Perla Audio Signature 50 Integrated Amplifier

Equipment report
Integrated amplifiers
Perla Audio Signature 50
Perla Audio Signature 50 Integrated Amplifier

A newcomer to the high-end audio scene, Perla Audio was founded in 2012 by Shane Duffy. It is a family-owned- and-operated corporation located in Sparks, Nevada, whose avowed mission is to design and manufacture the finest audio equipment in the world. Designing and refining musical playback components takes more than just a degree in electronics, says Duffy. “It takes passion, skill, and decades of searching for the highest quality parts and most sophisticated circuits.” The person responsible for the design of the Signature 50, chief engineer Ronald Van Robinson, has more than thirty years of experience in designing and engineering tube and solid-state amplifiers, ribbon and monitor loudspeakers, as well as source components such as digital-to-analog converters and phonostages. Duffy has been an audiophile since the early 80s when he started designing and building loudspeakers as a hobby. He began to design speakers again in 2010, and, as he was finalizing his design, realized that he needed someone with more experience to achieve the sound quality he was striving for. In 2011, he met Ronald Van Robinson, who had admired many renowned designers such as Dave Wilson, Alon Wolf, and the late Ken Shindo, and wished to design similar statement products for the high-end market. It turned out that Ron and Shane shared similar goals, and thus Ron was a perfect fit for Perla Audio.

Let’s start with the basic facts. The Signature 50 (S50) is a push-pull Class AB integrated amplifier. The front-panel volume control is a 48-step resistive-ladder attenuator that is remote controlled. Sadly, in line with the current trend toward minimalism, there is no balance control. The input selector controls three line-level inputs. The layout is said to be dual mono with every component being hand-soldered to a custom 3.35mm-thick, two-ounce-copper, 24k-gold-plated PCB board, using Cardas ultra-pure, quad-eutectic solder. A pair of ON Semiconductor complementary NPN-PNP bipolar power transistors is used per channel to generate 50 watts without any global feedback. The transistors are biased above cutoff so that the first watt is Class A. It’s a cool-running design, which accounts for the lack of external heatsinks. (Just about every solid-state amplifier in my listening room is outfitted with substantial heatsinks, the one notable exception being the GAS Ampzilla II, which is fan-cooled.) That’s a reflection of Class A operation or heavily biased Class AB designs, which generate a substantial amount of waste heat. In contrast, the S50 chassis is fully enclosed and quite substantial, being machined from a solid aluminum billet. It has been posited that Class AB is a compromise between the efficiency of Class B and the purity of Class A. Theoretically speaking, that may be true, but in the real world it ain’t necessarily so. Sonically, the end result very much depends on execution. A prime example of Class AB excellence is the McIntosh MC275, and to that category I can safely add the S50.

The power supply has received considerable attention, to the point where those outside the high-end arena might consider it overkill. But as Shane Duffy puts it: “While the rest of the world moves toward cheaper and smaller penny parts, we build our products with the highest-quality parts, no compromise!” For starters, there are two shielded 160VA toroidal power transformers. The bridge rectifier is built up by doubling the diodes at each position in the bridge to lower its impedance, and all of the diodes are low-noise Schottky types. There are no capacitors in the signal path. And that’s a good thing, as the best sounding cap is no cap at all. However, the power supply capacitance is totally on steroids—an amazing 376,000uF built up by using 80 electrolytic and 84 polypropylene caps in parallel. It’s a massive reservoir that continues to power the speakers for almost 30 seconds after the unit is switched off. The benefits include excellent low-frequency ripple filtering of the rectified DC voltage and a substantial low-impedance current reserve.

First impressions were positively surprising. My first solid-state amplifier in the early 1970s was the Citation 12, a consensus “best buy” in its day. In hindsight, it was one of the worst sounding amps I have endured over the years. It was quickly replaced by the Dynaco Stereo 70, which was less detailed but otherwise sounded shockingly more musical. Jim Bongiorno’s Ampzilla and Son of Ampzilla amps were giant steps in the right direction—a bit grainy, but far more dynamic than the Citation. The trend over the past 40 years has been has been toward a much smoother presentation, but emotional fire has been curiously absent from most modern transistorized designs. It struck me immediately that the Perla S50 was a definite exception to the smooth and sterile norm. It went the extra mile, combining ease of listening with plenty of expressive power, though it wasn’t the end-all for microdynamic persuasiveness. The winners in that category, being endowed by a potent first watt, are well-designed single-ended triode (SET) amps. In contrast, the S50 appears to be slightly muted dynamically, but I would still rate it as a respectable 7 on a scale of 1 to 10. However, to be fair, no solid-state amplifier I’ve auditioned to date has equaled a top-notch SET in microdynamic prowess, though Nelson Pass’s SIT-1 monoblocks have come very close. That’s one of the reasons why I believe that there will always be tubes in audio heaven.

