1、模数转换器主要内容1、ADC的性能指标2、SAR ADC的设计3、ADC的测试方法采样速率奈奎斯特ADC过采样ADCSAR ADCPipelined ADCCyclic ADCDual-slope ADCFlash ADCSigma-delta ADC品质因子（FOM-figure of merit）SAR ADC的设计SuccessiveApproximationRegister(SAR)N-Bit Search DACInputVoltageSample-and-Hold(S/H)AmplifierAnalogComparator 25 Advanced Analog IC DesignS
2、uccessive Approximation ADCProfessor Y.ChiuECE 581Fall 2009Charge Redistribution SA ADC4-bit binary-weighted capacitor array DACCapacitor array samples input when 1 is asserted(bottom-plate)26 Advanced Analog IC DesignSuccessive Approximation ADCProfessor Y.ChiuECE 581Fall 2009Charge Redistribution(
3、MSB)27 Advanced Analog IC DesignSuccessive Approximation ADCProfessor Y.ChiuECE 581Fall 2009Comparison(MSB)If VX VR/2,and MSB=1,C4 remains connected to VRIf VX 0,then Vi VR/2,and MSB=0,C4 is switched to ground 28 Advanced Analog IC DesignSuccessive Approximation ADCProfessor Y.ChiuECE 581Fall 2009Ch
4、arge Redistribution(MSB-1)29 Advanced Analog IC DesignSuccessive Approximation ADCProfessor Y.ChiuECE 581Fall 2009Comparison(MSB-1)If VX 3VR/4,and MSB-1=1,C3 remains connected to VRIf VX 0,then Vi 3VR/4,and MSB-1=0,C3 is switched to ground 30 Advanced Analog IC DesignSuccessive Approximation ADCProf
5、essor Y.ChiuECE 581Fall 2009Charge Redistribution(Other Bits)Test completes when all four bits are determined w/four charge redistributions and comparisons 31 Advanced Analog IC DesignSuccessive Approximation ADCProfessor Y.ChiuECE 581Fall 2009After Four Clock CyclesDAC设计比较器设计由于生产工艺过程中的误差造成电路结构不能和设计
6、中的那样对称，引起比较器的失调。比较器的失调电压会对比较器的性能和比较器后续电路的工作有影响，会造成误码现象。比较器一般有下面两种形式的失调消除方式：输入失调电压存储方式输出失调电压存储方式其电路包括前置放大器，失调存储电容，还有锁存单元。输入失调电压存储方式输出失调电压存储方式多级简单预放大+锁存结构低输入电容SAR ADC全差分输入SAR ADC时域比较器ADC的测试方法1、静态测试2、动态测试Dynamic TestingTypes of Dynamic Testsq Histogram or Code-Density Testq FFT TestFischer 0844Histogra
7、m or Code-Density Testq DNL appears as deviation of bin height from ideal value.q Integral nonlinearity(INL)is cumulative sum(integral)of DNL.q Offset is manifested by a horizontal shift of curve.q Gain error shows as horizontal compression or decompression of curve.Fischer 0845Histogram TestqInput
8、frequency must be selected carefully to avoid missing codes (fclk/fin must be non-integer ratio).q Input Swing is critical(cover full range)q Requires a large number of conversions(o 2n x 1,000).Fischer 0846Simulated Histogram TestFischer 0847FFT TestPros and Cons of FFT Testq Offers quantitative In
9、formation on output Noise,Signal-to-Noise Ratio(SNR),Spurious Free Dynamic Range(SFDR)and Harmonic Distortion(SNDR).q FFT test requires fewer conversions than histogram test.q Complete characterization requires multiple tests with various input frequencies.q Does not reveal actual code conversionsFischer 0848Simulated FFT TestSFDR=60 dBSNDR=49 dB ENOB=7.85
