Details

Parent Category: Products

Category: Headphone Amplifier Measurements

Created: 01 August 2019

Reviewed on: SoundStage! Solo, June 2019

I measured the Euterpe using a Clio 10 FW audio analyzer and a Neutrik NL-1 Minilyzer. Note that my focus with these tests is on measurements that confirm these devices’ basic functionality. I used the analog inputs; unfortunately, I’m currently unable to interface Clio’s coax digital output to USB-only DACs.

This chart shows the Euterpe’s frequency response with 1mW output into 32-, 250- and 600-ohm loads. The impedance switch on the amp was set to L for the 32-ohm load, and H for the 250- and 600-ohm loads. Into 32 ohms, response measures -8.24dB at 20Hz, -1.9dB at 20kHz, and -20.76dB at 75kHz. Into 250 ohms, the numbers are -7.74dB, -0.67dB, and -14.25dB, respectively. Into 600 ohms, the numbers are -9.91dB, -0.027dB, and -11.01dB, respectively. As you can see, the response curve basically shifts higher in frequency into higher-impedance loads, but in any case, this is an extreme amount of bass roll-off, and a substantial amount of treble roll-off.

I don’t normally include this chart because in most headphone amps, the channels are so closely matched that the difference isn’t worth noting. This difference here is, though. The right channel (measured into 32 ohms) is 0.36dB higher in level at 1kHz than the left channel is. Although it’s hard to see without normalizing the two curves at a certain frequency, you can see that the right channel’s frequency response is basically shifted to higher frequencies.

This is another chart I don’t usually show, but I thought it important to show the effects that the Euterpe’s high output impedance will have on the sounds of a couple of different headphones, so I compared the frequency response of two headphones driven by the Euterpe and by the Musical Fidelity V-CAN (output impedance 5 ohms). The lower traces show the response with the Audeze LCD-Xes, a planar-magnetic headphone that has a largely resistive impedance that makes it relatively insensitive to headphone amp output impedance. Still, the Euterpe’s output impedance (with the impedance switch set to L) is enough to reduce the LCD-Xes’ bass by 2.35dB at 50Hz. With the Beyerdynamic Amiron Homes, a dynamic-driver design, the effect is more pronounced -- the Euterpe reduces the Amiron Homes’ bass by 2.73dB at 50Hz, and also tilts the treble up by about 0.84dB. Bottom line: This amp is not neutral, and it will change the sound of your headphones relative to what you’d hear with most other headphone amps, especially ones with a low output impedance.

This chart shows the output of the Euterpe vs. total harmonic distortion (THD) into 32-, 250- and 600-ohm loads. Rated power is 0.9W, into an unspecified load at unspecified distortion at an unspecified frequency. Into 32 ohms, the lowest distortion I measured, at 0.01W, is 0.5%; the amp breaks my 1% THD max at 0.038W, and at the rated 0.9W max output, THD is 5.49%. Into 250 ohms, THD at 0.01W was 0.52%; output at 1% THD is 0.035W, and THD at the rated 0.9W is 5.89%. Surprisingly, the performance at 600 ohms easily bests the performance into lower-impedance loads -- output at 0.5% THD is 0.042W, and it’s 0.165W at 1% THD. At the rated 0.9W, THD measures 2.43%.

Here you can see the harmonic distortion spectrum and noise floor of the Euterpe, referenced to 3.1Vrms (0.3W) output at 600Hz into 32 ohms. This is a classic profile of the distortion of a single-ended tube amp, with the second-order distortion predominant. Because second-order harmonic distortion adds a harmonic precisely one octave above the fundamental, it’s less sonically offensive than third- or fifth-order harmonic distortion.

Output impedance at 1kHz measures 51 ohms with the impedance switch set to L, and 350 ohms with the switch set to H. This is extremely high output impedance relative to what I’m used to measuring; with any headphones that exhibit a significant impedance swing (such as earphones with balanced-armature drivers, and large over-ear headphones with dynamic drivers), the amp’s output impedance will interact with the reactance of the headphones or earphones to change the frequency response.

This is an amp with audible frequency response errors and high distortion. There are some audio writers who consider these idiosyncrasies a badge of honor, but I’m not one of them.

. . . Brent Butterworth
brentb@soundstagenetwork.com

Details

Parent Category: Products

Category: Headphone Amplifier Measurements

Created: 20 April 2019

Reviewed on: SoundStage! Solo, April 2019

I measured the Liquid Platinum using a Clio 10 FW audio analyzer and a Neutrik NL-1 Minilyzer. Note that my focus with these tests is on measurements that confirm these devices’ basic functionality. Because I didn’t have the necessary four-pin XLR adapter that would allow me to measure the balanced output, the measurements below are all with the amp in single-ended mode (using the 1/4” TRS headphone output). I have the parts on order to build the adapter and hope to add those results later. Meanwhile, I was able to measure the frequency response of the HiFiMan HE6se headphones from the balanced and unbalanced outputs, and the results were identical.

This chart shows the Liquid Platinum’s frequency response with 1mW output into 32-, 250-, and 600-ohm loads. Into 32 ohms, the response measures -0.014dB at 20Hz, -0.041dB at 20kHz, and -0.276dB at 75kHz. Into 250 ohms, the numbers are -0.015dB, -0.031dB, and -0.307dB, respectively. Into 600 ohms, the numbers are -0.015dB, -0.027dB, and -0.232dB, respectively. These are very good results.

This chart shows the single-ended output of the Liquid Platinum vs. total harmonic distortion (THD) into 32-, 250-, and 600-ohm loads. Note that Monoprice’s power ratings are specified at 33, 56, 150, and 300 ohms, so my measurements are not directly comparable, but Monoprice’s specs seem well in line with my results. Output into 32 ohms is 1.71W at 0.5% THD and 1.85W at 1% THD. (Monoprice’s most comparable rating is 1.78W into 33 ohms, THD unspecified.) Output into 250 ohms is 275mW at 0.5% THD and 289mW at 1% THD. (Monoprice’s most comparable rating is 230mW into 300 ohms, THD unspecified.) Output into 600 ohms is 117mW at 0.5% THD and 122mW at 1% THD.

Here you can see the harmonic distortion spectrum and noise floor of the Liquid Platinum, referenced to 1.5Vrms (1W) output at 600Hz into 32 ohms. Distortion is low, and no particular distortion harmonic dominates the spectrum; the first several harmonics are down in the -72dBFS range (plus or minus a couple of dB) relative to the level of the fundamental tone. (For reference, -70dBFS equates to 0.03% harmonic distortion.) We can see some 60-cycle AC hum and its harmonics, typically in the range of -70dBFS. The noise floor of the amp at this level is down around -92dBFS, which is pretty good for a device with tubes in the signal chain.

I measured output impedance of the 1/4” headphone jack at under 0.3 ohm at 1kHz, which is about as low as I can measure with my voltage divider; Monoprice rates it at 0.07 ohm. Either way, it’s easily low enough that the output impedance won’t react significantly with the reactance of the headphones, and thus won’t change their frequency response.

. . . Brent Butterworth
brentb@soundstagenetwork.com

Details

Parent Category: Products

Category: Headphone Amplifier Measurements

Created: 20 April 2019

Reviewed on: SoundStage! Solo, June 2019

I measured the Monoprice 24459 using a Clio 10 FW audio analyzer and a Neutrik NL-1 Minilyzer. I measured only the unbalanced output; for some reason I couldn’t figure out, the amp always went into protection mode when I connected the balanced output into a load resistor. Note that my focus with these tests is on measurements that confirm these devices’ basic functionality.

This chart shows the Monolith 24459’s frequency response with 1mW output into a 32-ohm load using the coaxial digital input. (Measurements with 250- and 600-ohm loads produced effectively identical results.) With the Normal digital-to-analog (DAC) filter, response measured -0.057dB at 20Hz, -0.227dB at 20kHz, and -1.470dB at 40kHz. With the Slow1 filter, the numbers were -0.057dB, -0.307dB, and -6.007dB, respectively. With the Slow2 filter, the numbers were -0.060dB, -0.975dB, and -4.307dB, respectively. These measurements were taken with a 192kHz digital signal, which the coax input accepts, but the digital circuitry is brick-wall filtered at about 40kHz (consistent with Monoprice’s published frequency response), so the effective resolution is actually 96kHz. Note the +1.2dB ringing of the Normal filter at 37kHz. From a technical standpoint, this isn’t impressive, but it won’t be audible. The ringing nearly disappears with the Slow1 and Slow2 filters.

This chart shows the effect of the two different analog-to-digital converter’s filter settings on the frequency response. Both were measured with 1mW output into a 32-ohm load using the unbalanced analog input, with the DAC filter set to Normal. (Measurements with 250- and 600-ohm loads produced effectively identical results.) With the Normal analog-to-digital (ADC) filter, response measured -0.057dB at 20Hz, -0.067dB at 10kHz, and -0.344dB at 30kHz. With the Slow1 filter, the numbers were -0.010dB, -0.139dB, and -0.922dB, respectively. Thus, the difference between the two filters might be just barely audible. (I cite the response here at 10kHz and 30kHz instead of my usual 20kHz and 40kHz because of the slightly non-smooth characteristics of the response curves.)

This chart shows the unbalanced output of the Monoprice 24459 vs. total harmonic distortion (THD) into 32-, 250-, and 600-ohm loads. Note that Monoprice’s power ratings are specified at 16, 32, 150, 300, and 600 ohms, so some of my measurements are not directly comparable. Output into 32 ohms was 1420mW at 0.5% THD and 1475mW at 1% THD. Output into 250 ohms was 183mW at 0.5% THD and 190mW at 1% THD. Output into 600 ohms was 77mW at 0.5% THD and 79mW at 1% THD. (Monoprice’s ratings are 1360mW into 32 ohms, 150mW into 300 ohms, and 73mW into 600 ohms, all with THD unspecified). These are very high numbers for a headphone amp, indicating that the Monolith 24459 should have no problem driving any headphones currently available.