The S50 was capable of generating slightly warm, sweet, and liquid harmonic textures, though it could never be mistaken for a tube amp simply because of its superior transient speed and the lack of a euphonic harmonic signature. In this respect it bettered even the much more expensive Lamm Audio M1.2 Reference monoblocks. However, the voicing of these two amps was significantly different. While the M1.2 was a master of tonal neutrality, the S50 consistently dished out darkish harmonic hues. Consequently, harmonic colors appeared less saturated relative to the M1.2 and a tad paler than the real thing. I found the M1.2’s sunnier disposition much easier to live with over the long haul, though I suspect that a slightly bright speaker would provide the S50 with a perfect tonal antidote. In compensation for its closed-in treble range, the S50 served up amazing soundstage transparency and image resolution.

The sense of space, while not quite tube-like, was well beyond the reach of most solid-state amps at this price point. In particular, image specificity was nothing short of spectacular. To put it into perspective, let me compare the S50 to one of my recent vintage acquisitions, NYAL’s Moscode Minuet in A. This is a fun vintage 1980s preamp with a distinctly tubey midrange. But it apparently can’t image to save its life. Imagine a soundstage filled with immense and gooey toasted-marshmallow-shaped image outlines bloated and blended into an indistinct mix. Well, the S50 gives a completely opposite impression. Image outlines were tightly focused within a transparent soundstage. For example, a lack of perceptible veiling made it incredibly easy to locate and follow operatic singers as they moved across the stage. In my book, transparency equates with the ability to zoom in and access the inner recesses of the soundstage, making it possible for the auditory system to construct a spatial impression complete with believable width and depth perspectives. And this the S50 did really well. One would think that there ought to be a linear association between passive parts-quality and transparency, and the S50 certainly makes a strong argument for such a relationship.

As one might expect, the beefy capacitive reservoir provided plenty of instant energy to handle macrodynamic peaks. In addition, bass range performance was impeccable with accurate pitch definition and a strong rhythmic drive. I measured a source impedance of 0.18 ohms, which gives a damping factor of 44 with an 8-ohm load. Although some designers have advocated for much higher damping factors, anything over about eight should be fine in practice. Current drive into low-impedance loads should not be an issue, so that a matching medium-sensitivity 4-ohm load would be expected to be more than satisfactory—at least when the amp/speaker is used in small to medium-sized rooms.

The Signature 50 represents Perla Audio’s grand entrance onto the high-end scene. It is not only a contender for top honors at its price point but it possesses the brawn and finesse to compete with much more expensive integrated amplifiers. Yes, it does need to be mated carefully to a speaker load that is comfortable with its darkish presentation, but the payoff is a soulful and vivacious connection with the music. It reminds me of one of my favorite classic solid-state stereo amps, the Class A Threshold 400A, sharing as it does a similar tonal balance. Mini-monitor aficionados rejoice! You will not be disappointed by the S50’s imaging specificity and soundstage transparency.


Power: 50Wpc into 8 ohms
Power consumption at standby:  6 watts
Frequency response: DC–40kHz +/-1dB
Inputs impedance: 100k ohms
Voltage gain: 27dB
Dimensions: 14.5" x 5.625" x 18"
Weight: 44.2 lbs.
Price: $9000

1575 San Pedro St.
Sparks, NV 89436
 (775) 722-1488

Associated Equipment
EnigmAcoustics Mythology M1 & Tannoy Revolution XT 8F loudspeakers; Kuzma Reference turntable; Kuzma Stogi Reference 313 VTA tonearm; Clearaudio Da Vinci V2 MC phono cartridge; Monarchy Audio NM24 and April Music Eximus DP1 DACs; FMS Nexus-2, Wire World, and Kimber KCAG  interconnects; Acoustic Zen Hologram speaker cable; Sound Application power line conditioners