Instruction Manual for DAC3 HGC and DAC3 L with 2.X Firmware Page 37
when loaded with real headphones. The
measurements use ideal resistive loads. Our
measurements show that headphones do not
behave like resistive loads.
Headphone Performance
In our tests we have measured substantial
distortion across resistors that are wired in
series with headphones. We have conducted
measurements with a variety of headphones.
In general, distortion increases as headphone
impedance decreases. This distortion can be
eliminated with a properly designed "0-Ohm"
headphone amplifier.
The performance of the HPA2™ does not
change when headphones are driven. THD+N
measurements for no-load, 30-Ohm resistive
loads, 30-Ohm headphone loads, and 600-
Ohm headphone loads are virtually identical.
The HPA2™ will substantially improve the
sound of 30 and 60-Ohm headphones. It will
make very noticeable improvements with
600-Ohm headphones.
Differential Amplifiers
Differential amplifiers remove common-mode
distortion components from the D/A converter
outputs. This feature is critical for achieving
low-distortion in down-stream devices.
Benchmark addresses common-mode
distortion so that it will not cause distortion in
power amplifiers and other connected
devices. Common-mode distortion can cause
audible distortion while escaping the scrutiny
of an audio analyzer. The balanced and
unbalanced outputs on the DAC3 deliver very
similar performance.
Many D/A converters omit the differential
amplifiers after the converters. Specifications
usually ignore common-mode distortion. A
balanced signal with high common-mode
distortion can measure just fine when feeding
a precisely balanced input on a high-quality
audio analyzer. However, any imbalance in a
downstream device will expose the common-
mode distortion.
Jitter-Immune UltraLock3™
UltraLock3™ is an improved version of the
UltraLock2™ clock system used in the
DAC2. The new UltraLock3™ system
provides faster lock times than the older
UltraLock2™ and UltraLock™ systems. The
DSP processing is 32-bits, DSP headroom is
3.5 dB, sample rate is 211 kHz, and jitter-
induced distortion and noise is at least 160 dB
below the level of the music - well below the
threshold of hearing. Benchmark's
UltraLock3™ system eliminates all audible
jitter artifacts.
The Importance of Eliminating Jitter
Accurate 24-bit audio conversion requires a
very low-jitter conversion clock. Jitter can
very easily turn a 24-bit converter into a 16-
bit converter (or worse). There is no point in
buying a 24-bit converter if clock jitter has
not been adequately addressed.
Jitter is present on every digital audio
interface. This type of jitter is known as
‘interface jitter’ and it is present even in the
most carefully designed audio systems.
Interface jitter accumulates as digital signals
travel down a cable from one digital device to
the next. If we measure interface jitter in a
typical system we will find that it is 10 to
10,000 times higher than the maximum
allowable level for accurate 24-bit conversion.
Fortunately, interface jitter has absolutely no
effect on the audio unless it influences the
conversion clock in an analog-to-digital
converter (A/D), the conversion clock in a
digital-to-analog converter (D/A), or the rate
estimator in an asynchronous sample rate
converter (ASRC).
Many converters use a single-stage Phase
Lock Loop (PLL) circuit to derive their
conversion clocks from AES/EBU, wordclock,
or superclock reference signals. Single-stage
PLL circuits provide some jitter attenuation
above 5 kHz but none below 5 kHz.
Unfortunately, digital audio signals often have
their strongest jitter components at 2 kHz.
Consequently, these converters can achieve
their rated performance only when driven
To Next Page
Loading...