Here you can see the harmonic distortion spectrum and noise floor of the Monolith 24459, referenced to 6.295Vrms (1.24W) output at 600Hz into 32 ohms. (I used this odd output number as a reference because, as best I can tell, the amplifier stage just barely starts to distort before the unit’s analog-to-digital converter stage clips. Any higher and the distortion becomes very high; any lower and there’s not enough distortion to see the harmonic content.) Harmonic distortion is predominantly odd-order, which is much more audible than even-order distortion, but with the 3rd harmonic at -78.2dBFS and the 5th harmonic at -79.4dBFS (both just slightly over 0.01% distortion), and the distortion occurring only at an extremely high output level, I think the chances of any listener actually hearing this are zero. Note also that the noise floor was generally at about -120dBFS. This is excellent performance.

I measured the output impedance of the unbalanced headphone jack at less than 0.5 ohm, which is as low as I could measure without triggering the amp’s protection circuit. In my opinion, an output impedance of less than 1 ohm is a good standard for headphone amps because it prevents the headphone amp from significantly interacting with the headphones’ impedance in a way that alters the headphones’ frequency response.

. . . Brent Butterworth
brentb@soundstagenetwork.com

Details

Parent Category: Products

Category: Headphone Amplifier Measurements

Created: 10 April 2019

Reviewed on: SoundStage! Solo, April 2019

I measured the Schiit Audio Fulla 2 using a Clio 10 FW audio analyzer and a Neutrik NL-1 Minilyzer. I used the Fulla 2’s analog input for all these measurements, because I haven’t yet found a way to get digital test signals from the Clio 10 FW to USB DACs. Note that my focus with these tests is on measurements that confirm these devices’ basic functionality.

This chart shows the Fulla 2’s frequency response with 1mW output into 32-ohm and 600-ohm loads. (Frequency response at 250 ohms is not shown because it almost perfectly overlapped with the response at 32 ohms.) Into 32 ohms, the response measures -0.011dB at 20Hz, -0.031dB at 20kHz, and -0.085dB at 75kHz. Into 600 ohms, the numbers are -0.009dB, -0.044dB, and -0.186dB, respectively. These are excellent results, comparable to those of a good high-end analog preamp.

This chart shows the output of the Fulla 2 vs. total harmonic distortion (THD) into 32-, 250- and 600-ohm loads. Note that Schiit’s power ratings are specified at 16, 50, 300, and 600 ohms, so some of my measurements are not directly comparable. Output into 32 ohms is 320mW at 0.5% THD and 340mW at 1% THD (Schiit’s rating is 360mW into 32 ohms, THD unspecified). Output into 250 ohms is 50mW at 0.5% THD and 51mW at 1% THD. Output into 600 ohms is 21mW at 0.5% THD and 22mW at 1% THD. These numbers are all very impressive for a $99 DAC-headphone amp.

Here you can see the harmonic distortion spectrum and noise floor of the Fulla 2, referenced to 1V RMS output at 600Hz into 32 ohms. Distortion is very low, with the second harmonic slightly higher in level than the third; I’d say this would make the Fulla 2 sound “tubey” if the distortion at this output level and load were high enough for you to hear, but that second harmonic is at -79dB. You can also see that the noise floor of the amp is way down around -110dB.

I measured output impedance of the headphone jack at 3.2 ohms at 1kHz; Schiit rates it at 0.5 ohm. Note that this measurement, made with a potentiometer used as a voltage divider, is not super-accurate, and any output impedance in the low single digits is low enough not to react significantly with the reactance of the headphones, and thus won’t change their frequency response.

. . . Brent Butterworth
brentb@soundstagenetwork.com

Details

Parent Category: Products

Category: Headphone Amplifier Measurements

Created: 10 January 2019

Reviewed on: SoundStage! Solo, January 2019

I measured the iFi Audio xCAN using a Clio 10 FW audio analyzer and a Neutrik NL-1 Minilyzer. Except as noted, I used the xCAN’s unbalanced analog input and unbalanced analog output, because I don’t yet have an adapter for 2.5mm balanced outputs I can use for measurements. Note that my focus with these tests is on measurements that confirm these devices’ basic functionality, and that gauge the efficacy of any special features and functions that might be measurable.

This chart shows the xCAN’s frequency response with all processing off, and with XBass II engaged in its three different modes (Bass, Presence, and Bass+Presence), with 1mW output into a 32-ohm load. With processing off, the response measures -0.14dB at 20Hz and -0.19dB at 20kHz. Bass mode boosts response by 9.96dB at 20Hz. Presence mode boosts response in a 4.12dB peak centered at 1288Hz. Frequency response did not change in 3D+ mode, and also did not change with 250-ohm and 600-ohm loads.

This chart shows the unbalanced output of the xCAN vs. total harmonic distortion (THD) into 32-, 250- and 600-ohm loads. Note that iFi’s power ratings are specified at 16, 50, 300 and 600 ohms, so some of my measurements are not directly comparable. Output into 32 ohms is 320mW at 0.5% THD and 336mW at 1% THD (iFi’s rating, in S-balanced/unbalanced mode, is 380mW into 32 ohms, THD unspecified). Output into 250 ohms is 46mW at 0.5% THD and 49mW at 1% THD. Output into 600 ohms is 20mW at 0.5% THD and 19mW at 1% THD.

Here you can see the harmonic distortion spectrum and noise floor of the xCAN, referenced to 3Vrms output at 600Hz into 32 ohms. The third harmonic at 1.8kHz is slightly more predominant than the second harmonic, which will sound a little more objectionable than an amp (like a typical tube amp) with predominantly second-harmonic distortion, but if you actually dare to listen at 3Vrms (280mW into 32 ohms), the distortion from the headphones will likely be far louder than the distortion from the amp.

I measured the unbalanced output impedance at 1.2 ohms at 1kHz; iFi rates impedance at <2 ohms for balanced and <1 ohm for unbalanced output. Regardless, the output impedance is low enough not to react significantly with the reactance of the headphones, and thus won’t change their frequency response.

. . . Brent Butterworth
brentb@soundstagenetwork.com

Details

Parent Category: Products

Category: Headphone Amplifier Measurements

Created: 01 October 2018

Reviewed on: SoundStage! Solo, October 2018

I measured the iFi Audio xDSD using a Clio 10 FW audio analyzer and a Neutrik NL-1 Minilyzer. For all of these tests, I used the xDSD’s coaxial digital input. Note that this is the first DAC-headphone amp I’ve measured for SoundStage! Solo; I’ve decided to focus my efforts on tests that confirm such devices’ basic functionality, and that gauge the efficacy of any special features and functions that might be measurable.

This chart shows the xDSD’s frequency response in its Listen and Measure modes, and with XBass+ engaged, with a 24-bit/192kHz S/PDIF signal and the xDSD set for 1mW output into a 32-ohm load. The response in both modes measured -0.16dB at 20Hz and -0.26dB at 20kHz. Listen mode actually measured slightly better here, with less rolloff above 65kHz; apparently, the switch is mislabeled. The bass boost in XBass+ mode was 6.48dB at 20Hz.

This chart shows the xDSD’s frequency response in Listen and Measure modes, and with XBass+ engaged, with a 16/48 S/PDIF signal and the xDSD set for 1mW output into a 32-ohm load. The treble response at 20kHz in Measure mode is -1.91dB, and in Listen mode -0.32dB. Definitely, the switch is mislabeled. According to the xDSD manual, the Listen filter is “transient-optimized minimum phase” and the Measure filter is “frequency response optimized,” but a filter with -1.91dB rolloff at 20kHz is certainly not “frequency response optimized.”

This chart shows the output of the xDSD vs. its total harmonic distortion (THD) into loads of 32, 250, and 600 ohms. Although iFi specifies the xDSD’s power output into 16, 50, 300, and 600 ohms, which renders most of my measurements not directly comparable, those measurements do suggest that iFi’s specs are on the mark. The xDSD’s output into 32 ohms is 291mW at 0.5% THD and 304mW at 1% THD; into 250 ohms, the output is 53mW at 0.5% THD and 54mW at 1% THD; and into 600 ohms, the xDSD puts out 22mW at 0.5% THD and 23mW at 1% THD.

Here you can see the xDSD’s spectrum of harmonic distortion and noise floor when driven by a 24/192 S/PDIF signal and referenced to 1.5V RMS output at 600Hz. Note that the distortion profile of the Measure and Listen modes is effectively the same.

I measured the xDSD’s output impedance as 0.8 ohm at 1kHz, which confirms iFi’s rating of <1 ohm. I prefer a headphone amp’s output impedance to be 1 ohm or less; the output impedance will then not react significantly with the reactance of the headphones, and thus won’t affect the ’phones’ frequency response.

. . . Brent Butterworth
brentb@soundstagenetwork.com

Details

Parent Category: Products

Category: Amplifier Measurements

Created: 01 April 2026

Link: reviewed by Doug Schneider on SoundStage! Hi-Fi on April 1, 2026

General Information

All measurements taken using an Audio Precision APx555 B Series analyzer.

The Simaudio Moon 371 was conditioned for 1 hour at 1/8th full rated power (~15W into 8 ohms) before any measurements were taken. All measurements were taken with both channels driven, using a 120V/20A dedicated circuit, unless otherwise stated.

The 371 offers two analog line-level inputs (RCA and XLR), four digital inputs (two coaxial, one optical, one HDMI), a phono input (RCA) configurable for both moving-magnet (MM) or moving-coil (MC), an ethernet connection for streaming, line-level pre-outs (RCA), and a pair of speaker-level outputs. Also on offer is a headphone output over 1/4″ TRS connector. For the purposes of these measurements, the following inputs were evaluated: coaxial digital (RCA), and the analog line-level (XLR) and phono (RCA) inputs. Comparisons were made between the balanced and unbalanced inputs, and no appreciable differences were seen (FFTs for both are included in this report).

Most measurements were made with a 2Vrms line-level analog input, 0dBFS digital input, and 5/0.5mVrms phono inputs (MM/MC). The signal-to-noise (SNR) measurements were made with the same input signal values but with the volume set to achieve the rated output power of 100W. For comparison, on the analog input, an SNR measurement was also made with the volume at maximum, but the input voltage reduced to achieve the same 100W output. The 371 offers a range of gain settings, though the default 6dB was maintained for these measurements.

