SPRACZ0A August   2021  – March 2023 TMS320F2800132 , TMS320F2800132 , TMS320F2800133 , TMS320F2800133 , TMS320F2800135 , TMS320F2800135 , TMS320F2800137 , TMS320F2800137 , TMS320F2800152-Q1 , TMS320F2800152-Q1 , TMS320F2800153-Q1 , TMS320F2800153-Q1 , TMS320F2800154-Q1 , TMS320F2800154-Q1 , TMS320F2800155 , TMS320F2800155 , TMS320F2800155-Q1 , TMS320F2800155-Q1 , TMS320F2800156-Q1 , TMS320F2800156-Q1 , TMS320F2800157 , TMS320F2800157 , TMS320F2800157-Q1 , TMS320F2800157-Q1 , TMS320F280021 , TMS320F280021 , TMS320F280021-Q1 , TMS320F280021-Q1 , TMS320F280023 , TMS320F280023 , TMS320F280023-Q1 , TMS320F280023-Q1 , TMS320F280023C , TMS320F280023C , TMS320F280025 , TMS320F280025 , TMS320F280025-Q1 , TMS320F280025-Q1 , TMS320F280025C , TMS320F280025C , TMS320F280025C-Q1 , TMS320F280025C-Q1 , TMS320F280033 , TMS320F280033 , TMS320F280034 , TMS320F280034 , TMS320F280034-Q1 , TMS320F280034-Q1 , TMS320F280036-Q1 , TMS320F280036-Q1 , TMS320F280036C-Q1 , TMS320F280036C-Q1 , TMS320F280037 , TMS320F280037 , TMS320F280037-Q1 , TMS320F280037-Q1 , TMS320F280037C , TMS320F280037C , TMS320F280037C-Q1 , TMS320F280037C-Q1 , TMS320F280038-Q1 , TMS320F280038-Q1 , TMS320F280038C-Q1 , TMS320F280038C-Q1 , TMS320F280039 , TMS320F280039 , TMS320F280039-Q1 , TMS320F280039-Q1 , TMS320F280039C , TMS320F280039C , TMS320F280039C-Q1 , TMS320F280039C-Q1 , TMS320F280040-Q1 , TMS320F280040-Q1 , TMS320F280040C-Q1 , TMS320F280040C-Q1 , TMS320F280041 , TMS320F280041 , TMS320F280041-Q1 , TMS320F280041-Q1 , TMS320F280041C , TMS320F280041C , TMS320F280041C-Q1 , TMS320F280041C-Q1 , TMS320F280045 , TMS320F280045 , TMS320F280048-Q1 , TMS320F280048-Q1 , TMS320F280048C-Q1 , TMS320F280048C-Q1 , TMS320F280049 , TMS320F280049 , TMS320F280049-Q1 , TMS320F280049-Q1 , TMS320F280049C , TMS320F280049C , TMS320F280049C-Q1 , TMS320F280049C-Q1 , TMS320F28075 , TMS320F28075 , TMS320F28075-Q1 , TMS320F28075-Q1 , TMS320F28076 , TMS320F28076 , TMS320F28374D , TMS320F28374D , TMS320F28374S , TMS320F28374S , TMS320F28375D , TMS320F28375D , TMS320F28375S , TMS320F28375S , TMS320F28375S-Q1 , TMS320F28375S-Q1 , TMS320F28376D , TMS320F28376D , TMS320F28376S , TMS320F28376S , TMS320F28377D , TMS320F28377D , TMS320F28377D-EP , TMS320F28377D-EP , TMS320F28377D-Q1 , TMS320F28377D-Q1 , TMS320F28377S , TMS320F28377S , TMS320F28377S-Q1 , TMS320F28377S-Q1 , TMS320F28378D , TMS320F28378D , TMS320F28378S , TMS320F28378S , TMS320F28379D , TMS320F28379D , TMS320F28379D-Q1 , TMS320F28379D-Q1 , TMS320F28379S , TMS320F28379S , TMS320F28384D , TMS320F28384D , TMS320F28384D-Q1 , TMS320F28384D-Q1 , TMS320F28384S , TMS320F28384S , TMS320F28384S-Q1 , TMS320F28384S-Q1 , TMS320F28386D , TMS320F28386D , TMS320F28386D-Q1 , TMS320F28386D-Q1 , TMS320F28386S , TMS320F28386S , TMS320F28386S-Q1 , TMS320F28386S-Q1 , TMS320F28388D , TMS320F28388D , TMS320F28388S , TMS320F28388S , TMS320F28P550SJ , TMS320F28P550SJ , TMS320F28P559SJ-Q1 , TMS320F28P559SJ-Q1 , TMS320F28P650DK , TMS320F28P650DK , TMS320F28P650SH , TMS320F28P650SH , TMS320F28P650SK , TMS320F28P650SK , TMS320F28P659DK-Q1 , TMS320F28P659DK-Q1

