AWR eBooks

RF Electronics Chapter 10: Operational Amplifiers Page 348 2022, C. J. Kikkert, James Cook University, ISBN 978-0-6486803-9-0. Hardware Realisation These design simulations are only part of the story. The circuit of figure 10.5 with Rg removed, Rs = 1 MΩ and Rp = 100 Ωk, was used for an early version of this amplifier. The circuit used AD8045 OpAmps, which are single OpAmps with a 1.3 pF input capacitance. Figure 10.9 shows the measured frequency response of this circuit when different size surface-mount 1 MΩ, 1% tolerance, resistors are used for Rs. The SMD resistors have a very small capacitance and inductance associated with them and that changes the high frequency response. These effects can only be measured using actual hardware. Since the 3216 Metric SMD resistors have a good frequency response and a 200V rating, they were used in the hardware realisation of figure 10.11. Figure 10.9. High input impedance OpAmp measured frequency response. When building the full circuit of figure 10.4, it was found that the common mode input specification for the hardware, required a very good gain and phase match of OpAmps A1 and A2. This can only be obtained using Dual OpAmps. As a result, the AD8058, Dual OpAmp shown in figure 10.4 were used. As part of this hardware, another amplifier was used to inject a PLC frequency onto the power line. The Voltage Sensor of figure 10.4 was then used to measure that PLC frequency voltage on that power line. Since there is some electromagnetic coupling between the PLC frequency signal generator and the Voltage Sensor, the attenuation due to Rp should be minimised, by keeping Rp as large as possible. For the hardware of figure 10.4, a satisfactory frequency response, noise performance and isolation between the different parts of the equipment was obtained using Rp = 68 kΩ. This example shows that a high frequency OpAmp design involves many compromises to ensure that the components used and the OpAmp characteristics cause the least degradation in system noise and bandwidth performance. Figure 10.10 shows the measured and simulated frequency response of the whole circuit. Even though both A1 and A2 are in the same package, the layout is made as symmetrical as possible and the matched resistors from the same batches are used, the frequency response of the +ve input (A1) and the –ve input (A2) are different. A prototype made with single AD8045 amplifiers had bigger differences between the inputs. RF Electronics: Design and Simulation 348 www.cadence.com/go/awr

• Cover