Based on the accuracy and repeatability at various volume levels of the left/right channel matching (see table below), the 371 volume control is likely digitally controlled but operating in the analog domain. The volume range is from -41dB to +37.5dB (line-level inputs, speaker level outputs). Below volume level 20, steps are 1dB each; above 20, volume step size is 0.5dB.

Our typical input bandwidth filter setting of 10Hz-22.4kHz was applied for all measurements, except for frequency response (DC to 1MHz) and for FFTs and frequency sweeps (10Hz to 90kHz).

Volume-control accuracy (measured at speaker outputs): left-right channel tracking

Published specifications vs. our primary measurements

The table below summarizes the measurements published by Simaudio for the Moon 371 compared directly against our own. The published specifications are sourced from Simaudio’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, a measurement input bandwidth of 10Hz to 22.4kHz, and the worst-case measured result between the left and right channels.

Our primary measurements revealed the following using the analog input (unless specified, assume a 1kHz sinewave at 2Vrms or 0dBFS, 10W output, 8-ohm loading, 10Hz to 22.4kHz bandwidth):

For the continuous dynamic power test, the Moon 371 was able to sustain 263W (1% THD) into 4 ohms using an 80Hz tone for 500ms, alternating with a signal at -10dB of the peak (10W) for 5 seconds, for 5 continuous minutes without inducing a fault or the initiation of a protective circuit. This test is meant to simulate sporadic dynamic bass peaks in music and movies. During the test, the 371 was very hot to the touch.

Our primary measurements revealed the following using the phono-level input, MM configuration (unless specified, assume a 1kHz 5mVrms sinewave input, 10W output, 8-ohm loading, 10Hz to 22.4kHz bandwidth):

Our primary measurements revealed the following using the phono-level input, MC configuration (unless specified, assume a 1kHz 0.5mVrms sinewave input, 10W output, 8-ohm loading, 10Hz to 22.4kHz bandwidth):

Our primary measurements revealed the following using the analog input at the headphone output (unless specified, assume a 1kHz sinewave, 2Vrms input/output, 300 ohms loading, 10Hz to 22.4kHz bandwidth):

Frequency response (8-ohm loading, line-level input)

In our measured frequency-response chart above, the 371 is essentially perfectly flat within the audioband (20Hz to 20kHz). At the extremes the 371 is 0dB at 5Hz and -1.8dB at 200kHz. The 371 appears to be DC-coupled, as we see no attenuation even at 5Hz.  In the graph above and most of the graphs below, only a single trace may be visible. This is because the left channel (blue or purple trace) is performing identically to the right channel (red or green trace), and so they perfectly overlap, indicating that the two channels are ideally matched.

Phase response (8-ohm loading, line-level input)

Above is the phase response plot from 20Hz to 20kHz for the analog input. The 371 does not invert polarity and exhibits about -15 degrees of phase shift at 20kHz.

Frequency response (8-ohm loading, MM phono input)

The chart above shows the frequency response for the phono input configured for MM. What is shown is the deviation from the RIAA curve, where the input signal sweep is EQ’d with an inverted RIAA curve supplied by Audio Precision (i.e., zero deviation would yield a flat line at 0dB). We see an exceptionally flat response from 20Hz to 20kHz, with essentially zero channel-to-channel deviations. At the extremes, the 371 is -1dB at 5Hz and -0.5dB at 90kHz. This is an exceptional example of RIAA EQ implementation, especially when we consider it’s implemented in the analog domain.

Phase response (8-ohm loading, MM phono input)

Above is the phase response plot from 20Hz to 20kHz for the phono input configured for MM, measured across the speaker outputs at 10W into 8 ohms. The 371 does not invert polarity. For the phono input, since the RIAA equalization curve must be implemented, which ranges from +19.9dB (20Hz) to -32.6dB (90kHz), phase shift at the output is inevitable. Here we find a worst case of about +40 degrees at 20Hz and -10 degrees at 200Hz, and +60 degrees at 20kHz.

Frequency response (8-ohm loading, MC phono input)

The chart above shows the frequency response for the phono input configured for MC. We see essentially the same result as with the MM configuration.

Phase response (8-ohm loading, MC phono input)

Above is the phase response plot from 20Hz to 20kHz for the phono input configured for MC, measured across the speaker outputs at 10W into 8 ohms. The 371 does not invert polarity. For the phono input, since the RIAA equalization curve must be implemented, which ranges from +19.9dB (20Hz) to -32.6dB (90kHz), phase shift at the output is inevitable. Here we find a worst case of about -60 degrees at 200Hz and -100 degrees at 20kHz.

Frequency response vs. input type (8-ohm loading)

The chart above shows the 371’s frequency response as a function of input type. The green traces are the same analog input data from the previous graph (limited to 80kHz). The blue/red traces are for a 16-bit/44.1kHz dithered digital input signal from 5Hz to 22kHz, the purple/green traces are for a 24/96 dithered digital input signal from 5Hz to 48kHz, and finally orange/pink is 24/192 from 5Hz to 96kHz. The behavior at low frequencies is the same across all input types: 0dB at 5Hz. The behavior at high frequencies for all digital input types is typical, sharp filtering around half the sample rate. The -3dB points are at roughly 22kHz, 46kHz, and 91kHz.

Digital linearity (16/44.1 and 24/96 data)

The chart above shows the results of a linearity test for the coaxial digital input for both 16/44.1 (blue/red) and 24/96 (purple/green) input data, measured at the line-level pre-outs of the 371. For this test, the digital input is swept with a dithered 1kHz input signal from -120dBFS to 0dBFS and the output is analyzed by the APx555. The ideal response would be a straight flat line at 0dB. Both digital input types are essentially perfect from -110 dBFS down to 0dBFS. At -120dBFS, the 16/44.1 input data overshot the ideal output signal amplitude by 2dB (right channel only), while the 24/96 data remained perfect. The sweep was extended down to -140dBFS to . . .

. . . verify the performance of the 24/96 data. The traces show only a +2/3dB overshoot at -140dBFS. A superb result.

Impulse response (24/44.1 data)

The graph above shows the impulse response for a looped 24/44.1 test file that moves from digital silence to full 0dBFS (all “1”s) for one sample period then back to digital silence, measured at the line-level pre-outs of the 371. We can see that the 371 utilizes a reconstruction filter with no pre-ringing but obvious post-ringing.

J-Test (coaxial)

The plot above shows the results of the J-Test test for the coaxial digital input measured at the line-level pre-outs of the 371. J-Test was developed by Julian Dunn the 1990s. It is a test signal: specifically a -3dBFS undithered 12kHz squarewave sampled (in this case) at 48kHz (24bit). Since even the first odd harmonic (i.e., 36kHz) of the 12kHz squarewave is removed by the bandwidth limitation of the sampling rate, we are left with a 12kHz sinewave. In addition, an undithered 250Hz squarewave at -144dBFS is mixed with the signal. This test file causes the 22 least-significant bits to constantly toggle, which produces strong jitter spectral components at the 250Hz rate and its odd harmonics. The test file shows how susceptible the DAC and delivery interface are to jitter, which would manifest as peaks above the noise floor at 500Hz intervals (e.g., 250Hz, 750Hz, 1250Hz, etc.). Note that the alternating peaks are in the test file itself, but at levels of -144dBFS and below.  The test file can also be used in conjunction with artificially injected sinewave jitter by the Audio Precision, to show how well the DAC rejects jitter.

We see only very small peaks in the audioband at -140dBFS and below. This is a very good J-Test result and an indication that the 371 DAC has good jitter immunity.

J-Test (optical)

The plot above shows the results of the J-Test test for the optical digital input measured at the line-level pre-outs of the 371. We see essentially the same result as with the coaxial input above.

J-Test (coaxial, 10ns jitter)

The chart above shows the results of the J-Test test for the coaxial digital input measured at the line-level output of the 371, with an additional 10ns of 2kHz sinewave jitter injected by the APx555. The telltale peaks at 10kHz and 12kHz cannot be seen. The optical input yielded the same result.

J-Test (coaxial, 100ns jitter)

The chart above shows the results of the J-Test test for the coaxial digital input measured at the line-level output of the 371, with an additional 100ns of 2kHz sinewave jitter injected by the APx555. The telltale peaks at 10kHz and 12kHz cannot be seen above the -160dBFS noise floor. Further evidence of the 371 DAC’s superb jitter rejection. The optical input produced a similar result.

Wideband FFT spectrum of white noise and 19.1kHz sine-wave tone (coaxial input)

The plot above shows a fast Fourier transform (FFT) of the 371’s line-level pre-outs with white noise at -4 dBFS (blue/red) and a 19.1 kHz sinewave at 0dBFS fed to the coaxial digital input, sampled at 16/44.1. The sharp roll-off above 20kHz in the white-noise spectrum shows the implementation of a brickwall-type reconstruction filter. There are no aliased image peaks in the audioband above the -135dBrA noise floor. The main 25kHz alias peak is at -100dBrA. The second, third, and fourth distortion harmonics (38.2, 57.3, 76.4kHz) of the 19.1kHz tone are at roughly the same level.

RMS level vs. frequency vs. load impedance (1W, left channel only)

The chart above shows RMS level (relative to 0dBrA, which is 1W into 8ohms or 2.83Vrms) as a function of frequency, for the analog input swept from 10Hz to 100kHz. The blue plot is into an 8-ohm load, the purple is into a 4-ohm load, the pink plot is an actual speaker (Focal Chora 806, measurements can be found here), and the cyan plot is no load connected. The chart below . . .

. . . is the same but zoomed in to highlight differences. Zoomed in, we can see a maximum deviation within the audioband of roughly 0.04dB from 4 ohms to no load, which is an indication of a very high damping factor, or very low output impedance. The maximum variation in RMS level into a real speaker was about the same.