 

  1.   Abstract
  2.   Trademarks
  3. 1Introduction
    1. 1.1 Resources
      1. 1.1.1 TINA-TI SPICE-Based Analog Simulation Program
      2. 1.1.2 PSPICE for TI Design and Simulation Tool
      3. 1.1.3 Application Report: ADC Input Circuit Evaluation for C2000 MCUs
      4. 1.1.4 TI Precision Labs - SAR ADC Input Driver Design Series
      5. 1.1.5 Analog Engineer's Calculator
      6. 1.1.6 TI Precision Labs - Op Amps: Stability Series
      7. 1.1.7 Related Application Reports
      8. 1.1.8 Comparison of Schematic Capture and Simulation Tools
      9. 1.1.9 PSpice for TI ADC Input Models
  4. 2Charge-Sharing Concept
    1. 2.1 Traditional High-Speed ADC Driving Circuits
    2. 2.2 Increased Cs in High-Speed ADC Driving Circuits
    3. 2.3 Very Large Cs in ADC Driving Circuits
    4. 2.4 Charge-Sharing Operation
    5. 2.5 Sample Rate and Source Impedance vs. Tracking Error
    6. 2.6 Analytical Solution to Tracking Error
    7. 2.7 Charge-Sharing in Multiplexed ADCs
    8. 2.8 Charge-Sharing Circuit Advantages
    9. 2.9 Charge-Sharing Circuit Disadvantages
  5. 3Charge Sharing Design Flow
    1. 3.1 Gather Required Information
    2. 3.2 Size Cs
    3. 3.3 Verify Sample Rate, Source Impedance, and Bandwidth
    4. 3.4 Simulate Circuit Settling Performance
    5. 3.5 Input Design Worksheet
  6. 4Charge-Sharing Circuit Simulation Methods
    1. 4.1 Simulation Components
      1. 4.1.1 Vin
      2. 4.1.2 Voa , Voa_SS, and Verror
      3. 4.1.3 Rs, Cs, and Vcont
      4. 4.1.4 Ch, Ron, and Cp
      5. 4.1.5 S+H Switch, Discharge Switch, tacq, and tdis
    2. 4.2 Configure the Simulation Parameters
    3. 4.3 Bias Point Analysis to Determine Voa_ss
    4. 4.4 Transient Analysis to Determine Voa_ss
    5. 4.5 Measure the Settling Error
    6. 4.6 Sweeping Source Resistance
  7. 5Example Circuit Designs
    1. 5.1 Example 1: Determining Maximum Sample Rate
      1. 5.1.1 Example 1: Analysis
      2. 5.1.2 Example 1: Simulation
      3. 5.1.3 Example 1: Worksheet
    2. 5.2 Example 2: Adding an Op-amp
      1. 5.2.1 Example 2: Analysis
      2. 5.2.2 Example 2: Simulation
      3. 5.2.3 Example 2: Worksheet
    3. 5.3 Example 3: Reduced Settling Target
      1. 5.3.1 Example 3: Analysis
      2. 5.3.2 Example 3: Simulation
      3. 5.3.3 Example 3: Worksheet
    4. 5.4 Example 4: Voltage Divider
      1. 5.4.1 Example 4: Analysis
      2. 5.4.2 Example 4: Simulation
      3. 5.4.3 Example 4: Worksheet
  8. 6Summary
  9.   A Appendix: ADC Input Settling Motivation
    1.     A.1 Mechanism of ADC Input Settling
    2.     A.2 Symptoms of Inadequate Settling
      1.      A.2.1 Distortion
      2.      A.2.2 Memory Cross-Talk
      3.      A.2.3 Accuracy
      4.      A.2.4 C2000 ADC Architecture
  10.   References
  11.   Revision History

