Precision, Unipolar, Inverting Conversion Using the AD5546/AD5556 DAC
Figure 1: Unipolar 2-Quadrant Multiplying Mode, VOUT = 0 to -VREF
This circuit uses the ADR03, which is a highly accuracy, high stability, 2.5 V precision voltage reference. As voltage reference temperature coefficient and long-term drift are primary considerations for applications requiring high precision conversion, this device is an ideal candidate.
An op amp is used in the current-to-voltage (I-V) stage of this circuit. An op amp’s bias current and offset voltage are both important selection criteria for use with precision current output DACs. Therefore, this circuit employs the AD8628 auto-zero op amp, which has ultralow offset voltage (1 μV typical) and bias current (30 pA typical). The value of C7 for this circuit is 2.2 pF, which is optimized to compensate for the external output capacitance of the DAC.
Note that the AD8628 has rail-to-rail input and output stages, but the output can only come within a few millivolts of either rail depending on load current. For the circuit shown, the output can swing from –2.5 V to approximately –1 mV.
The input offset voltage of the op amp is multiplied by the variable noise gain (due to the code-dependent output resistance of the DAC) of the circuit. A change in this noise gain between two adjacent digital codes produces a step change in the output voltage due to the amplifier’s input offset voltage. This output voltage change is superimposed on the desired change in output between the two codes and gives rise to a differential linearity error, which, if large enough, could cause the DAC to be nonmonotonic. In general, the input offset voltage should be a fraction of an LSB to ensure monotonic behavior when stepping through codes. For the ADR03 and the AD5546, the LSB size is
The input offset voltage of the AD8628 auto-zero op amp is typically 1 μV, which is negligible compared to the LSB size.
The input bias current of an op amp also generates an offset at the voltage output as a result of the bias current flowing through the feedback resistor, RFB. In the case of the AD8628, the input bias current is only 30 pA typical, which flowing through the RFB resistor (10 kΩ typical), produces an error of only 0.3 μV.
The AD5546/AD5556 DAC architecture uses a current-steering R-2R ladder design that requires an external reference and op amp to generate the output voltage. VOUT can be calculated for the AD5546 using the equation
where D is the decimal equivalent of the input code. VOUT can be calculated for the AD5556 using the equation
where D is the decimal equivalent of the input code.
These circuits can also be used as a variable gain element by utilizing the multiplying bandwidth nature of the R-2R structure of the AD5546/AD5556 DAC. In this configuration, remove the external precision reference and apply the signal to be multiplied to the reference input pins of the DAC.