Circuit functions and advantages

The circuit shown in Figure 1 provides a 20-bit programmable voltage with an output range of −10 V to +10 V, an integral nonlinearity of ±1 LSB, a differential nonlinearity of ±1 LSB, and low noise characteristics.

The circuit's digital inputs are serial inputs and are compatible with standard SPI, QSPI™, MICROWIRE® and DSP interface standards. For high-precision applications, this circuit can provide high-precision and low-noise performance by combining precision devices such as the AD5791 , AD8675 , and AD8676 .

Reference buffering is critical to the design because the input impedance of the DAC reference input is highly code-dependent, resulting in linearity errors if the DAC reference source is not adequately buffered. The AD8676 has an open loop gain of up to 120 dB and has been verified and tested to meet the settling time, offset voltage, and low impedance drive capability requirements of this circuit application. The AD5791 has been characterized and factory calibrated so that its reference voltage input can be buffered using the dual-channel operational amplifier AD8676, further enhancing the reliability of the accompanying device.

This combination of devices delivers industry-leading 20-bit resolution, ±1 LSB integral nonlinearity (INL) and ±1 LSB differential nonlinearity (DNL), ensuring monotonicity, with low power consumption, small PCB size, and high Cost-effectiveness and other characteristics.

Circuit description

The digital-to-analog converter (DAC) shown in Figure 1 is the AD5791, a 20-bit high-voltage converter with an SPI interface that provides ±1 LSB INL, ±1 LSB DNL performance, and 7.5 nV/√Hz noise spectral density. In addition, the AD5791 also has extremely low temperature drift (0.05 ppm/°C). The precision architecture of the AD5791 requires forced sensing to buffer its reference voltage input to ensure specified linearity. The amplifiers chosen to buffer the reference input (B1 and B2) should have low noise, low temperature drift, and low input bias current. The AD8676 amplifier is recommended for this function, an ultra-precision, 36 V, 2.8 nV/√Hz dual op amp with low offset drift of 0.6 μV/°C and 2 nA input bias current. In addition, the AD5791 is characterized and factory calibrated to use the dual op amp to buffer its voltage reference input, further enhancing the reliability of the companion device.

Figure 1 shows that the AD5791 is configured with independent positive and negative reference voltages, so the output voltage range is from the negative reference voltage to the positive reference voltage, in this case −10 V to +10 V. The output buffer is the AD8675, which is a single-channel version of the AD8676 with low noise and low drift. The AD8676 amplifier (A1 and A2) is also used to amplify the +5 V reference voltage to +10 V and −10 V. R2, R3, R4, and R5 in these amplifier circuits are precision metal film resistors with a tolerance of 0.01% and a temperature coefficient of 0.6 ppm/°C. For optimal performance over the entire temperature range, resistor networks such as the Vishay 300144 or VSR144 series can be used. The resistor values ​​chosen are low (1 kΩ and 2 kΩ) to keep system noise low. R1 and C1 form a low-pass filter with a cutoff frequency of approximately 10 Hz. This filter is used to attenuate reference noise.

Linearity measurement

The following data further demonstrates the precision performance of the circuit shown in Figure 1. Figures 2 and 3 show integral nonlinearity and differential nonlinearity as a function of DAC code. It is evident from the figure that these two characteristics are within the specifications of ±1 LSB and ±1 LSB respectively.

The total unregulated error of this circuit is composed of various DC errors, namely INL error, zero-scale error, and full-scale error. Figure 4 shows a plot of total unadjusted error versus DAC code. The maximum errors occur at DAC codes of 0 (zero-scale error) and 1,048,575 (full-scale error). This is as expected and is caused by mismatches in resistor pairs R2 and R3, R4 and R5, and offset errors in amplifiers A1, A2, B1, and B2 (see Figure 1).

In this example, the resistor pair has a maximum nominal mismatch of 0.02% (typically much less than this). The amplifier offset error is 75 μV (max) or 0.000375% of full-scale range, which is negligible relative to the error due to resistor mismatch. Therefore, the maximum expected full-scale and zero-scale errors are approximately 0.02% or 210 LSB. Figure 4 shows a measured full-scale error of 1 LSB and a measured zero-scale error of 4 LSB or 0.0003% of the full-scale range, indicating that all devices performed significantly better than their rated maximum tolerances.

Noise measurement

To achieve high accuracy, the peak-to-peak noise at the output of the circuit must remain below 1 LSB, which is 19.07 μV for 20-bit resolution and a 20 V peak-to-peak voltage range. Figure 5 shows the peak-to-peak noise measured over a 0.1 Hz to 10 Hz bandwidth over 10 seconds. The peak-to-peak values ​​under the three conditions are 1.48 μV (mid-level output), 4.66 μV (full-scale output), and 5.45 μV (zero-level output). The mid-level output has the lowest noise, where the noise comes only from the DAC core. When mid-level codes are selected, the DAC attenuates the noise contribution of each reference voltage path.

However, a practical application will not have a high-pass cutoff frequency at 0.1 Hz to attenuate 1/f noise, but will include frequencies as low as DC in its passband; therefore, the measured peak-to-peak noise is more realistic, as shown in Figure 6 shown. In this example, the noise at the output of the circuit was measured over 100 seconds, and the measurement fully covers frequencies as low as 0.01Hz. The upper cutoff frequency is approximately 14 Hz and is limited by the measurement setup. For the three conditions shown in Figure 6, the corresponding peak-to-peak values ​​are 4.07 μV (mid-level output), 11.85 μV (full-scale output), and 15.37 μV (zero-level output). The worst-case peak-to-peak value (15.37 μV) is roughly equivalent to 0.8 LSB.

As the measurement time becomes longer, lower frequencies will be included and the peak-to-peak values ​​will become larger. At lower frequencies, temperature drift and thermocouple effects become sources of error. These effects can be minimized by selecting devices with lower thermal coefficients such as the AD5791, AD8675, and AD8676, and by carefully considering the circuit structure, see the linked documents in the "Learn More Information" section.