THD ratio (unweighted) vs. frequency vs. output power

The chart above shows THD ratios at the output into 8 ohms as a function of frequency (20Hz to 6kHz) for a sinewave stimulus at the analog input. The blue and red plots are for the left and right channels at 1W output into 8 ohms, purple/green at 10W, and pink/orange at roughly 100W. The power was varied using the volume control. The 1W data exhibited the highest THD values, with values varying around 0.0007% at 20Hz, down to 0.0005% from 60Hz to 600Hz, then up to 0.01% at 20kHz. The 10W data varied from 0.0004% at 20Hz, down to 0.0003% from 30Hz to 600Hz, then up to 0.004% at 20kHz. At 100W, THD values varied from 0.0006% at 20Hz, down to 0.0002% from 200Hz to 500Hz, then up to 0.02% at 20kHz.

THD ratio (unweighted) vs. frequency at 10W (phono input, MM and MC)

The chart above shows THD ratios as a function of frequency plots for the phono input measured across an 8-ohm load at 10W. The blue/red traces are for the MM configuration, purple/green for MC. For this test, the input sweep is EQ’d with an inverted RIAA curve. The THD values for the MM configuration vary from around 0.04% at 20Hz, down to 0.0006-0.001% from 300Hz to 3kHz (right channel), then up to 0.005% at 20kHz. The left channel for the MM configuration yielded THD ratios roughly 10dB higher than the right channel between 300Hz and 3kHz. The THD values for the MC configuration vary from around 0.2% at 20Hz, down to 0.005-0.01% from 300Hz to 8kHz (right channel), then up to 0.05% at 20kHz. The left channel for the MC configuration yielded THD ratios roughly 10dB higher than the right channel between 300Hz and 6kHz, but 15dB lower at 20kHz.

THD ratio (unweighted) vs. output power at 1kHz into 4 and 8 ohms

The chart above shows THD ratios measured at the output of the 371 as a function of output power for the analog line level-input, for an 8-ohm load (blue/red for left/right) and a 4-ohm load (purple/green for left/right), with the volume set to maximum. The 8-ohm data ranges from 0.003% at 50mW, down to 0.0003% from 20W to the “knee,” at about 130W.  The 4-ohm data ranges from 0.005% at 50mW, down to 0.0006-0.0008% from 5W to the “knee,” at about 230W. The 1% THD marks were hit at 145W and 264W for the 8-ohm and 4-ohm loads respectively.

THD+N ratio (unweighted) vs. output power at 1kHz into 4 and 8 ohms

The chart above shows THD+N ratios measured at the output of the 371 as a function of output power for the analog line level-input, for an 8-ohm load (blue/red for left/right) and a 4-ohm load (purple/green for left/right), with the volume set to maximum. The 8-ohm data ranges from 0.03% at 50mW, down to 0.0007% at the “knee.”  The 4-ohm data ranges from 0.05% at 50mW, down to 0.001% at the “knee.”

THD ratio (unweighted) vs. frequency at 8, 4, and 2 ohms (left channel only)

The chart above shows THD ratios measured at the output of the 371 as a function of load (8/4/2 ohms) for a constant input voltage that yields 20W at the output into 8 ohms (and roughly 40W into 4 ohms and 80W into 2 ohms) for the analog line-level input. The 8-ohm load is the blue trace, the 4-ohm load the purple trace, and the 2-ohm load the pink trace. We find increasing levels of THD from 8 to 4 to 2 ohms, with about a 5-8dB increase when halving the load beyond 1kHz. Overall, even with a 2-ohm load at roughly 80W, THD values were fairly low, between 0.0002-0.0003% from 20Hz to 50Hz, then up to 0.02% at 20kHz.

THD ratio (unweighted) vs. frequency into 8 ohms and real speakers (left channel only)

The chart above shows THD ratios measured at the output of the 371 as a function of frequency into an 8-ohm load and two different speakers for a constant output voltage of 2.83Vrms (1W into 8 ohms) for the analog line-level input. The 8-ohm load is the blue trace, the purple plot is a two-way speaker (Focal Chora 806, measurements can be found here), and the pink plot is a three-way speaker (Paradigm Founder Series 100F, measurements can be found here). Generally, THD ratios into the real speakers were higher than those measured across the resistive dummy load. The differences ranged from 0.03% at 20Hz for the two-way speaker versus 0.0007% for the resistive load, and 0.015% at 20kHz into the three-way speaker versus 0.01% for the resistive load. Between the important frequencies of 300Hz to 6kHz, all three THD traces were very close, between 0.0005% and 0.003% mark, with the speaker THD ratios outperforming the resistive load by a few dB.

IMD ratio (CCIF) vs. frequency into 8 ohms and real speakers (left channel only)

The chart above shows intermodulation distortion (IMD) ratios measured at the output of the 371 as a function of frequency into an 8-ohm load and two different speakers for a constant output voltage of 2.83Vrms (1W into 8 ohms) for the analog line-level input. Here the CCIF IMD method was used, where the primary frequency is swept from 20kHz (F1) down to 2.5kHz, and the secondary frequency (F2) is always 1kHz lower than the primary, with a 1:1 ratio. The CCIF IMD analysis results are the sum of the second (F1-F2 or 1kHz) and third modulation products (F1+1kHz, F2-1kHz). The 8-ohm load is the blue trace, the purple plot is a two-way speaker (Focal Chora 806, measurements can be found here), and the pink plot is a three-way speaker (Paradigm Founder Series 100F, measurements can be found here). We find that all three IMD traces are close to one another, with the three-way speaker yielding 5dB higher results in the 10-20kHz range. Most of the IMD results are hovering between the 0.001% and 0.01% level.

IMD ratio (SMPTE) vs. frequency into 8 ohms and real speakers (left channel only)

The chart above shows IMD ratios measured at the output of the 371 as a function of frequency into an 8-ohm load and two different speakers for a constant output voltage of 2.83Vrms (1W into 8 ohms) for the analog line-level input. Here, the SMPTE IMD method was used, where the primary frequency (F1) is swept from 250Hz down to 40Hz, and the secondary frequency (F2) is held at 7kHz with a 4:1 ratio. The SMPTE IMD analysis results consider the second (F2 ± F1) through the fifth (F2 ± 4xF1) modulation products. The 8-ohm load is the blue trace, the purple plot is a two-way speaker (Focal Chora 806, measurements can be found here), and the pink plot is a three-way speaker (Paradigm Founder Series 100F, measurements can be found here). We find very similar IMD ratios into all three loads, between 0.004 and 0.005% across the sweep.

FFT spectrum – 1kHz (line-level input, XLR)

Shown above is the fast Fourier transform (FFT) for a 1kHz input sinewave stimulus, measured at the output across an 8-ohm load at 10W for the analog balanced line-level input. We see that the signal’s second (2kHz) and third (3kHz) harmonics dominate at a low -110dBrA, or 0.0003%. The remaining signal harmonics are at and below -120dBrA, or 0.0001%. On the left side of the main signal peak, we see only minor power-supply noise-related peaks from the left channel only at -120dBrA to -130dBrA, or 0.0001% to 0.00003%, at 60/300/420 Hz.

FFT spectrum – 1kHz (line-level input, RCA)

Shown above is the fast Fourier transform (FFT) for a 1kHz input sinewave stimulus, measured at the output across an 8-ohm load at 10W for the analog unbalanced line-level input. We see that the signal’s second (2kHz) and third (3kHz) harmonics dominate and are higher than the balanced input FFT above at -100dBrA, or 0.001%. The remaining harmonic and noise peaks are very similar to the balanced input FFT above, other than spurious high-frequency peaks at 16-17kHz at -120dBrA, or 0.0001%.

FFT spectrum – 1kHz (MM phono input)

Shown above is the fast Fourier transform (FFT) for a 1kHz input sinewave stimulus, measured at the output across an 8-ohm load at 10W for the phono input configured for MM. We see that the signal’s second (2kHz) and fourth (4kHz) harmonics dominate at -90dBrA and -105dBrA respectively, or 0.003% and 0.0006%. These peaks are for the left channel, the right channel peaks are up to 20dB lower in amplitude. On the left side of the main signal peak, we see power-supply noise-related peaks at 60Hz  at -80dBrA, or 0.001%, followed by smaller peaks (right channel only) at 120/180/240Hz at -90dBrA to -100dBrA, or 0.003% to 0.001%.

FFT spectrum – 1kHz (MC phono input)

Shown above is the fast Fourier transform (FFT) for a 1kHz input sinewave stimulus, measured at the output across an 8-ohm load at 10W for the phono input configured for MC. We see that the signal’s second (2kHz) and fourth (4kHz) harmonics dominate at -70dBrA and -85dBrA respectively, or 0.03% and 0.006%. These peaks are for the left channel, the right channel peaks are up to 30dB lower in amplitude. On the left side of the main signal peak, we see power-supply noise-related peaks at 60Hz at -60dBrA, or 0.01%, followed by smaller peaks (right channel only) at 120/180/240/360Hz at -70dBrA to -85dBrA, or 0.03% to 0.006%.

FFT spectrum – 1kHz (digital input, 16/44.1 data at 0dBFS)

Shown above is the fast Fourier transform (FFT) for a 1kHz input sinewave stimulus, measured at the output across an 8-ohm load at 10W for the coaxial digital input, sampled at 16/44.1. We see that the signal’s second (2kHz) and third (3kHz) harmonics dominate at -110dBrA and -100dBrA, respectively, or 0.0003% and 0.001%. The remaining signal harmonics are below -110dBrA, or 0.0003%. On the left side of the main signal peak, we see only minor power-supply noise-related peaks from the left channel only at -120dBrA to -130dBrA, or 0.0001% to 0.00003%, at 60/300/420 Hz.

FFT spectrum – 1kHz (digital input, 24/96 data at 0dBFS)

Shown above is the fast Fourier transform (FFT) for a 1kHz input sinewave stimulus, measured at the output across an 8-ohm load at 10W for the coaxial digital input, sampled at 24/96. We see effectively the same FFT as the 16/44.1 result above, but with a lower noise floor due to the increased bit depth.

FFT spectrum – 1kHz (digital input, 16/44.1 data at -90dBFS)

Shown above is the FFT for a 1kHz -90dBFS dithered 16/44.1 input sinewave stimulus at the coaxial digital input, measured at the output across an 8-ohm load. We see the 1kHz primary signal peak, at the correct amplitude, no signal harmonics above the noise floor, and power-supply noise-related peaks from the left channel only at -125dBrA to -130dBrA, or 0.00006% to 0.00003%, at 60/180/300Hz.

