ZHCSBB1D July   2013  – March 2018 UCC28740


  1. 特性
  2. 应用
  3. 说明
    1.     Device Images
      1.      简化应用示意图
      2.      典型伏安图
  4. 修订历史记录
  5. Pin Configuration and Functions
    1.     Pin Functions
  6. Specifications
    1. 6.1 Absolute Maximum Ratings
    2. 6.2 ESD Ratings
    3. 6.3 Recommended Operating Conditions
    4. 6.4 Thermal Information
    5. 6.5 Electrical Characteristics
    6. 6.6 Switching Characteristics
    7. 6.7 Typical Characteristics
  7. Detailed Description
    1. 7.1 Overview
    2. 7.2 Functional Block Diagram
    3. 7.3 Feature Description
      1. 7.3.1 Detailed Pin Description
      2. 7.3.2 Valley-Switching and Valley-Skipping
      3. 7.3.3 Startup Operation
      4. 7.3.4 Fault Protection
    4. 7.4 Device Functional Modes
      1. 7.4.1 Secondary-Side Optically Coupled Constant-Voltage (CV) Regulation
      2. 7.4.2 Primary-Side Constant-Current (CC) Regulation
  8. Application and Implementation
    1. 8.1 Application Information
    2. 8.2 Typical Application
      1. 8.2.1 Design Requirements
      2. 8.2.2 Detailed Design Procedure
        1. Custom Design With WEBENCH® Tools
        2. Standby Power Estimate and No-Load Switching Frequency
        3. Input Bulk Capacitance and Minimum Bulk Voltage
        4. Transformer Turns-Ratio, Inductance, Primary Peak Current
        5. Transformer Parameter Verification
        6. VS Resistor Divider, Line Compensation
        7. Output Capacitance
        8. VDD Capacitance, CVDD
        9. Feedback Network Biasing
      3. 8.2.3 Application Curves
  9. Power Supply Recommendations
  10. 10Layout
    1. 10.1 Layout Guidelines
      1. 10.1.1 VDD Pin
      2. 10.1.2 VS Pin
      3. 10.1.3 FB Pin
      4. 10.1.4 GND Pin
      5. 10.1.5 CS Pin
      6. 10.1.6 DRV Pin
      7. 10.1.7 HV Pin
    2. 10.2 Layout Example
  11. 11器件和文档支持
    1. 11.1 器件支持
      1. 11.1.1 开发支持
        1. 使用 WEBENCH® 工具定制设计方案
      2. 11.1.2 器件命名规则
        1.  电容术语(以法拉为单位)
        2.  占空比术语
        3.  频率术语(以赫兹为单位)
        4.  电流术语(以安培为单位)
        5.  电流和电压调节术语
        6.  变压器术语
        7.  功率术语(以瓦特为单位)
        8.  电阻术语(以 Ω 为单位)
        9.  时序术语(以秒为单位)
        10. 电压术语(以伏特为单位)
        11. 交流电压术语(以 VRMS 为单位)
        12. 效率术语
    2. 11.2 文档支持
      1. 11.2.1 相关文档
    3. 11.3 接收文档更新通知
    4. 11.4 社区资源
    5. 11.5 商标
    6. 11.6 静电放电警告
    7. 11.7 Glossary
  12. 12机械、封装和可订购信息


机械数据 (封装 | 引脚)
散热焊盘机械数据 (封装 | 引脚)

Feedback Network Biasing

Achieving very low standby power while maintaining high-performance load-step transient response requires careful design of the feedback network. Optically coupled secondary-side regulation is used to provide the rapid response needed when a heavy load step occurs during the no-load condition. One of the most commonly used devices to drive the optocoupler is the TL431 shunt-regulator, due to its simplicity, regulation performance, and low cost. This device requires a minimum bias current of 1 mA to maintain regulation accuracy. Together with the UCC28740 primary-side controller, careful biasing will ensure less than 30 mW of standby power loss at room temperature. Where a more stringent standby loss limit of less than 10 mW is required, the TLV431 device is recommended due to its minimum 80-µA bias capability.

Facilitating these low standby-power targets is the approximate 23-µA range of the FB input for full to no-load voltage regulation. The control-law profile graph (see Figure 15) shows that for FB-input current greater than 22 µA, no further reduction in switching frequency is possible. Therefore, minimum power is converted at fSW(min). However, the typical minimum steady-state operating frequency tends to be in the range of several-hundred Hertz, and consequently the maximum steady-state FB current at no-load will be less than IFBMAX. Even so, prudent design practice dictates that IFBMAX should be used for conservative steady-state biasing calculations. At this current level, VFBMAX can be expected at the FB input.

Referring to the Design Procedure Application Example in Figure 18, the main purpose of RFB4 is to speed up the turnoff time of the optocoupler in the case of a heavy load-step transient condition. The value of RFB4 is determined empirically due to the variable nature of the specific optocoupler chosen for the design, but tends to fall within the range of 10 kΩ to 100 kΩ. A tradeoff must be made between a lower value for faster transient response and a higher value for lower standby power. RFB4 also serves to set a minimum bias current for the optocoupler and to drain dark current.

