Link: reviewed by Philip Beaudette on SoundStage! Hi-Fi on June 1, 2023
General information
All measurements taken using an Audio Precision APx555 B Series analyzer.
The S3 B was conditioned for 30 minutes at 2Vrms (1Vrms unbalanced) at the output before any measurements were taken.
The S3 B offers one pair of unbalanced (RCA) and balanced (5-pin mini XLR) inputs, for a moving-magnet (MM) or moving coil (MC) cartridge, selectable by a gain switch (40/45/60/65 dB, unbalanced output) on the front panel. There are both unbalanced (RCA) and balanced (XLR) outputs. Besides the extra 6dB in gain between the balanced and unbalanced outputs, we found no appreciable differences in terms of THD+N; however, 1kHz FFTs are still shown in this report for every input/output configuration for comparison. Also included are a grounding post, a subsonic filter, four capacitive loading settings (50/150/300/400pf), and five resistive loading settings (10, 50, 100, 1000, 47k ohms).
Unless otherwise specified, the balanced inputs and outputs were used for all measurements, with the subsonic filter off, capacitive loading at 50pF. For the MM configuration, gain was set to 40dB (46dB with balanced outputs) and 47k ohms loading, while the MC configuration was set to 60dB (66dB with balanced outputs) and 100-ohm loading. Using the default settings above, to achieve the reference output voltage of 2Vrms (1Vrms unbalanced) at 1kHz, 10mVrms was required with the MM configuration, and 1.4mVrms with the MC configuration. A custom female 5-pin mini XLR to dual female 3-pin XLR cable adapter was required to interface between the Audio Precision’s balanced outputs and the S3 B’s balanced inputs.
Published specifications vs. our primary measurements
The table below summarizes the measurements published by Pro-Ject for the S3 B compared directly against our own. The published specifications are sourced from Pro-Ject’s website, either directly or from the manual available for download, or a combination thereof. With the exception of frequency response, where the Audio Precision bandwidth was set at its maximum (DC to 1MHz), assume, unless otherwise stated, 2Vrms balanced output into 200k ohms (100k ohms unbalanced) and a measurement input bandwidth of 10Hz to 90kHz, and the worst-case measured result between the left and right channels. For the gain setting measurements, the input impedance was set to 47k ohms.
| Parameter | Manufacturer | SoundStage! Lab |
| Input impedance | 10/50/100/1k/47k ohms | 10.2/56.2/99.4/981/52k ohms |
| Gain | 40/45/60/65dB | 40.0/44.7/59.9/64.5dB |
| Signal-to-noise ratio (MM/40dB, max output voltage) | 110dB (A-weighted) | 110dB (A-weighted) |
| Signal-to-noise ratio (MC/60dB, max output voltage) | 90dB (A-weighted) | 90.8dB (A-weighted) |
| THD at 1kHz (MM) | <0.001% | <0.0005% |
| THD at 1kHz (MC) | <0.005% | <0.004% |
| THD (MM, 20Hz-20kHz) | <0.008% | <0.004% |
| THD (MC, 20Hz-20kHz) | <0.01% | <0.02% |
| RIAA response accuracy | ±0.3dB (20Hz-20kHz) | +0.1/-0.38dB (20Hz-20kHz) |
| Subsonic filter | -3dB at 20Hz (18dB/Oct) | -3dB at 20Hz (18dB/Oct) |
Our primary measurements revealed the following using the balanced MM configuration (unless specified, assume a 1kHz sinewave, 2Vrms output into a 200k ohms load, 10Hz to 90kHz bandwidth):
| Parameter | Left channel | Right channel |
| Crosstalk, one channel driven (10kHz) | -115.6dB | -114.4dB |
| DC offset | <9.5mV | <9.5mV |
| Gain (default) | 46.0dB | 46.1dB |
