SBOA524 July   2021 ALM2402-Q1 , ALM2402F-Q1 , ALM2403-Q1 , INA1620

 

  1.   Trademarks
  2. 1Circuit Theory
    1. 1.1 Overview of Parallel Amplifier Approaches
    2. 1.2 Considerations
    3. 1.3 Current Mismatch Equations
    4. 1.4 Stability
  3. 2Qualitative and Empirical Comparisons
    1. 2.1 Overview Comparison
    2. 2.2 Monte Carlo Comparison
  4. 3References

Overview of Parallel Amplifier Approaches

When implementing a multiple-buffer arrangement, it is theoretically possible to use a very simple circuit where the desired input signal is applied to the noninverting input of each driving amplifier, with the feedback of both amplifiers taken at the load. However, this can lead to stability issues as the amplifiers' differing offsets cause them to fight over the proper load voltage, and can even force the amplifiers into current limit as they source/sink current between each other. Thus, to be practical the buffer approach usually requires the use of ballast resistors to isolate the amplifier outputs and feedback from each other, effectively limiting the output voltage swing when sourcing or sinking significant currents.

Additionally, high output currents will cause a voltage drop across the ballast resistance (as per Ohm’s law), degrading the accuracy of the system. To fix this problem, an error amplifier (with its feedback taken at the load) is often used to drive the non-inverting inputs of the driving amplifiers instead. Input and feedback resistors will typically be required for impedance matching and input protection of the drivers, while feedback capacitances are often required (both for the driving amplifiers and for the error amplifier) to improve the circuit phase margin and make the circuit stable. Thus, in practice the conventional parallel buffer arrangement actually looks more like the circuit shown in Figure 1-1, where RtraceX is the parasitic resistance of the PCB or cabling.

Figure 1-1 Conventional Buffer Circuit

By simply adding two additional resistors per channel to the circuit of Figure 1-1, it is possible to dramatically modify the mechanics of the circuit. Each driving amplifier will now be operated as an improved Howland current pump, regulating the voltage across (and thus the current through) the ballast resistance instead of simply regulating the voltage at one node of the ballast resistance. The resulting circuit is shown in Figure 1-2.

Figure 1-2 Parallel Improved Howland Pump Circuit

This approach makes the circuit practically immune to mismatches in the trace resistances of the two channels, at the expense of only two additional components. Typically, the pumps will be used in an attenuating gain configuration – this has the benefit of reducing the circuit’s susceptibility to mismatches in the ballast resistors, as compared to the conventional buffer arrangement, as well as reducing the sensitivity of the circuit to offset mismatch. Limiting the driving amplifier bandwidth (often required for stability purposes) can confer additional benefits in the form of limiting the circuit’s integrated noise.