Jun-Suk Bang
KAIST
Network
Latest external collaboration on country level. Dive into details by clicking on the dots.
Publication
Featured researches published by Jun-Suk Bang.
international solid-state circuits conference | 2015
Min-Yong Jung; Sang-Hui Park; Jun-Suk Bang; Dongchul Park; Se-Un Shin; Gyu-Hyeong Cho
Reducing the number of large external components, especially inductors, is a very important issue for Power-Management ICs (PMICs). Single-Inductor Multiple-Output (SIMO) converters are excellent candidates to meet this requirement [1-3]. However, there are several issues with SIMO converters, such as cross regulation, instability and inefficiency at light load. Under normal load conditions, comparator-based controlled SIMO converters [1,2] show good cross regulation performance due to the fast response of the comparator. However, the switching loss remains constant and degrades light load efficiency due to the fixed switching frequency of output switches. The low-efficiency characteristic when any output is under light load condition is a critical issue that must be solved because a SIMO converter is very suitable for light load applications. In addition, the cross regulation issue appears again when any output is under no load because the output receives energy from the inductor every cycle despite the load condition. To solve these issues, a SIMO converter was previously reported to support Pulse Frequency Modulation (PFM) mode [3]. However, the mode change control method increases the complexity of the control loop, which makes it unsuitable for a multi-output SIMO converter. In this paper, an Error Based Controlled (EBC) SIMO converter is presented to resolve the problems raised above using load-dependent Adaptive Pulse Modulation (APM). A hybrid topology composed of a switching converter and a linear regulator is also presented to minimize the cross regulation issue. To highlight the advantages, a 10-output SIMO converter is designed.
IEEE Journal of Solid-state Circuits | 2016
Jun-Suk Bang; Hyun-Sik Kim; Ki-Duk Kim; Oh-Jo Kwon; Choong-Sun Shin; Joo-hyung Lee; Gyu-Hyeong Cho
An active matrix organic light emitting diode (AMOLED) display driver IC, enabling real-time thin-film transistor (TFT) nonuniformity compensation, is presented with a hybrid driving method to satisfy fast driving speed, high TFT current accuracy, and a high aperture ratio. The proposed hybrid column-driver IC drives a mobile UHD (3840 × 2160) AMOLED panel, with one horizontal time of 7.7 μs at a scan frequency of 60 Hz, simultaneously senses the TFT current for back-end TFT variation compensation. Due to external compensation, a simple 3T1C pixel circuit is employed in each pixel. Accurate current sensing and high panel noise immunity is guaranteed by a proposed current-sensing circuit. By reusing the hybrid column-driver circuitries, the driver embodies an 8 bit current-mode ADC to measure OLED V -I transfer characteristic for OLED luminance-degradation compensation. Measurement results show that the hybrid driving method reduces the maximum current error between two emulated TFTs with a 60 mV threshold voltage difference under 1 gray-level error of 0.94 gray level (37 nA) in 8 bit gray scales from 12.82 gray level (501 nA). The circuit-reused current-mode ADC achieves 0.56 LSB DNL and 0.75 LSB INL.
IEEE Journal of Solid-state Circuits | 2015
Min-Yong Jung; Sang-Hui Park; Jun-Suk Bang; Gyu-Hyeong Cho
An error-based control method is proposed for a single-inductor multiple-output (SIMO) converter that maintains high efficiency under various load conditions. A peak efficiency of 88.7% was achieved with one output (IO1) of 5 mA, while the other outputs are at 432 mA. A hybrid topology composed of a switching converter and a linear regulator is also presented for fast load transient response. The proposed SIMO buck converter was fabricated with a 1P4M 0.35 μm BCD process and has 10 independently regulated outputs. The measured load transient waveform shows cross regulation of 0.1 mV/mA and a load transient of 0.17 mV/mA. This was achieved under load transient conditions for one output (IO9) in between 1 mA and 100 mA with the other outputs at 422 mA.
IEEE Transactions on Power Electronics | 2016
Si Duk Sung; Sung-Wan Hong; Jun-Suk Bang; Ji-Seon Paek; Seung-Chul Lee; Thomas Byunghak Cho; Gyu-Hyeong Cho
An envelope modulator (EM) is presented to increase the efficiency of an RF power amplifier. In order to supply an output voltage higher than the input voltage while providing low-frequency power in the EM, a single-inductor dual-output (SIDO) converter is introduced. By employing the SIDO converter, the EM does not require an additional boost converter. In addition, a high-frequency converter (HFC) with a wide-bandwidth capability is also proposed. These two converters, the SIDO converter and the HFC, are combined in parallel without an ac coupling capacitor by employing a low-frequency current-balancing technique. The chip is implemented in a 0.18-μm CMOS process and achieves 86.5% peak efficiency while tracking a 10-MHz long-term evolution envelope signal.
symposium on vlsi circuits | 2015
SiDuk Sung; Sung-Wan Hong; Jun-Suk Bang; Ji-Seon Paek; Seung-Chul Lee; Thomas Byunghak Cho; Gyu-Hyeong Cho
For achieving boost capability and wideband with high efficiency in Envelope Modulator (EM), a newly proposed topology is introduced in this paper. The proposed EM consists of two converters: one is Low Frequency Converter (LFC) with Single Inductor Dual Output (SIDO) and the other is High Frequency Converter (HFC) with wideband capability. The two converters are combined directly in parallel without AC coupling capacitor by employing Low Frequency Current Balancing (LFCB) technique. The chip is implemented in 0.18μm CMOS process achieving 86.55% peak efficiency while tracking a 10MHz LTE envelope signal.