FFT spectrum – 1kHz (digital input, 24/96 data at -90dBFS)

Shown above is the FFT for a 1kHz -90dBFS dithered 24/96 input sinewave stimulus at the coaxial digital input, measured at the output across an 8-ohm load. We see the 1kHz primary signal peak, at the correct amplitude, no signal harmonics above the noise floor, and power-supply noise-related peaks from the left channel only at -125dBrA to -130dBrA, or 0.00006% to 0.00003%, at 60/180/300Hz.

FFT spectrum – 50Hz (line-level input)

Shown above is the FFT for a 50Hz input sinewave stimulus measured at the output across an 8-ohm load at 10W for the analog input. The X axis is zoomed in from 40 Hz to 1kHz, so that peaks from noise artifacts can be directly compared against peaks from the harmonics of the signal. The most predominant (non-signal) peaks are the signal third (150Hz) harmonic just below -110dBrA, or 0.0003%, and power-supply-related noise peaks from the left channel at -120dBrA to -130dBrA, or 0.0001% to 0.00003%.

FFT spectrum – 50Hz (MM phono input)

Shown above is the FFT for a 50Hz input sinewave stimulus measured at the output across an 8-ohm load at 10W for the phono input configured for MM. The X axis is zoomed in from 40 Hz to 1kHz, so that peaks from noise artifacts can be directly compared against peaks from the harmonics of the signal. The most predominant (non-signal) peak is the main power-supply-related noise peak at 60Hz at 80dBrA, or 0.001%. Signal harmonic peaks and subsequent power-supply-related noise peaks can be seen between -90dBrA and -110dBrA, or 0.003% and 0.0003%.

FFT spectrum – 50Hz (MC phono input)

Shown above is the FFT for a 50Hz input sinewave stimulus measured at the output across an 8-ohm load at 10W for the phono input configured for MC. The X axis is zoomed in from 40 Hz to 1kHz, so that peaks from noise artifacts can be directly compared against peaks from the harmonics of the signal. The most predominant (non-signal) peak is the main power-supply-related noise peak at 60Hz at 60dBrA, or 0.01%. Signal harmonic peaks and subsequent power-supply-related noise peaks can be seen (mostly from the right channel) between -70dBrA and -90dBrA, or 0.03% and 0.003%.

Intermodulation distortion FFT (18kHz + 19kHz summed stimulus, line-level input)

Shown above is an FFT of the intermodulation distortion (IMD) products for an 18kHz + 19kHz summed sinewave stimulus tone measured at the output across an 8-ohm load at 10W for the analog input. The input RMS values are set at -6.02dBrA so that, if summed for a mean frequency of 18.5kHz, would yield 10W (0dBrA) into 8 ohms at the output. We find that the second-order modulation product (i.e., the difference signal of 1kHz) is at -120dBrA, or 0.0001%, while the third-order modulation products, at 17kHz and 20kHz, are at -105dBrA, or 0.0006%.

Intermodulation distortion FFT (line-level input, APx 32 tone)

Shown above is the FFT of the speaker-level output of the 371 with the APx 32-tone signal applied to the analog input. The combined amplitude of the 32 tones is the 0dBrA reference, and corresponds to 10W into 8 ohms. The intermodulation products—i.e., the “grass” between the test tones—are distortion products from the amplifier and are at and below the very low -125dBrA, or 0.00006%, level.

Intermodulation distortion FFT (18kHz + 19kHz summed stimulus, coaxial digital input, 16/44.1)

Shown above is an FFT of the intermodulation distortion (IMD) products for an 18kHz + 19kHz summed sinewave stimulus tone measured at the output across an 8-ohm load at 10W for the digital coaxial input at 16/44.1. We find that the second-order modulation product (i.e., the difference signal of 1kHz) is at -120dBrA, or 0.0001%, while the third-order modulation products, at 17kHz and 20kHz, are at -100dBrA, or 0.001%.

Intermodulation distortion FFT (18kHz + 19kHz summed stimulus, coaxial digital input, 24/96)

Shown above is an FFT of the intermodulation distortion (IMD) products for an 18kHz + 19kHz summed sinewave stimulus tone measured at the output across an 8-ohm load at 10W for the digital coaxial input at 24/96. We find that the second-order modulation product (i.e., the difference signal of 1kHz) is at -120dBrA, or 0.0001%, while the third-order modulation products, at 17kHz and 20kHz, are at -100dBrA, or 0.001%.

Intermodulation distortion FFT (18kHz + 19kHz summed stimulus, MM phono input)

Shown above is an FFT of the intermodulation distortion (IMD) products for an 18kHz + 19kHz summed sinewave stimulus tone measured at the output across an 8-ohm load at 10W for the phono input configured for MM. The input RMS values are set at -6.02dBrA so that, if summed for a mean frequency of 18.5kHz, would yield 10W (0dBrA) into 8 ohms at the output. We find that the second-order modulation product (i.e., the difference signal of 1kHz) is at -100dBrA, or 0.001%, while the third-order modulation products, at 17kHz and 20kHz, are at roughly the same level.

Intermodulation distortion FFT (18kHz + 19kHz summed stimulus, MC phono input)

Shown above is an FFT of the intermodulation (IMD) products for an 18kHz + 19kHz summed sinewave stimulus tone measured at the output across an 8-ohm load at 10W for the phono input configured for MC. The input RMS values are set at -6.02dBrA so that, if summed for a mean frequency of 18.5kHz, would yield 10W (0dBrA) into 8 ohms at the output. We find that the second-order modulation product (i.e., the difference signal of 1kHz) is at -80dBrA (left channel only), or 0.01%, while the third-order modulation products, at 17kHz and 20kHz, are at -105dBrA, or 0.0006%.

Squarewave response (10kHz)

Above is the 10kHz squarewave response using the analog input, at roughly 10W into 8 ohms. Due to limitations inherent to the Audio Precision APx555 B Series analyzer, this graph should not be used to infer or extrapolate the 371’s slew-rate performance. Rather, it should be seen as a qualitative representation of its extended bandwidth. An ideal squarewave can be represented as the sum of a sinewave and an infinite series of its odd-order harmonics (e.g., 10kHz + 30kHz + 50kHz + 70kHz . . .). A limited bandwidth will show only the sum of the lower-order harmonics, which may result in noticeable undershoot and/or overshoot, and softening of the edges. Here we see a very clean result, with only very mild softening in the corners.

Damping factor vs. frequency (20Hz to 20kHz)

The final plot above is the damping factor as a function of frequency. Both channels show a relatively constant damping factor from 20Hz to 3kHz at roughly 425. The damping factor then dips to roughly 200 at 20kHz.

Diego Estan
Electronics Measurement Specialist

Details

Parent Category: Products

Category: Amplifier Measurements

Created: 15 March 2026

Link: reviewed by Dennis Burger on SoundStage! Simplifi on March 15, 2026

General Information

All measurements taken using an Audio Precision APx555 B Series analyzer.

The SA45 was conditioned for 1 hour at 1/8th full rated power (~20W into 8 ohms) before any measurements were taken. All measurements were taken with both channels driven, using a 120V/20A dedicated circuit, unless otherwise stated.

The SA45 offers four sets of line-level analog inputs (three single-ended RCA, one balanced XLR), two phono inputs (MM and MC over RCA), two digital coaxial inputs (RCA), two digital optical inputs (TosLink), left/right pre-outs and sub-outs (RCA and XLR), one set of speaker-level outputs, and one headphone output over 1/8″ TRS connector. Bluetooth, HDMI and streaming inputs are also offered. For the purposes of these measurements, the following inputs were evaluated: digital coaxial, analog (XLR) line-level and phono (MM and MC), as well as the headphone output.

Most measurements were made with a 4Vrms line-level balanced analog input and 0dBFS digital input. The phono inputs were 5mVrms (MM) and 0.5mVrms (MC). The signal-to-noise ratio (SNR) measurements were made with the default input signal values but with the volume set to achieve the achievable output power of 177W into 8 ohms. For comparison, on the line-level input, a SNR measurement was also made with the volume at maximum. The SA45 offers four digital filters for the digital inputs, labelled as follows in this report:

• Filter 1: Linear phase apodising (default)
• Filter 2: Linear phase slow roll-off
• Filter 3: Minimum phase slow roll-off
• Filter 4: Minimum phase

The SA45 also offers Dirac Live room correction. If no subwoofers are enabled and Dirac Live is off, the SA45 does not digitize incoming analog signals. For comparisons, frequency response and FFTs of analog inputs (line-level and phono) are shown in this report with the subwoofer setting toggled on and off (which engages and disengages the ADC at the input).

Based on the accuracy and randomness of the left/right volume channel matching (see table below), the SA45 volume control is digitally controlled but operating in the analog domain. The SA45 overall volume range is from -81dB to +35dB (line-level input, speaker output). It offers 3dB increments from position 0 to 10, 2dB increments from positions 11 to 20, 1dB from 21 to 50, 0.5dB from 51 to 85, and 0.25dB from 86 to 100. 

Our typical input bandwidth filter setting of 10Hz-22.4kHz was used for all measurements except FFTs and THD versus frequency, where a bandwidth of 10Hz-90kHz was used. Frequency response measurements utilize a DC to 1 MHz input bandwidth.

Volume-control accuracy (measured at speaker outputs): left-right channel tracking

Published specifications vs. our primary measurements

The table below summarizes the measurements published by Arcam for the SA45 compared directly against our own. The published specifications are sourced from Arcam’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 extended to 1MHz, assume, unless otherwise stated, 10W into 8 ohms and a measurement input bandwidth of 10Hz to 22.4kHz, and the worst-case measured result between the left and right channels.

