Shindo Aurieges Equalizer Amplifier Measurements

Sidebar 2: Measurements

In his review of this expensive, two-box, tubed phono preamplifier in October 2015, Art Dudley concluded that were he "the owner of a lesser full-function preamplifier, or virtually any line-only preamp, the new Aurieges would be at the top of my list. In its sound, its musicality, and its distinction as a handmade, artisanal component of exceptional quality and consequently high value, it is every inch a Shindo."

However, when he used the Aurieges phono preamplifier with April Music's Aura Note V2 CD receiver while reviewing the latter for the April 2016 issue, Art was disturbed to hear crackles on percussion instruments while playing an LP. (The rest of the system was an EMT TSD 15 cartridge with a nominal output of 1.05mV at 1kHz with a recorded velocity of 5cm/s, and an Auditorium 23 Hommage T2 transformer.) As I wrote in the measurements sidebar to AD's review, I found that the Aura Note's line input overloaded at levels greater than 3.3V. Art shipped to me the Aurieges Equalizer Amplifier (serial no.0001) so that I could investigate the matter.

The Aurieges has a single pair of input jacks (RCA) for stereo cartridges. A pair of transformers provide the additional gain needed for moving-coil (MC) cartridges; these transformers can be disconnected—as was done in Art's review sample—leaving the stereo input compatible with moving-magnet (MM) cartridges, with a measured gain of 39.9dB. (There is also a single RCA jack for mono MC cartridges. This input has a single transformer; with a 1kHz signal, the total gain was 59dB.) I conjectured, in my measurements sidebar for the Aura Note, that the 60dB gain of the combination of the Shindo Aurieges phono preamplifier's MM stage and the Hommage T2 transformer would give an overload margin for the receiver's line inputs of 10dB compared with the amplified voltage of the EMT cartridge at 5cm/s. As peaks on LPs can exceed 10dB ref. the same 5cm/s, this was very likely the source of the crackles.

But as I now had the Shindo Aurieges on my test bench, I ran a complete set of measurements, using my Audio Precision SYS2722 system (see the January 2008 As We See It"). As always with phono stages, I experimented with the grounding between the Shindo and the Audio Precision system to get the lowest level of noise. I ended up running a separate ground wire from the Shindo's chassis to the SYS2722's ground connection.

The MM input impedance varied from 52k ohms at low frequencies to 87k ohms at 1kHz to 73k ohms at the top of the audioband; the MC input impedance (mono) varied from 204 ohms at 20Hz to 360 ohms at 1kHz to 349 ohms at 20kHz. Both the stereo and mono inputs preserved absolute polarity (ie, were non-inverting). The output impedance at 20kHz was very low, at 23 ohms. It rose to 139 ohms at 1kHz, which is still low, but at 100Hz it measured a very high 3700 ohms. It rose even further as the frequency dropped, to 11k ohms at 50Hz and 16.5k ohms at 20Hz. I assume that this drastic rise in impedance is due to the Aurieges Equalizer Amplifier's output being taken from the tubes' plates via a coupling capacitor. With line preamplifiers having an input impedance of less than 100k ohms, this increasing output impedance at low frequencies will roll off the bass.

Fig.1 Shindo Aurieges, response with RIAA correction (left channel blue, right red) (1dB/vertical div.).

This should be audible, except that the Aurieges's RIAA correction (fig.1) also rolls off the low frequencies, reaching –6dB at 39Hz in the right channel (red trace) and 32Hz in the left (blue) compared with the level at 1kHz. Perhaps compensating for this when it comes to sound quality, there is a 3dB boost centered just above 100Hz. There is also a broad 0.5dB depression in the treble that deepens when the Aurieges is set to "Columbia" equalization, and flattens out when the preamp is set to "78."

Fig.2 Shindo Aurieges, spectrum of 1kHz sinewave, DC–1kHz, at 1V into 100k ohms (linear frequency scale).

Fig.3 Leema Acoustics Essentials, spectrum of 1kHz sinewave, DC–1kHz, at 1V into 100k ohms (linear frequency scale).

Measuring channel separation was difficult because of high levels of infrasonic noise, but it appeared to be close to 50dB across the audioband. Even with the ground wire linking the preamp to the analyzer, the Aurieges was very noisy. The MM input's unweighted, wideband signal/noise ratio, ref. 1kHz at 5mV, was just 39dB in the left channel, 43dB in the right. These ratios hardly changed when the measurement bandwidth was restricted to the audioband, but did improve to 62.5 and 64.5dB, respectively, when A-weighted. Fig.2 shows that the noise floor, ref. 1kHz at 10mV input voltage, rises dramatically below 400Hz. For reference, fig.3 shows the spectrum of the noise floor, taken under identical conditions, of the inexpensive Leema Essentials phono preamplifier ($749), which AD reviewed in October 2015. While some low-level power-supply artifacts are present in the Leema's output, its level of noise in the upper midrange is around 15dB lower than the Aurieges—and below 100Hz, the tubed preamplifier is a whopping 30dB noisier than the solid-state design. This kind of noise is called "1/f" noise because its spectral content continues to rise with decreasing frequency, right down to DC. It's also called "flicker noise," because of the characteristic bouncing behavior of the signal trace on an oscilloscope.

Fig.4 Shindo Aurieges, spectrum of 1kHz sinewave, DC–10kHz, at 1V into 100k ohms (linear frequency scale).

Where the Aurieges excels is in the linearity of its circuit, which turns out to have the highest overload margin I have measured: an extraordinary 43.2dB at 1kHz, ref. 5mV, equivalent to an output voltage of 70.7V! This was the MM stereo input; the mono MC input margin was even greater, at 48.5dB at 1kHz ref. 500µV. Even at 20kHz, the MM margin was a very high 34.7dB. As a result, distortion at normal signal levels was minuscule. Fig.4 shows a wider-band spectrum taken under the same conditions as fig.2. The only distortion harmonic that can be seen is the second, at –91dB in the left channel and –94dB in the right, the latter equivalent to just 0.002%.

Fig.5 Shindo Aurieges, HF intermodulation spectrum, DC–30kHz, 19+20kHz at 1V peak into 100k ohms (linear frequency scale).

Intermodulation distortion (fig.5) was similarly low in level. These graphs were taken into a high 100k ohm load. With loads below a few thousand ohms, the distortion rose significantly.

On the test bench, Shindo Laboratory's Aurieges Equalizer Amplifier produced mixed results. The very low levels of distortion and extraordinarily high overload margins must be set against its nonflat RIAA response, its high level of flicker noise, and its high output impedance at low frequencies, which will make system optimization tricky.—John Atkinson

Shindo Laboratory/Tone Imports
(646) 425-7800

junker's picture

How is it loaded with the resistors on the Lundahls? What is the gain? What cartridges did you use it with? EMT? Shindo? SPU? Where was the Masseto volume control at, with your amps at their typical gain position? There just isn't enough info here about what gear it would work well with?

Despite the ability to play 78's and do mono, it appears to be specifically targeted to the Aurieges-L or Vosne - Romanee with the SPU Shindo cartridge. It just seems way too limited to me vs. using a SUT in most other cases.