| IMD ratio (18kHz and 19kHz stimulus tones) | <-82dB | <-82dB |
| IMD ratio (3kHz and 4kHz stimulus tones) | <-97dB | <-97dB |
| Input impedance | 51.7k ohms | 56.4k ohms |
| Input impedance (unbalanced) | 51.6k ohms | 52.0k ohms |
| Maximum output voltage (at clipping 1% THD+N) | 21.2Vrms | 21.2Vrms |
| Noise level (A-weighted) | <55uVrms | <55uVrms |
| Noise level (unweighted) | <120uVrms | <120uVrms |
| Output impedance | 199.6 ohms | 200 ohms |
| Output impedance (unbalanced) | 122.0 ohms | 122.4 ohms |
| Overload margin (relative 5mVrms input, 1kHz) | 26.6dB | 26.6dB |
| Overload margin (relative 5mVrms input, 20Hz) | 7.6dB | 7.53dB |
| Overload margin (relative 5mVrms input, 20kHz) | 33.9dB | 33.9dB |
| Signal-to-noise ratio (A-weighted) | 90.2dB | 90.2dB |
| Signal-to-noise ratio (unweighted) | 84.5dB | 84.9dB |
| THD (unweighted) | <0.0005% | <0.0005% |
| THD+N (A-weighted) | <0.0027% | <0.0027% |
| THD+N (unweighted) | <0.0057% | <0.0065% |
Our primary measurements revealed the following using the balanced MC configuration (unless specified, assume a 1kHz sinewave, 2Vrms output into a 200k ohms load, 10Hz to 90kHz bandwidth):
| Parameter | Left channel | Right channel |
| Crosstalk, one channel driven (10kHz) | -100.8dB | -100.1dB |
| DC offset | <9.5mV | <9.5mV |
| Gain (default) | 65.8dB | 65.9dB |
| IMD ratio (18kHz and 19kHz stimulus tones) | <-81dB | <-81dB |
| IMD ratio (3kHz and 4kHz stimulus tones) | <-78dB | <-78dB |
| Input impedance | 100 ohms | 99.7 ohms |
| Input impedance (unbalanced) | 99.6 ohms | 99.2 ohms |
| Maximum output voltage (at clipping 1% THD+N) | 21.1Vrms | 21.1Vrms |
| Noise level (A-weighted) | <530uVrms | <530uVrms |
| Noise level (unweighted) | <1.1mVrms | <1.1mVrms |
| Output impedance | 199.6 ohms | 200 ohms |
| Output impedance (unbalanced) | 122.0 ohms | 122.4 ohms |
| Overload margin (relative 0.5mVrms input, 1kHz) | 29.7dB | 29.7dB |
| Overload margin (relative 0.5mVrms input, 20Hz) | 11.1dB | 11.1dB |
| Overload margin (relative 0.5mVrms input, 20kHz) | 50.1dB | 50.1dB |
| Signal-to-noise ratio (A-weighted) | 70.1dB | 70.4dB |
| Signal-to-noise ratio (unweighted) | 64.7dB | 65.2dB |
| THD (unweighted) | <0.0043% | <0.0043% |
| THD+N (A-weighted) | <0.027% | <0.027% |
| THD+N (unweighted) | <0.06% | <0.06% |
Frequency response - MM input
In our measured frequency-response plots above for the MM configuration measured at the balanced output, the blue/red traces are with the subsonic filter disengaged, while the purple and green represent the responses with the subsonic filter. An inverse RIAA EQ is applied to the input sweep, so that if a device were to track the RIAA curve perfectly, a flat line would emerge. The S3 B is within +/-0.1dB or so of flat from 20Hz to 10kHz. These data just about corroborate Proj-Ject’s claim of 20Hz to 20kHz +/-0.3dB—we measured -0.38dB at 20kHz. With the subsonic filter engaged, there is steep attenuation below 30Hz at 18dB/octave with the corner frequency at 20Hz, as advertised. In the graph above and some of the graphs below, we see two visible traces: the left channel (blue or purple) and the right channel (red or green). On other graphs, only one trace may be visible, this is because the left and right channels are tracking extremely closely, so they do not show a difference with the chosen axis’ scales.
Frequency response - MC input
In our measured frequency-response plot above for the MC configuration, the S3 B yields virtually the same results as with the MM configuration above.