Our primary measurements revealed the following using the line-level analog input and digital coaxial input (unless specified, assume a 1kHz sinewave at 4Vrms or 0dBFS, 10W output, 8-ohm loading, 10Hz to 22.4kHz bandwidth):

For the continuous dynamic power test, the SA45 was able to sustain 290W into 4 ohms (~4% THD) using an 80Hz tone for 500ms, alternating with a signal at -10dB of the peak (29W) for 5 seconds, for 5 continuous minutes without inducing a fault protection circuit. This test is meant to simulate sporadic dynamic bass peaks in music and movies. During the test, the top of the SA45 was slightly warm to the touch.

Our primary measurements revealed the following using the phono-level input, MM configuration (unless specified, assume a 1kHz 5mVrms sinewave input, 10W output, 8-ohm loading, 10Hz to 22.4kHz bandwidth):

Our primary measurements revealed the following using the phono-level input, MC configuration (unless specified, assume a 1kHz 0.5mVrms sinewave input, 10W output, 8-ohm loading, 10Hz to 22.4kHz bandwidth):

Our primary measurements revealed the following using the analog input at the headphone output (unless specified, assume a 1kHz sinewave, 2Vrms input/output, 300 ohms loading, 10Hz to 22.4kHz bandwidth):

Frequency response (8-ohm loading, line-level input)

In our frequency-response plots above (relative to 1kHz), measured across the speaker outputs at 10W into 8 ohms, the SA45 is essentially perfectly flat within the audioband (20Hz to 20kHz, 0/-0.1dB). The -3dB point is at roughly 80kHz, and 0dB at 5Hz. The SA45 appears to be DC-coupled. In the graph above and most of the graphs below, only a single trace may be visible. This is because the left channel (blue or purple trace) is performing identically to the right channel (red or green trace), and so they perfectly overlap, indicating that the two channels are ideally matched.

Frequency response (8-ohms loading, line-level analog input, ADC on)

Above are the frequency response plots (relative to 1kHz), measured across the speaker outputs at 10W into 8 ohms, with the ADC at the input on. The main difference here versus the plot above with an analog signal chain is the sharp high-frequency attenuation around 30kHz. The -3dB point is at roughly 46kHz.The ADC appears to be sampling at 96kHz.

Phase response (8-ohm loading, line-level input)

Above are the phase-response plots from 20Hz to 20kHz for the line-level input, measured across the speaker outputs at 10W into 8 ohms. The SA45 yielded only about -30 degrees of phase shift at 20kHz.

Frequency response (8-ohm loading, MM phono input)

The chart above shows the frequency response (relative to 1kHz) for the MM phono input. The red/blue plots are with the ADC off, purple/green with the ADC on. We see a very flat response from 100Hz to 20kHz (up to 10kHz when the ADC is on), and essentially no channel-to-channel deviations. Below 100Hz, there is steep attenuation (-3dB at roughly 17Hz), as Arcam appears to have implemented an anti-rumble filter on their phono input. What is shown is the deviation from the RIAA curve, where the input signal sweep is EQ’d with an inverted RIAA curve supplied by Audio Precision (i.e., zero deviation would yield a flat line at 0dB).

Phase response (MM input)

Above is the phase response plot from 20Hz to 20kHz for the MM phono input, measured across the speaker outputs at 10W into 8 ohms. The SA45 does not invert polarity. For the phono input, since the RIAA equalization curve must be implemented, which ranges from +19.9dB (20Hz) to -32.6dB (90kHz), phase shift at the output is inevitable. Here we find a worst case of about +20 degrees at 20Hz and -100 degrees at 20kHz.

Frequency response (8-ohm loading, MC phono input)

The chart above shows the frequency response (relative to 1kHz) for the MC phono input. The red/blue plots are with the ADC off, purple/green with the ADC on. We see essentially the same result as with the MM input.

Phase response (MC input)

Above is the phase response plot from 20Hz to 20kHz for the MC phono input, measured across the speaker outputs at 10W into 8 ohms. We see essentially the same result as with the MM input.

Frequency response vs. input type (8-ohm loading, left channel only)

The chart above shows the SA45’s frequency response (relative to 1kHz) as a function of input type measured across the speaker outputs at 10W into 8 ohms. The two green traces are the same analog input data from the speaker-level frequency-response graph above. The blue and red traces are for a 16-bit/44.1kHz dithered digital input signal from 5Hz to 22kHz using the coaxial input, the purple and green traces are for a 24/96 dithered digital input signal from 5Hz to 48kHz, and the pink and orange traces are for a 24/192 dithered digital input signal. At low frequencies, the digital signals yielded a flat response down to 5Hz, same as the analog response. The -3dB points are: 21kHz for the 16/44.1 data, 46kHz for the 24/96, 70kHz for the 24/192 data, and 80kHz for the analog input.

Frequency response vs. filter type (16/44.1, left channel only)

The chart above shows the SA45’s frequency response (relative to 1kHz) as a function of filter type measured across the speaker outputs at 10W into 8 ohms for a 16/44.1 digital input (left channel only). The blue plot is the default linear phase apodizing filter (#1), the red is the linear phase slow roll-off filter (#2), green is minimum phase slow roll-off (#3), and pink is the minimum phase filter (#4). We can see how filters 1 and 4 are brickwall filters showing only 0.2dB of attenuation at 20kHz, while the other two filters (2 and 3) are at -4dB at 20kHz.

Digital linearity (16/44.1 and 24/96 data)

The chart above shows the results of a linearity test for the coaxial digital input for both 16/44.1 (blue/red) and 24/96 (purple/green) input data, measured at the line-level pre-outputs of the SA45, where 0dBFS was set to yield 2Vrms. For this test, the digital input is swept with a dithered 1kHz input signal from -120dBFS to 0dBFS, and the output is analyzed by the APx555. The ideal response would be a straight flat line at 0dB. Both data were essentially perfect as of -100dBFS up to 0dBFS. The 24/96 data were flat down to about -115dBFS below which a muting circuit appears to engage, while the 16/44.1 data were +2/+4dB at -120dBFS.

Intersample headroom (+3dB 11.025kHz sinewave at 24/44.1 at -3/-1.5/0dBFS) — PASS

The chart above shows the results of an intersample headroom test for the coaxial digital input, measured at the line-level pre-outputs of the SA25, where a standard 0dBFS sinewave was set to yield 2Vrms (2.83Vp or 5.66Vpp). For this test, the DAC is fed a test file consisting of a 11.025kHz sinewave sampled at 24/44.1 at +3.01dB. This is achieved without digital clipping by using a sinewave at exactly one quarter the sample rate, completely avoiding sampling at the peaks and troughs of the waveform. The test file is then run through the DAC at -3, -1.5, and 0dBFS (purple/green/orange plots). A DAC with built-in headroom will be able to reconstruct all three sinewaves cleanly with no distortion, with the highest amplitude sinewave at 4Vp (3.01dB above the standard 2.83Vp for a 0dBFS input signal). A DAC without built-in headroom will show significant clipping (up to ~10% THD) when the test file is fed at -1.5 and 0dBFS. The DAC in the SA45 passed this test.

Impulse response (24/44.1 data, Filter 1)

The graph above shows the impulse response for a looped 24/44.1 test file that moves from digital silence to full 0dBFS (all “1”s) for one sample period then back to digital silence, measured at the line-level pre-outs of SA45, for Filter 1 (linear phase apoziding). We find a standard symmetrical sinc function response.

Impulse response (24/44.1 data, Filter 2)

The graph above shows the impulse response for a looped 24/44.1 test file that moves from digital silence to full 0dBFS (all “1”s) for one sample period then back to digital silence, measured at the line-level pre-outs of SA45, for Filter 2 (linear phase slow roll-off). We find a standard symmetrical sinc function response with less ringing than Filter 1.

Impulse response (24/44.1 data, Filter 3)

The graph above shows the impulse response for a looped 24/44.1 test file that moves from digital silence to full 0dBFS (all “1”s) for one sample period then back to digital silence, measured at the line-level pre-outs of SA45, for Filter 3 (minimum phase slow roll-off). We find a response with minimal pre-ringing and some post-ringing.

Impulse response (24/44.1 data, Filter 4)

The graph above shows the impulse response for a looped 24/44.1 test file that moves from digital silence to full 0dBFS (all “1”s) for one sample period then back to digital silence, measured at the line-level pre-outs of SA45, for Filter 4 (minimum phase slow roll-off). We find a response with no pre-ringing and significant post-ringing.

J-Test (coaxial)

The chart above shows the results of the J-Test test for the coaxial digital input measured at the line-level pre-outputs of the SA45 where 0dBFS is set to 2Vrms. J-Test was developed by Julian Dunn the 1990s. It is a test signal—specifically, a -3dBFS undithered 12kHz squarewave sampled (in this case) at 48kHz (24bit). Since even the first odd harmonic (i.e., 36kHz) of the 12kHz squarewave is removed by the bandwidth limitation of the sampling rate, we are left with a 12kHz sinewave (the main peak). In addition, an undithered 250Hz squarewave at -144dBFS is mixed with the signal. This test file causes the 22 least-significant bits to constantly toggle, which produces strong jitter spectral components at the 250Hz rate and its odd harmonics. The test file shows how susceptible the DAC and delivery interface are to jitter, which would manifest as peaks above the noise floor at 500Hz intervals (e.g,. 250Hz, 750Hz, 1250Hz, etc.). Note that the alternating peaks are in the test file itself, but at levels of -144dBrA and below. The test file can also be used in conjunction with artificially injected sinewave jitter by the Audio Precision, to show how well the DAC rejects jitter.

Here we see a strong J-Test result, with only a few very low-level peaks in the audioband, ranging from -145dBFS down to -160dBFS. This is an indication that the SA45 DAC should have strong jitter immunity.

J-Test (optical)

The chart above shows the results of the “J-test” test for the optical digital input measured at the line-level pre-outputs of the SA45. The optical input yielded very similar results compared to the coaxial input.

J-Test (coaxial, 10ns jitter)

The chart above shows the results of the J-Test test for the coaxial digital input measured at the line-level output of the SA45, with an additional 10ns of 2kHz sinewave jitter injected by the APx555. The telltale peaks at 10kHz and 12kHz can be seen but at a very low -145dBFS. The optical input behaved similarly.