It is important to understand the distinction between steady-state no-load bias currents and voltages which affect standby power, and the varying extremes of these same currents and voltages which affect regulation during transient conditions. Design targets for minimum standby loss and maximum transient response often result in conflicting requirements for component values. Trade-offs, such as for RFB4 as discussed previously, must be made.

During standby operation, the total auxiliary current (used in Equation 8) is the sum of IWAIT into the IC and the no-load optocoupler-output current ICENL. This optocoupler current is given by Equation 30.

Equation 30. UCC28740 q_Icenl_lusbf3.gif

For fast response, the optocoupler-output transistor is biased to minimize the variation of VCE between full-load and no-load operation. Connecting the emitter directly to the FB input of the UCC28740 is possible, however, an unload-step response may unavoidably drive the optocoupler into saturation which will overload the FB input with full VDD applied. A series-resistor RFB3 is necessary to limit the current into FB and to avoid excess draining of CVDD during this type of transient situation. The value of RFB3 is chosen to limit the excess IFB and RFB4 current to an acceptable level when the optocoupler is saturated. Like RFB4, the RFB3 value is also chosen empirically during prototype evaluation to optimize performance based on the conditions present during that situation. A starting value may be estimated using Equation 31.

Equation 31. UCC28740 q_Rfb3_lusbf3.gif

Note that RFB3 is estimated based on the expected no-load VDD voltage, but full-load VDD voltage will be higher resulting in initially higher ICE current during the unload-step transient condition. Because RFB3 is interposed between VE and the FB input, the optocoupler transistor VCE varies considerably more as ICE varies and transient response time is reduced. Capacitor CFB3 across RFB3 helps to improve the transient response again. The value of CFB3 is estimated initially by equating the RFB3CFB3 time constant to 1 ms, and later is adjusted higher or lower for optimal performance during prototype evaluation.

The optocoupler transistor-output current ICE is proportional to the optocoupler diode input current by its current transfer ratio, CTR. Although many optocouplers are rated with nominal CTR between 50% and 600%, or are ranked into narrower ranges, the actual CTR obtained at the low currents used with the UCC28740 falls around 5% to 15%. At full-load regulation, when IFB is near zero, VFB is still approximately 0.4 V and this sets a minimum steady-state current for ICE through RFB4. After choosing an optocoupler, the designer must characterize its CTR over the range of low output currents expected in this application, because optocoupler data sheets rarely include such information. The actual CTR obtained is required to determine the diode input current range at the secondary-side shunt-regulator.

Referring again to Figure 18, the shunt-regulator (typically a TL431) current must be at least 1 mA even when almost no optocoupler diode current flows. Since even a near-zero diode current establishes a forward voltage, ROPT is selected to provide the minimum 1-mA regulator bias current. The optocoupler input diode must be characterized by the designer to obtain the actual forward voltage versus forward current at the low currents expected. At the full-load condition of the converter, IFB is around 0.5 µA, ICE may be around (0.4 V / RFB4), and CTR at this level is about 10%, so the diode current typically falls in the range of 25 µA to 100 µA. Typical opto-diode forward voltage at this level is about 0.97 V which is applied across ROPT. If ROPT is set equal to 1 kΩ, this provides 970 µA plus the diode current for IOPT.

As output load decreases, the voltage across the shunt-regulator also decreases to increase the current through the optocoupler diode. This increases the diode forward voltage across ROPT. CTR at no-load (when ICE is higher) is generally a few percent higher than CTR at full-load (when ICE is lower). At steady-state no-load condition, the shunt-regulator current is maximized and can be estimated by Equation 30 and Equation 32. IOPTNL, plus the sum of the leakage currents of all the components on the output of the converter, constitute the total current required for use in Equation 7 to estimate secondary-side standby loss.

Equation 32. UCC28740 q_Ioptnl_lusbf3.gif

The shunt-regulator voltage can decrease to a minimum, saturated level of about 2 V. To prevent excessive diode current, a series resistor, RTL, is added to limit IOPT to the maximum value necessary for regulation. Equation 33 provides an estimated initial value for RTL, which may be adjusted for optimal limiting later during the prototype evaluation process.

Equation 33. UCC28740 q_Rtl_lusbf3.gif

The output-voltage sense-network resistors RFB1 and RFB2 are calculated in the usual manner based on the shunt-regulator reference voltage and input bias current. Having characterized the optocoupler at low currents and determined the initial values of RFB1, RFB2, RFB3, RFB4, CFB3, ROPT and RTL using the above procedure, the DC-bias states of the feedback network can be established for steady-state full-load and no-load conditions. Adjustments of these initial values may be necessary to accommodate variations of the UCC28740, optocoupler, and shunt-regulator parameters for optimal overall performance.

The shunt-regulator compensation network, ZFB, is determined using well-established design techniques for control-loop stability. Typically, a type-II compensation network is used. The compensation design procedure is beyond the scope of this datasheet.