Phase response - MM input
Above is the phase response of the S3 B for the MM configuration, from 20Hz to 20kHz. The S3 B does not invert polarity. Since phono preamplifiers must implement the RIAA equalization curve, which ranges from +19.9dB (20Hz) to -32.6dB (90kHz), phase shift at the output is inevitable. Here we find a worst case -50 to -60 degrees around 200Hz and 3-5kHz.
Phase response - MC input
Above is the phase response of the S3 B for the MC configuration, from 20Hz to 20kHz. The results are virtually identical to the MM configuration above.
THD ratio (unweighted) vs. frequency - MM and MC inputs
The chart above shows THD ratios as a function of frequency, where the input sweep is EQ’d with an inverted RIAA curve. The balanced output voltage is maintained at the refrence 2Vrms. The red/blue (L/R) traces represent the MM configuration, and purple/green for MC. For the MM configuration, THD values are very low, ranging from 0.003% at 20Hz down to 0.0003% at 1kHz (left channel), then up to 0.003% at 20kHz. The MC configuration yielded higher THD ratios, ranging from 0.03% at 20Hz, down to around 0.002% at 1 to 4kHz, then back up to 0.01% at 20kHz.
THD ratio (unweighted) vs output voltage at 1kHz - MM and MC inputs
The chart above shows THD ratios as a function of output voltage for the balanced output. The red/blue (L/R) traces represent the MM configuration, and purple/green for MC. For the MM configuration, THD values at 100mVrms are at 0.005%, then dip as low as 0.0003% at 2Vrms, then a steady rise up to 0.02% at the “knee” right around 20Vrms. For the MC configuration, THD values at 100mVrms are at 0.05%, then steadily decrease down to 0.002% at 3 to 5Vrms, then a steady rise up to 0.02% at the “knee” right around 18Vrms The 1% THD values for the both configurations are reached at 21.1Vrms at the output. It’s important to mention that anything above 1-2Vrms is not typically required for most line-level preamps or integrated amps.
THD+N ratio (unweighted) vs output voltage at 1kHz - MM and MC inputs
Above we can see a plot of THD+N ratios as a function of output voltage for the balanced output. The red/blue (L/R) traces represent the MM configuration, and purple/green for MC. For the MM configuration, THD+N values at 100mVrms are at 0.1%, then dip as low as 0.002% around 5 to 7Vrms, then a steady rise up to 0.02% at the “knee” right around 20Vrms. For the MC configuration, THD+N values at 100mVrms are at 1%, then dip as low as 0.015% around 10Vrms until the “knee”at 18Vrms.
THD+N ratio (A-weighted) vs output voltage at 1kHz - MM and MC inputs
Above we can see a plot of THD+N (A-weighted) ratios as a function of output voltage for the balanced output. The red/blue (L/R) traces represent the MM configuration, and purple/green for MC. For the MM configuration, THD+N values at 100mVrms are at 0.05%, then dip as low as 0.0015% around 5Vrms. For the MC configuration, THD+N values at 100mVrms are at 0.5%, then dip as low as 0.01% around 7 to 10Vrms.
FFT spectrum, 1kHz - MM input (balanced in, balanced out)
Shown above is a fast Fourier Transform (FFT) of a 1kHz input sinewave stimulus for the MM configuration, which results in the reference voltage of 2Vrms (0dBrA) at the balanced output using the balanced inputs. Here we see clean results. Signal harmonics are low and can be seen at the second (2kHz) and third (3kHz) positions at -115/110dBrA (left/right), or 0.0002/0.0003%, and -120dBrA, or 0.0001%, respectively. On the left side of the signal peak, there are no power-supply-related harmonics visible.
FFT spectrum, 1kHz - MM input (balanced in, unbalanced out)
Shown above is a fast Fourier Transform (FFT) of a 1kHz input sinewave stimulus for the MM configuration, which results in the reference voltage of 2Vrms (0dBrA) at the unbalanced output using the balanced inputs and the 45dB gain setting. Here we see clean results, but with signal harmonics at the second (2kHz) and third (3kHz) positions slightly higher (5-10dB) than with the balanced outputs and 40dB gain setting above. On the left side of the signal peak, there are no power-supply-related harmonics visible.
