SCDA060 May   2025 CD4053B

 

  1.   1
  2.   Abstract
  3.   Trademarks
  4. 1Introduction
    1. 1.1 Demultiplexing Application Example in Solar System
    2. 1.2 Application Issue of Unwanted Voltage at OFF Channels
  5. 2OFF Channel Model Analysis
    1. 2.1 Channel Structure
    2. 2.2 Equivalent Resistor Model
  6. 3Fix OFF Channel Output Voltage to Ground
    1. 3.1 Pull-down Resistor
    2. 3.2 Pull-down Capacitor
    3. 3.3 Bleeder Resistor With a Switch
  7. 4Test and Measurement
    1. 4.1 Measurement Considerations
    2. 4.2 Test Result
  8. 5Summary
  9. 6References

Measurement Considerations

As mentioned in Section 1.2, the voltage readings of the multimeter and oscilloscope are different at the same OFF-channel output node. This is mainly due to the difference in input resistance of two instruments. As shown in Figure 4-1, when the probe is placed at two ends of the device under test (DUT) to detect voltage, the input resistance of multimeter and oscilloscope is actually in parallel with DUT.

 Loading Effect of Test EquipmentFigure 4-1 Loading Effect of Test Equipment

Usually, the input resistance of test equipment is large enough so that the effect of paralleling can be neglected. However, when the resistance of DUT is comparable to or even larger than the input resistance of test equipment, the effect of paralleling cannot be neglected and error can occur.

Such is the case in CD4053B’s application, where the OFF-channel equivalent resistance Roff,up, Roff,down, and op amp input resistance Rin,amp are enormous. The effect of test equipment on measurement is modeled by Rtest in Figure 4-2.

 Effect of Test Equipment on MeasurementFigure 4-2 Effect of Test Equipment on Measurement

Rtest is in parallel with Roff,down and Rin,amp. The paralleled resistors Roff,down ∥ Rtest ∥ Rin,amp divide the supply voltage with Roff,up. Therefore, different value of Rtest can lead to different divided voltage at the OFF-channel output.

Take Tektronix TPP0500B probe for oscilloscope as an example. The nominal probe input resistance Rtest 10MΩ. Assume Roff,up and Roff,down to be 60GΩ and Rin,amp to be 6.1 TΩ (typical value calculated in Section 3.1). Then Rtest is far less than Roff,up, Roff,down and Rin,amp. Therefore, the reading of oscilloscope is nearly 0V according to Equation 6 which is not the correct value.

For multimeter, the input resistance in voltage measurement mode is usually defined by the resistor network inside the network, which is not specified in the data sheet and can vary among different products. Keysight 34401A supports setting of input resistance to >10GΩ level. Testing with this setting can lead to a voltage reading of around 1V . Normally, the multimeter’s input resistance is fixed at 10MΩ, and the voltage reading is almost zero, similar to the result with oscilloscope.

To eliminate the effect of input resistance of test equipment, test the output voltage at the output of the unity gain buffer. According to the principle of unity gain buffer, the output voltage equals to the input voltage, and the input is isolated from the output, which means that the output load of unity gain buffer does not affect input. And no loading error can occur because the output resistance of unity gain buffer is much smaller compared to the input resistance of test equipment (For TLV9004, the open-loop output impedance in the data sheet is 1.2kΩ).

 Measuring Voltage With Help of Unity Gain BufferFigure 4-3 Measuring Voltage With Help of Unity Gain Buffer