J-Test (coaxial, 100ns jitter)

The chart above shows the results of the J-Test test for the coaxial digital input measured at the line-level output of the SA45, with an additional 100ns of 2kHz sinewave jitter injected by the APx555. The telltale peaks at 10kHz and 12kHz can be seen but are still low in amplitude at -125dBFS. The optical input behaved similarly.

Wideband FFT spectrum of white noise and 19.1kHz sine-wave tone (Filter 1, coaxial input)

The chart above shows a fast Fourier transform (FFT) of the SA45’s line-level pre-outputs with white noise at -4dBFS (blue/red) and a 19.1 kHz sinewave at 0dBFS fed to the coaxial digital input, sampled at 16/44.1 using Filter 1 (linear phase apodizing). The steep roll-off around 20kHz in the white-noise spectrum shows brickwall-type attenuation. There are no low-level aliased image peaks within the audioband above the -135dBrA noise floor. The primary aliasing signal at 25kHz is highly suppressed at -120dBrA, while the second and third distortion harmonics (38.2, 57.3kHz) of the 19.1kHz tone are near the same level.

Wideband FFT spectrum of white noise and 19.1kHz sine-wave tone (Filter 2, coaxial input)

The chart above shows a fast Fourier transform (FFT) of the SA45’s line-level pre-outputs with white noise at -4dBFS (blue/red) and a 19.1 kHz sinewave at 0dBFS fed to the coaxial digital input, sampled at 16/44.1 using Filter 2 (linear phase slow roll-off). The slow roll-off around 20kHz in the white-noise spectrum is as expected and advertised. There are two low-level aliased image peaks within the audioband at -120dBrA and below. The primary aliasing signal at 25kHz is barely suppressed at -15dBrA, while the second and third distortion harmonics (38.2, 57.3kHz) of the 19.1kHz tone are around the -110 to -120dBrA level.

Wideband FFT spectrum of white noise and 19.1kHz sine-wave tone (Filter 3, coaxial input)

The chart above shows a fast Fourier transform (FFT) of the SA45’s line-level pre-outputs with white noise at -4dBFS (blue/red) and a 19.1 kHz sinewave at 0dBFS fed to the coaxial digital input, sampled at 16/44.1 using Filter 3 (minimum phase slow roll-off). The slow roll-off around 20kHz in the white-noise spectrum is as expected and advertised. There are two low-level aliased image peaks within the audioband at -120dBrA and below. The primary aliasing signal at 25kHz is barely suppressed at -15dBrA, while the second and third distortion harmonics (38.2, 57.3kHz) of the 19.1kHz tone are around the -110 to -120dBrA level.

Wideband FFT spectrum of white noise and 19.1kHz sine-wave tone (Filter 3, coaxial input)

The chart above shows a fast Fourier transform (FFT) of the SA45’s line-level pre-outputs with white noise at -4dBFS (blue/red) and a 19.1 kHz sinewave at 0dBFS fed to the coaxial digital input, sampled at 16/44.1 using Filter 4 (minimum phase). The roll-off around 20kHz in the white-noise spectrum is steep but not as sharp as Filter 1. There are no low-level aliased image peaks within the audioband above the -135dBrA noise floor. The primary aliasing signal at 25kHz is highly suppressed at -105dBrA, while the second and third distortion harmonics (38.2, 57.3kHz) of the 19.1kHz tone are near the same level.

RMS level vs. frequency vs. load impedance (1W, left channel only)

The chart above shows RMS level (relative to 0dBrA, which is 1W into 8ohms or 2.83Vrms) as a function of frequency, for the analog line-level input swept from 5Hz to 50kHz. The blue plot is into an 8-ohm load, the purple is into a 4-ohm load, the pink plot is an actual speaker (Focal Chora 806, measurements can be found here), and the cyan plot is no load connected. The chart below . . .

. . . is the same but zoomed in to highlight differences. Here we see that the deviations between no load and 4 ohms are very small at roughly 0.04dB. This is a strong result and an indication of a very low output impedance, or high damping factor. With a real speaker load, deviations measured lower at roughly 0.03dB.

THD ratio (unweighted) vs. frequency vs. output power

The chart above shows THD ratios at the speaker-level outputs into 8 ohms as a function of frequency for a sinewave stimulus at the analog line-level input. The blue and red plots are for the left and right channels at 1W output into 8 ohms, purple/green at 10W, and pink/orange at 137W. The power was varied using the SA45’s volume control. The 10W THD ratios were the lowest (right channel), ranging from 0.00015% from 30Hz to 2kHz, then up to 0.0008% at 20kHz. The rest of the 1 and 10W data were close to these levels (within 5dB). At 137W, THD ratios ranged from 0.0005% from 30Hz to 1kHz, then up to 0.01% at 20kHz.

THD ratio (unweighted) vs. frequency at 10W (MM phono input)

The chart above shows THD ratios as a function of frequency plots for the MM (blue/red) and MC (purple/green) phono inputs measured across an 8-ohm load at 10W. For this test, the input sweep is EQ’d with an inverted RIAA curve. The THD values for the MM input vary from around 0.01% (20Hz) down to 0.0005% at 2kHz, then up to 0.002/0.001% (left/right) at 20kHz. The THD values for the MC input vary from around 0.05% (20Hz) down to 0.001% at 2 to 8kHz, then up to 0.0025% at 20kHz.

THD ratio (unweighted) vs. output power at 1kHz into 4 and 8 ohms

The chart above shows THD ratios measured at the speaker-level outputs of the SA45 as a function of output power for the analog line-level input, for an 8-ohm load (blue/red for left/right) and a 4-ohm load (purple/green for left/right), with the volume at maximum. THD ratios into 4 and 8 ohms are close (within 5dB). The 8-ohm data range from 0.002% at 50mW, down to 0.0003% in the 5 to 30W range. The “knee” into 8 ohms can be found right around 140W, while the 4-ohm “knee” can be seen around 210W. The 1% THD marks were hit at 177W and 275W into 8 and 4 ohms.

THD+N ratio (unweighted) vs. output power at 1kHz into 4 and 8 ohms

The chart above shows THD+N ratios measured at the speaker-level outputs of the SA45 as a function of output power for the analog line-level input, for an 8-ohm load (blue/red for left/right) and a 4-ohm load (purple/green for left/right). THD+N ratios into 4 and 8 ohms are remarkably close (with 2-3dB). The 8-ohm data range from 0.03% at 50mW, down to 0.0007% at the “knee.”

THD ratio (unweighted) vs. frequency at 8, 4, and 2 ohms (left channel only)

The chart above shows THD ratios measured at the output of the SA45 as a function of frequency into three different loads (8/4/2 ohms) for a constant input voltage that yields 20W at the output into 8 ohms (and roughly 40W into 4 ohms, and 80W into 2 ohms) for the analog line-level input. The 8-ohm load is the blue trace, the 4-ohm load the purple trace, and the 2-ohm load the pink trace. We find a roughly 5dB increase in THD from 8 to 4 to 2 ohms across the sweep (a strong result). These ranged from 0.00025% to 0.002% for the 8-ohm load and 0.001% to 0.005% for the 2-ohm load.

THD ratio (unweighted) vs. frequency into 8 ohms and real speakers (left channel only)

The chart above shows THD ratios measured at the output of the SA45 as a function of frequency into an 8-ohm load and two different speakers for a constant output voltage of 2.83Vrms (1W into 8 ohms) for the analog line-level input. The 8-ohm load is the blue trace, the purple plot is a two-way speaker (Focal Chora 806, measurements can be found here), and the pink plot is a three-way speaker (Paradigm Founder Series 100F, measurements can be found here). Generally, THD ratios into the real speakers were higher than those measured across the resistive dummy load. The differences ranged from 0.03% at 20Hz for the two-way speaker versus 0.00025% for the resistive load, and 0.006% at 20kHz into the three-way speaker versus 0.001% for the resistive load. Between the important frequencies of 500Hz to 4kHz, all three THD traces were very close, around the 0.0002-0.0005% level.

IMD ratio (CCIF) vs. frequency into 8 ohms and real speakers (left channel only)

The chart above shows intermodulation distortion (IMD) ratios measured at the output of the SA45 as a function of frequency into an 8-ohm load and two different speakers for a constant output voltage of 2.83Vrms (1W into 8 ohms) for the analog line-level input. Here the CCIF IMD method was used, where the primary frequency is swept from 20kHz (F1) down to 2.5kHz, and the secondary frequency (F2) is always 1kHz lower than the primary, with a 1:1 ratio. The CCIF IMD analysis results are the sum of the second (F1-F2 or 1kHz) and third modulation products (F1+1kHz, F2-1kHz). The 8-ohm load is the blue trace, the purple plot is a two-way speaker (Focal Chora 806, measurements can be found here), and the pink plot is a three-way speaker (Paradigm Founder Series 100F, measurements can be found here). Generally, IMD ratios into the real speakers were higher than those measured across the resistive dummy load, with the three-way speaker yielding the highest results at 0.001 to 0.003%. The two-way speaker yielded results between 0.0004 and 0.002%, while the resistive load yielded more constant results at roughly 0.0003% across the sweep.

IMD ratio (SMPTE) vs. frequency into 8 ohms and real speakers (left channel only)

The chart above shows IMD ratios measured at the output of the SA45 as a function of frequency into an 8-ohm load and two different speakers for a constant output voltage of 2.83Vrms (1W into 8 ohms) for the analog line-level input. Here, the SMPTE IMD method was used, where the primary frequency (F1) is swept from 250Hz down to 40Hz, and the secondary frequency (F2) is held at 7kHz with a 4:1 ratio. The SMPTE IMD analysis results consider the second (F2 ± F1) through the fifth (F2 ± 4xF1) modulation products. The 8-ohm load is the blue trace, the purple plot is a two-way speaker (Focal Chora 806, measurements can be found here), and the pink plot is a three-way speaker (Paradigm Founder Series 100F, measurements can be found here). We find very similar IMD ratios into all three loads, a constant  0.002% across the sweep.