FFT spectrum, 1kHz - MM input (unbalanced in, balanced out)
Shown above is a fast Fourier Transform (FFT) of a 1kHz input sinewave stimulus for the MM configuration, which results in the reference voltage of 2Vrms (0dBrA) at the balanced output using the unbalanced inputs. Here we see clean results, with the signal harmonics at the second (2kHz) and third (3kHz) positions essentially the same as with the balanced input and output configuration above. However, unlike the balanced input and output configuration above, on the left side of the signal peak, there is a visible but small power-supply related peak at 60Hz at -100dBrA, or 0.001%.
FFT spectrum, 1kHz - MM input (unbalanced in, unbalanced out)
Shown above is a fast Fourier Transform (FFT) of a 1kHz input sinewave stimulus for the MM configuration, which results in the reference voltage of 2Vrms (0dBrA) at the unbalanced output using the unbalanced inputs and the 45dB gain setting. Here we see clean results, but with signal harmonics at the second (2kHz) and third (3kHz) positions slightly higher (5-10dB) than with the balanced outputs and 40dB gain setting above. Like the unbalanced input and balanced output configuration above, on the left side of the signal peak, there is a visible but small power-supply-related peak at 60Hz at -100dBrA, or 0.001%.
FFT spectrum, 1kHz - MC input
Shown above is a fast Fourier Transform (FFT) of a 1kHz input sinewave stimulus for the MC configuration, which results in the reference voltage of 2Vrms (0dBrA) at the balanced output using the balanced inputs. Here we see clean results. Signal harmonics are virtually non-existent above the noise floor, with only the second (2kHz) harmonic peak visible just barely visible from the right channel at -100dBrA, or 0.001%. On the left side of the signal peak, there are no power-supply related harmonics visible above the higher noise floor due to the extra gain.
FFT spectrum, 50Hz - MM input
Shown above is the FFT for a 50Hz input sinewave stimulus measured at the balanced output for the MM configuration. The X axis is zoomed in from 40Hz to 1kHz, so that peaks from noise artifacts can be directly compared against peaks from the harmonics of the signal. Only the second (100Hz) signal harmonic can be seen at around -110dBrA, or 0.0003%. There are no power-supply-related harmonics visible.
FFT spectrum, 50Hz - MC input
Shown above is the FFT for a 50Hz input sinewave stimulus measured at the balanced output for the MC configuration. The X axis is zoomed in from 40Hz to 1kHz, so that peaks from noise artifacts can be directly compared against peaks from the harmonics of the signal. Only the second (100Hz) signal harmonic is barely visible above the noise floor at just below -90dBrA, or 0.003%. There are no power-supply-related harmonics visible.
Intermodulation distortion FFT (18kHz + 19kHz summed stimulus) - MM input
Above is an FFT of the IMD products for an 18kHz and 19kHz summed sinewave stimulus tone for the MM configuration measured at the balanced output. The input RMS values are set so that if summed (for a mean frequency of 18.5kHz), would yield 2Vrms (Reference or 0dBRa) at the output. Here we find the second order modulation product (i.e., the difference signal of 1kHz) at -100dBrA, or 0.001%. We can also see the third-order modulation products (i.e., 17kHz and 20kHz) sitting just above the -100dBrA level.
Intermodulation distortion FFT (18kHz + 19kHz summed stimulus) - MC
The last graph is an FFT of the IMD products for an 18kHz and 19kHz summed sinewave stimulus tone for the MC input. Here we find that the second-order modulation product (i.e., the difference signal of 1kHz) is not visible above the -100dBrA noise floor. We can also see the third-order modulation products (i.e., 17kHz and 20kHz) are lower than with the MM configuration above, sitting at around -110dBRa, or 0.0003%.
Diego Estan
Electronics Measurement Specialist