international solid-state circuits conference | 2017
Yong-Min Ju; Se-Un Shin; Yeunhee Huh; Sang-Hui Park; Jun-Suk Bang; Ki-Duk Kim; Sung-Won Choi; Ji-Hun Lee; Gyu-Hyeong Cho
The number of mobile device users increases every year. Each mobile device is usually equipped with a Li-ion battery having voltage that varies from a minimum of 2.7V to a maximum of 4.2V. Therefore, as the battery voltage decreases with time, a DC-DC converter is required for a regulated supply lower or higher than the battery voltage. A simple buck converter is not suited for this case, since step-up conversion is not available [1]. Instead, a non-inverting buck-boost converter can be a solution over the entire range of the battery voltage [1–4]. Many research studies related to buck-boost converters operated on Li-ion batteries set the target output voltage at around 3.4V [3,4]. Since Li-ion batteries have a wide plateau from 3.6V to 3.8V and a small energy storage below the plateau, DC-DC converters are generally operated on step-down mode at most of the battery voltage range, as shown in Fig. 10.4.1 top. Notwithstanding, step-up conversion is also required for extracting the energy below the plateau even if it is a small amount in the battery. Therefore, in DC-DC converters, it is critical to maintain high efficiency over the whole range of the battery voltage when it operates on both step-down and step-up modes to prolong the battery usage effectively. However, if the conventional buck-boost topology of Fig. 10.4.1 bottom-left is used for step-up and step-down purposes, there are always two switches (S1 and S3) conducting in the main current path through the inductor. Thus, the switches become large in size to minimize the conduction loss. As the switching loss also increases when the switch size is larger, the efficiency of this structure is usually lower than that of the simple buck (or boost) converter [1]. In this respect, this paper proposes a topology named a flying-capacitor buck-boost (FCBB) converter suitable for such an application by obtaining both step-up and step-down operations with high efficiency throughout the whole range of the battery voltage.
symposium on vlsi technology | 2016
Changsik Shin; Wonji Lee; Sw Lee; Byeong Ho Jeong; Jong Min Lee; Unyong Jang; Yong-Goo Kim; Sang-Han Lee; Jun-Suk Bang; Gyu-Hyeong Cho
A phase-cut dimmable AC-DC buck LED driver is presented with a sine-reference band (SRB)-controlled average current technique as a new control scheme. The proposed SRB-controlled average current technique can perform both sine current control and phase dimming control. The sine current control regulates the average LED current as a sine-wave maintaining high power factor. The phase dimming control removes visible flicker at AC line frequency without an electrolytic capacitor and makes the driver compatible with two types of phase-cut dimmer. Experimental results show line regulation <; ±2.4% (90-260Vac), load regulation <; ±0.7% (10-36 LEDs), PF > 0.95, and 90.7% peak efficiency.
symposium on vlsi circuits | 2015
Jun-Suk Bang; Hyun-Sik Kim; Sang-Hui Park; Ki-Duk Kim; Sung-Won Choi; Oh-Jo Kwon; Choong-Sun Shin; Joo-hyung Lee; Gyu-Hyeong Cho
An UHD AMOLED display driver IC, enabling real-time TFT non-uniformity compensation, is presented with a hybrid driving scheme. The proposed hybrid driving scheme drives a mobile UHD (3840×1920) AMOLED panel, whose scan time is 7.7μs at a scan frequency of 60Hz, through the load of 30kohm resistance and 30pF capacitance. A proposed accurate current sensor embedded in the column driver and a back-end compensation scheme reduce maximum current error between emulated TFTs within 0.94 LSB (37nA) of 8-bit gray scales. Since the TFT variation is externally compensated, a simple 3T1C pixel circuit is employed in each pixel.
Journal of information display | 2014
Geon-Hee Kim; Hyun-Sik Kim; Sang-Hui Park; Jun-Suk Bang; Gyu-Hyeong Cho
This paper describes the design of a gamma-type current-mode digital-to-analog converter (DAC) for active-matrix organic light-emitting diode display (AMOLED) drivers. The proposed nonlinear DAC is composed of division-based bypassing DAC units for gamma correction. This multiple stack of DAC units dramatically reduces the chip size of the display driver while realizing more natural gamma correction. Moreover, it does not require an additional memory, such as a digital look-up table block and a DAC with a higher resolution, which can achieve significant manufacturing cost savings. Consequently, the proposed multi-stacked gamma-type DAC provides high area efficiency, low cost, and more natural gamma correction. The prototype 8-bit DAC is implemented in a TowerJazz 3.3 V CMOS process. The simulated gamma values of the double- and triple-stacked DAC units are varied from 1 to 2 and from 1 to 3, respectively. In addition, the Monte-Carlo simulation demenstrates that the proposed design shows uniform gamma transfer curves regardless of the process variation.
IEEE Journal of Solid-state Circuits | 2018
Yeunhee Huh; Sung-Wan Hong; Sang-Hui Park; Changsik Shin; Jun-Suk Bang; Changbyung Park; Sung-soo Park; Gyu-Hyeong Cho
A high-signal-to-noise ratio (SNR) inductor-free 3-D hover sensor is presented. This paper solved the low-signal component issue, which is the biggest problem in 3-D hover sensing. For this purpose, we propose a power- and cost-effective high-voltage driving technique in the self-capacitance sensing scheme (SCSS) and lateral resolution optimization of a touch panel. In addition, the huge panel offsets in the SCSS from both vertical panel capacitance (CSV) and horizontal panel capacitance (CSH) can effectively be eliminated by exploiting the panel’s natural characteristics, without using other costly resources. Therefore, in the proposed design, the total calibration block is minimized only for parasitic capacitance mismatches. Last, by adopting new driving scheme, two-phase simultaneous sensing is enabled to increase the SNR further. The proposed hover sensing system achieved a 39-dB SNR at a 1-cm hover point under a 240-Hz scan rate condition in noise experiments, while consuming 183