5 uses ten tubes (four 6922s, six 12BH7As, footnote 1).
The impressive power supply fills the entire left-hand side of the chassis. A toroidal transformer is mounted vertically, supplying five bridge rectifiers from separate secondary windings. The 290V B+ supply uses discrete regulation, with Rel-Cap and Wonder Cap capacitors.
The LS5’s build quality is excellent. Between the extensive power supply and circuitry using ten vacuum tubes, the LS5 is jampacked with components. As you might expect, it runs quite warm.
BL2 input controller
The fully balanced LS5 has no single-ended inputs. Audio Research believed that the LS5’s fully balanced design would be compromised by including single-ended inputs in the same chassis. Consequently, they designed the companion BL2 balanced line driver to convert single-ended source signals to balanced. If your system has any single-ended source components, you’ll need the BL2 as a companion to the LS5.
The BL2 is a slim chassis about half the height of Audio Research’s preamps. The BL2 looks and functions just like a preamplifier, but with only 6dB of gain and no volume control. Seven pairs of single-ended inputs are provided, along with one balanced output and two unbalanced tape outputs.
Inside, the circuit uses JFETs at the input/phase-splitter stage, followed by an LT1223 op-amp for each phase of the balanced signal. The output line driver consists of a complementary pair of bipolar transistors, again one pair for each phase of the balanced signal. The circuit is direct-coupled, with DC offset trimmed out.
The power supply generates ±20V for the BL2’s gain stages. A combination of IC and discrete regulation is employed, with FETs used as the series pass transistors on the ±20V rails. The power supply uses WIMA caps for the audio circuitry supply, and electrolytics for the timer, mute, and other housekeeping circuits. The circuit and power supply are quite simple and don’t take up much real estate; the BL2 has a fairly simple task.
Listening to the LS5
I’ve used Audio Research preamps nearly continuously for the past three years and greatly enjoy their straightforward layout, ease of use, and quality feel of the controls. The LS5 continues this tradition.
When I put the LS5 in my system in place of the LS2B, I didn’t expect much of a change in the sound; the LS2B is an excellent preamplifier. It was immediately obvious, however, that the LS5 was a significant step up over the LS2B.
First, the LS5 had a much warmer, richer, more full-bodied sound compared to the LS2B. The entire bottom end was bigger, with a tremendous sense of bloom in the bass. The LS2, by comparison, was leaner and drier, with less air. This isn’t to say the LS5 was tubey or overly ripe; instead, the LS5 revealed more of the space and bloom in recordings (when called for), with a fuller rendering. Bass extension was also deeper, with a greater sense of weight and power.
The LS5’s treble was a significant improvement over the LS2B’s. The latter’s slightly whitish grain was replaced by the LS5’s liquidity and smoothness. The all-tubed LS5 was outstanding at presenting instrumental timbre without grain or etch. This contributed to the LS5’s less analytical-sounding rendering. Despite being smoother, more relaxed, and less up-front in the treble, the LS5 revealed a full measure of musical detail.
I also liked the LS5’s more laid-back perspective. The music was less forward—particularly in the mids—and set back behind the loudspeakers. The LS5 put more distance between listener and music, giving a feeling of ease and relaxation. Although the LS5’s perspective was easygoing, music had a sense of palpability, life, and immediacy—a rare combination.
The LS5’s soundstaging was spectacular—transparent, deep, layered, and infused with a beautiful sense of bloom. This is another quality that set the LS5 apart from the LS2B: the LS5 had a much greater sense of spaciousness, air, transparency, and bloom. The impression of individual instruments hanging in space was extraordinary. I heard none of the thickness and congestion sometimes heard from even high-priced preamps. Images were clearly delineated and focused, yet not in an analytical way. What made the LS5’s soundstage truly special, however, was its sense of air surrounding instrumental outlines. This bloom infused the music with a feeling of realness; the LS5’s presentation was the antithesis of sterile, canned, or “hi-fi”—the music had space to breathe and come to life.
Longer-term listening, particularly in comparison with the new $3495 Sonic Frontiers SFL-2 [currently under review by Russ Novak—Ed.] revealed the LS5 still to have a very slight whitish quality to the upper mids and treble. Nevertheless, I found the LS5’s combination of qualities—liquid mids, smooth treble, warm bass, slightly laid-back perspective, and wonderful openness—musically addictive. The LS5 also excelled at preserving the differences in perspective on different recordings.
Listening to the BL2
Evaluating the BL2 was a challenge. Had I simply run the balanced outputs of a digital processor into the LS5, and the unbalanced outputs into the LS5 via the BL2, I’d have ended up comparing the processor’s balanced and unbalanced outputs (and an additional run of interconnect) more than the BL2. To resolve this quandary, I compared the unbalanced outputs of the SFD-2 processor and the Vendetta, feeding the BL2/LS5 to the Sonic Frontiers SFL-2. (I used the ARC-supplied balanced interconnect between the BL2 and the LS5.) With a good feel for the differences between the LS5 and SFL-2 in balanced mode, this comparison would highlight the BL2’s effect on the sound.
I found the BL2 fairly neutral, although it did add a slight brightness to the sound. The treble lost some of its purity, sounding more forward and a little coarse. The fine layer of grain was marginal, but audible when compared to the purity of the LS5 on its own. The sound was a little drier, and some of the bloom was gone. Some of the LS5’s liquidity was diminished, and the presentation was more analytical and less relaxed. The BL2 made the LS5 sound a little more like the LS2B than the LS5 by itself—but still significantly better than the LS2B.
The Audio Research LS5 preamplifier is a superbly musical product. It provides the best of what tubes can do without the euphonic colorations of some tubed preamps. In relation to Audio Research’s very successful LS2 and LS2B, the LS5 clearly operates at another level of performance (footnote 2).
Footnote 1: LS2 and LS2B owners can significantly improve the sound of these preamps by replacing the standard Chinese 6DJ8 with the excellent Sovtek 6922.
Footnote 2: As this issue went to press, we were informed by Audio Research that the LS5 has been upgraded. Robert Harley will be comparing the two generations of the LS5 in a Follow-Up.—John Atkinson