2.5 Sample Rate and Source Impedance vs. Tracking Error

Designers of high speed op-amp based ADC signal conditioning circuits are required to trade-off settling speed against op-amp bandwidth and external component sizes. Meeting the charge-sharing criterion for source capacitor size ensures that the minimum ADC S+H window duration can be used regardless of source impedance. However, this creates a new trade-off that needs to be optimized: sample rate and source impedance vs. tracking error.

Significant tracking error occurs when the ADC samples too fast relative to the source's ability to recharge Cs through Rs in the time between samples. This is illustrated by simulating the circuit shown in Figure 2-3. The results from the this simulation can be seen in Figure 2-4: each step down in the sawtooth waveform is a charge equalization due to ADC sampling. In this situation, the first sample causes the voltage on the pin to drop by about 0.5 LSBs, resulting in a good sample of the DC input voltage. However, the external source is only able to partially recharge Cs before the next sample occurs. Each sample subsequently bleeds off some of the charge on Cs. Eventually, an equilibrium is reached when the delta between the pin voltage and the source voltage increases enough to drive 0.5 LSBs of recovery in the time between samples. In the simulation results of Figure 2-4, the equilibrium is reached at about 3.6 mV of tracking error.

GUID-A728EAC6-81D4-4064-B87C-04ADFBB0FD30-low.jpg Figure 2-3 ADC Input Circuit With 1 kΩ Rs and 100 kHz Sample Rate
GUID-1AEEA0CE-C920-45E9-9CFB-507C3A00E42B-low.jpg Figure 2-4 Simulation Results for 1kΩ Rs and 100 kHz Sample Rate

The settling error target is usually set to 0.5 LSBs, but applications may be able to tolerate more or less settling error. If a 1 LSB settling error target is assumed, the previous example circuit is still far out of target range with 0.5 LSBs of charge-share error plus 3.6 mV of tracking error (about 5 LSBs). To get the tracking error within tolerance, either the sample rate needs to be reduced or the source impedance, Rs, needs to be reduced (or both).

Figure 2-5 and Figure 2-6 show that by significantly reducing the source impedance or the sample rate the tracking error can be made to be of a similar magnitude to the error from charge equalization, resulting in an overall settling error within the target of 1 LSBs.

GUID-93FBA639-3FAF-4F0E-95C9-4EF6471AA82E-low.jpg Figure 2-5 Simulation Results for 100Ω Rs and 100 kHz Sample Rate
GUID-3005FD05-D970-4D8A-B105-5B0F634FCE26-low.jpg Figure 2-6 Simulation Results for 1kΩ Rs and 10kHz Sample Rate
Note: The example above shows a DC source voltage with a S+H capacitor that always starts fully discharged to 0 V. This makes the resulting tracking error look like the ADC input pin is drooping to a lower voltage. In the case of an AC input signal the voltage on the pin will lag behind the applied voltage. In the case the S+H capacitor starts at a voltage other than 0 V (likely in multiplexed ADCs) the pin voltage will be pulled towards that other voltage (possibly the previously sampled channel in the mux sampling sequence).