Input Circuit

In an amplifier, the input circuit is often a crucial component that determines the overall noise performance.

If the input circuit itself generates high noise, no amount of downstream noise reduction measures can fully eliminate it.
This noise ultimately dominates the residual noise of the entire amplifier.

Furthermore, if the input circuit, which directly interfaces with the external environment, is unable to effectively eliminate noise, it may pick up electromagnetic interference or noise leakage from the player—resulting in unpleasant noise from earphones or headphones.
This type of noise is mainly known as common-mode noise, and it can be eliminated using an amplifier configured with a balanced circuit.

At Brise Audio, we place great importance on three key elements in the input circuit: low residual noise, high common-mode noise rejection ratio (CMRR), and low distortion—aiming to meet all of them at a high level.

This goal exists to achieve one fundamental principle: “fidelity to the source.”

BIS1.0
This circuit was first implemented in TSURANAGI, Brise Audio’s inaugural electronics product.
It is a circuit that uses a low-distortion balanced line receiver IC designed for audio applications.

To enhance CMRR, it is crucial to precisely subtract the positive and negative sides of the input signal.
To achieve this, we adopted an IC that integrates the entire circuit into a single package.
This approach offers significantly better component matching than constructing the circuit from discrete parts.

As a result, it achieves outstanding performance with over 80dB of CMRR at 1kHz.


BIS2.0
FUGAKU and WATATSUMI are both equipped with this circuit.

In recent years, as earphones on the market have become increasingly low-impedance and high-sensitivity, we have recognized the growing need for lower residual noise.
This refers to the phenomenon where white noise becomes audible from earphones due to the residual noise generated by the circuit itself.

To address this issue, we lowered the cutoff frequency of the filter in the first stage—originally designed to block high-frequency noise—so it targets the audio band, thereby reducing the amount of noise passed on to subsequent stages.
Although it is tuned for the audio band, the cutoff frequency is set between 200kHz and 400kHz to avoid degrading transient response.

Furthermore, the previously integrated amplifier within a single IC was separated into two distinct stages.
This was done because the first-stage amplifier contributes significantly to residual noise, and separating it allows for the use of a higher-performance component.

However, using only discrete components would make it difficult to achieve the precise component matching needed to enhance CMRR. Therefore, we adopted an audio-grade amplifier that integrates thin-film resistors and the amplifier into a single IC.

As a result, we succeeded in reducing residual noise to less than one-fourth of that in BIS1.0, while maintaining a CMRR of 80dB.

Output Circuit

In an amplifier, the output circuit is a critical component that determines distortion under load, waveform fidelity, and transient response.

FUGAKU is a rare example of an amplifier that drives a dedicated earphone as a fixed load. In most cases, however, amplifiers are required to handle a wide range of load impedances—from compact earphones to large headphones.


With low-impedance loads, sufficient sound pressure can be achieved with a small output voltage, but such loads also tend to be more sensitive to noise.
Additionally, since less voltage is needed, the amplifier is often required to supply more current.

In general, the more current an amplifier is required to deliver, the more prone it becomes to distortion, and higher current is perceived as a heavier load.
Maintaining low distortion under heavy load requires advanced circuit design and PCB layout techniques.


With high-impedance loads, relatively high output voltage is required to achieve sufficient sound pressure.
Since devices like smartphones operate on relatively low power supply voltages, it is difficult for their built-in amplifiers to generate sufficient sound pressure, making high-impedance headphones challenging to drive.

However, high impedance also means that less current is required.
In other words, it means that the conditions that typically cause amplifier distortion are less likely to occur.

To drive high output voltages, the amplifier's power supply voltage must be set sufficiently high.
In this regard, portable amplifiers with high supply voltages designed to accommodate high-impedance loads offer significant advantages.

Higher supply voltage increases power consumption, leading to trade-offs such as greater heat generation and shorter battery life, so simply raising the supply voltage is not always the optimal solution.


In addition to impedance, earphones and headphones also have a parameter known as sensitivity (or efficiency).
1mWの電力をかけたときに音圧が何dBSPL得られるか、という指標です。

インピーダンスがある程度高くても能率が極端に低い場合、電力(電圧と電流の掛け算)が求められるため、結果的に高い電圧と高い電流の両方が要求されることもあります。
このようなケースが最もアンプにとって負荷が高く、本当の意味で"鳴らしづらい"イヤホン・ヘッドホンと言えるでしょう。

Additionally, because earphones and headphones use voice coils, the amplifier sees them as complex loads consisting of inductive (inductance), capacitive (capacitance), and resistive components.

Capacitive loads are particularly troublesome, as they can cause abnormal oscillations in amplifiers, making it essential to design with such conditions in mind.
They are primarily divided into the capacitance introduced by the cable and the parasitic capacitance inherent in the driver.

The resistive component mainly comes from the wiring resistance of the cable between the amplifier and the earphones or headphones, as well as from components within the crossover network.


In recent years, high-end earphones have been trending toward lower impedance due to the increasing number of drivers, while the headphone market has seen a rise in low-impedance planar magnetic models and low-sensitivity units.
At Brise Audio, our goal is to develop amplifiers that can drive all types of earphones and headphones on the market without compromise.

BOS1.0
This circuit was first implemented in TSURANAGI, Brise Audio’s inaugural electronics product.

The signal is converted from balanced to unbalanced in the input circuit, then passed through the electronic volume and voltage amplifier before entering this output circuit.
To output a balanced signal, an unbalanced-to-balanced conversion is performed at this stage using an FD Amp (fully differential operational amplifier).

Since a fully differential OPAMP alone cannot adequately drive the load, a current feedback amplifier is placed downstream to handle the drive.

The current feedback amplifier operates independently as a unity gain buffer, and its output is fed back into the fully differential OPAMP, forming a composite amplifier configuration.
While the current feedback amplifier alone can drive with relatively low distortion, additional distortion cancellation is performed by the fully differential OPAMP to further suppress distortion.

The isolator is a circuit designed to prevent oscillation caused by the capacitive component of the load.
Placing impedance between the amplifier and the load helps to isolate them, but using resistance in the audio band lowers the damping factor and weakens driving force. Therefore, components with high impedance are used only in the high-frequency range where oscillation is likely to occur.

Since the isolator must also handle high current, components are carefully selected to avoid introducing distortion.

The DC servo circuit is used to cancel the amplifier’s DC offset and is implemented to eliminate the need for AC coupling capacitors at the output.

Because it must cancel only the DC offset without affecting the main audio band, careful attention is given to ensure it does not introduce unwanted noise or distortion.


BOS1.5
This circuit, first introduced in WATATSUMI, is an improved version based on BOS1.0.
A new version of the component used in the isolator has been released, and it has been replaced with one that supports more than twice the rated current.
As a result, driving capability has been further improved, and distortion at near-maximum output levels has been reduced.

Additionally, the DC servo’s feedback configuration has been revised to eliminate one resistor, and the OPAMP has been replaced with a state-of-the-art model that delivers ultra-high performance—meeting all criteria for low noise, low distortion, low offset, and low power consumption.

Furthermore, the PCB layout philosophy has been updated by positioning the ground (GND)—which serves as the reference for the circuit—symmetrically near the output and equidistant from both left and right channels.
The copper thickness of the PCB has also been doubled, among other enhancements, to ensure the amplifier remains stable and unwavering even during high current output—reflecting the meticulous care taken in its design.

As a result of these improvements, distortion at a 16Ω load has been reduced by more than one-third, and maximum output has been tripled.