FFT spectrum – 1kHz (XLR line-level input)

Shown above is the fast Fourier transform (FFT) for a 1kHz input sinewave stimulus, measured at the output across an 8-ohm load at 10W for the analog line-level XLR input. We see that the signal’s second (2kHz) and third (3kHz) harmonics dominate at a very low -115dBrA, or 0.0002%, and -125dBrA, or 0.00006% . There are subsequent signal harmonics visible at and below the extremely low -140dBrA, or 0.00001%, level. On the right side of the signal peak, we find power-supply-related noise peaks, but only at and below -130dBrA, or 0.00003%. Overall, this is a very clean FFT.

FFT spectrum – 1kHz (RCA line-level input)

Shown above is the fast Fourier transform (FFT) for a 1kHz input sinewave stimulus, measured at the output across an 8-ohm load at 10W for the analog line-level RCA input. The main difference between this FFT and the FFT above using the XLR input is slightly higher power-supply-related noise peaks, reaching -120dBrA, or 0.0001%, at the primary peak of 60Hz.

FFT spectrum – 1kHz (XLR line-level input, ADC on)

Shown above is the fast Fourier transform (FFT) for a 1kHz input sinewave stimulus, measured at the output across an 8-ohm load at 10W for the analog line-level XLR input, with the ADC sampling the incoming signal for the purposes of internal DSP. The main difference between this FFT and the FFT using the XLR input without any digitization is a few more higher order signal harmonics here. For example, the fifth (5kHz) and seventh (7kHz) harmonics can be seen here around the -120dBrA, or 0.0001%, level. What can also be seen are peaks at 95kHz and 97kHz, which is evidence of 96kHz sampling (the two IMD peaks between the signal and the sampling frequency).

FFT spectrum – 1kHz (MM phono input)

Shown above is the fast Fourier transform (FFT) for a 1kHz input sinewave stimulus, measured at the output across an 8-ohm load at 10W for the phono MM input. The signal harmonics are difficult to detect amongst the power-supply-related noise peaks, but the second (2kHz) and third (3kHz) harmonics are there just below -110dBrA, or 0.0003%. Power-supply-related noise peaks can be seen throughout the FFT ranging from -95dBrA, or 0.002%, down to -120dBrA, or 0.0001%.

FFT spectrum – 1kHz (MC phono input)

Shown above is the fast Fourier transform (FFT) for a 1kHz input sinewave stimulus, measured at the output across an 8-ohm load at 10W for the phono MC input. The signal harmonics are difficult to detect amongst the power-supply-related noise peaks, but the second (2kHz) and third (3kHz) harmonics are there just below -105dBrA, or 0.0006%. Power-supply-related noise peaks can be seen throughout the FFT ranging from -70dBrA, or 0.03%, down to -110dBrA, or 0.0003%.

FFT spectrum – 1kHz (digital input, 16/44.1 data at 0dBFS)

Shown above is the fast Fourier transform (FFT) for a 1kHz input sinewave stimulus, measured at the output across an 8-ohm load at 10W for the coaxial digital input, sampled at 16/44.1. We see that the signal’s second (2kHz) and third (3kHz) harmonics dominate at a very low -115dBrA, or 0.0002%, and -125dBrA, or 0.00006%, same as the analog FFT. Noise peaks are essentially the same as with the analog FFT above (at and below -130dBrA).

FFT spectrum – 1kHz (digital input, 24/96 data at 0dBFS)

Shown above is the fast Fourier transform (FFT) for a 1kHz input sinewave stimulus, measured at the output across an 8-ohm load at 10W for the coaxial digital input, sampled at 24/96. We see essentially the same result as with the 16/44.1 FFT above, but with a lower noise floor due to the increased bit depth.

FFT spectrum – 1kHz (digital input, 16/44.1 data at -90dBFS)

Shown above is the FFT for a 1kHz -90dBFS dithered 16/44.1 input sinewave stimulus at the coaxial digital input, measured at the output across an 8-ohm load. We see the 1kHz primary signal peak, at the correct amplitude, with no signal harmonics above the -135dBrA noise floor.

FFT spectrum – 1kHz (digital input, 24/96 data at -90dBFS)

Shown above is the FFT for a 1kHz -90dBFS dithered 24/96 input sinewave stimulus at the coaxial digital input, measured at the output across an 8-ohm load. We see the 1kHz primary signal peak, at the correct amplitude, no signal harmonics above the -145dBrA noise floor, and power-supply-related noise peaks at and below -135dBrA, or 0.00002%.

FFT spectrum – 50Hz (line-level input)

Shown above is the FFT for a 50Hz input sinewave stimulus measured at the output across an 8-ohm load at 10W for the analog line-level input. 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. The most predominant (non-signal) peak is the second (100Hz) signal harmonic at a low -115dBrA, or 0.0002%, followed by the third (150Hz) signal harmonic at -120dBrA, or 0.0001%. Other peaks (both signal harmonics and power-supply noise-related harmonics) can be seen at -130dBrA and below.

FFT spectrum – 50Hz (MM phono input)

Shown above is the FFT for a 50Hz input sinewave stimulus measured at the output across an 8-ohm load at 10W for the MM phono input. The X axis is zoomed in from 40 Hz to 1kHz, so that peaks from noise artifacts can be directly compared against peaks from the harmonics of the signal. The most predominant (non-signal) peaks are the signal’s second (100Hz) harmonic and the 60Hz fundamental power-supply noise peak and its second (120Hz left channel) and fifth (300Hz left channel) harmonics, all at roughly -95dBrA, or 0.002%.

FFT spectrum – 50Hz (MC phono input)

Shown above is the FFT for a 50Hz input sinewave stimulus measured at the output across an 8-ohm load at 10W for the MC phono input. 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. The most predominant (non-signal) peak is the 180Hz power-supply third-harmonic noise peak at -75dBrA, or 0.02%. Other power-supply-related noise peaks can be seen at -80dBrA, or 0.01%, and below. The highest signal harmonic is at 100Hz, at -90dBrA, or 0.003%.

Intermodulation distortion FFT (18kHz + 19kHz summed stimulus, line-level input)

Shown above is an FFT of the intermodulation distortion (IMD) products for an 18kHz + 19kHz summed sinewave stimulus tone measured at the output across an 8-ohm load at 10W for the analog line-level input. The input RMS values are set at -6.02dBrA so that, if summed for a mean frequency of 18.5kHz, would yield 10W (0dBrA) into 8 ohms at the output. We find that the second-order modulation product (i.e., the difference signal of 1kHz) is at -120dBrA, or 0.0001%, while the third-order modulation products, at 17kHz and 20kHz, are at -115dBrA, or 0.0002%. This is a very clean IMD result.

Intermodulation distortion FFT (line-level input, APx 32 tone)

Shown above is the FFT of the speaker-level output of the SA45 with the APx 32-tone signal applied to the analog input. The combined amplitude of the 32 tones is the 0dBrA reference, and corresponds to 10W into 8 ohms. The intermodulation products—i.e., the “grass” between the test tones—are distortion products from the amplifier and are at and below the very low -140dBrA, or 0.00001%, level.

Intermodulation distortion FFT (18kHz + 19kHz summed stimulus, coaxial digital input, 16/44.1)

Shown above is an FFT of the intermodulation distortion (IMD) products for an 18kHz + 19kHz summed sinewave stimulus tone measured at the output across an 8-ohm load at 10W for the digital coaxial input at 16/44.1 (0dBFS). We find that the second-order modulation product (i.e., the difference signal of 1kHz) is at -120/130dBrA (left/right), or 0.0001/0.00003%, while the third-order modulation products, at 17kHz and 20kHz, are very low at just under -120dBrA, or 0.0001%.

Intermodulation distortion FFT (18kHz + 19kHz summed stimulus, coaxial digital input, 24/96)

Shown above is an FFT of the intermodulation distortion (IMD) products for an 18kHz + 19kHz summed sinewave stimulus tone measured at the output across an 8-ohm load at 10W for the digital coaxial input at 24/96 (0dBFS). We find that the second-order modulation product (i.e., the difference signal of 1kHz) is at -120/130dBrA (left/right), or 0.0001/0.00003%, while the third-order modulation products, at 17kHz and 20kHz, are very low at just under -120dBrA, or 0.0001%.

Intermodulation distortion FFT (18kHz + 19kHz summed stimulus, MM phono input)

Shown above is an FFT of the intermodulation distortion (IMD) products for an 18kHz + 19kHz summed sinewave stimulus tone measured at the output across an 8-ohm load at 10W for the MM phono input. We find that the second-order modulation product (i.e., the difference signal of 1kHz) is at around -95dBrA, or 0.002%, while the third-order modulation products, at 17kHz and 20kHz, are lower at -110dBrA, or 0.0003%.

Intermodulation distortion FFT (18kHz + 19kHz summed stimulus, MC phono input)

Shown above is an FFT of the intermodulation distortion (IMD) products for an 18kHz + 19kHz summed sinewave stimulus tone measured at the output across an 8-ohm load at 10W for the MC phono input. We find that the second-order modulation product (i.e., the difference signal of 1kHz) is at around -90dBrA, or 0.003%, while the third-order modulation products, at 17kHz and 20kHz, are lower at -110dBrA, or 0.0003%.

Squarewave response (10kHz)

Above is the 10kHz squarewave response using the analog line-level input, at roughly 10W into 8 ohms. Due to limitations inherent to the Audio Precision APx555 B Series analyzer, this graph should not be used to infer or extrapolate the SA45’s slew-rate performance. Rather, it should be seen as a qualitative representation of the SA45’s mid-to-high bandwidth. An ideal squarewave can be represented as the sum of a sinewave and an infinite series of its odd-order harmonics (e.g., 10kHz + 30kHz + 50kHz + 70kHz . . .). A limited bandwidth will show only the sum of the lower-order harmonics, which may result in noticeable undershoot and/or overshoot, and softening of the edges. In this case, we find relatively clean corners, with some softening and overshoot.

Damping factor vs. frequency (20Hz to 20kHz)

The final graph above is the damping factor as a function of frequency. We can see here a constant damping factor just around 400 from 20Hz to 2kHz, then a decline to just under 200 at 20kHz. This is a strong result for a medium-powered solid-state integrated amplifier.

Diego Estan
Electronics Measurement Specialist