With this circuit, precise 10V output can be easily achieved

Analog output systems in industrial applications such as programmable logic controllers (PLCs), process control, or motor control require unipolar or bipolar voltage swings from 0 V to 10 V or more. One possible solution is to choose a bipolar output DAC that can directly generate the required output voltage; another is to use a low-voltage single-supply (LVSS) DAC and amplify its output voltage to the required output level. In order to choose the method that best suits the application, you must understand the output requirements and know the advantages or disadvantages of each solution.

Main advantages:

• Simple. Board design is simplified because the required output levels from 0 V to 10 V or above 10 V can be directly obtained through hardware or software configuration. In addition, fault protection modes are usually integrated, which simplifies system design.

• Manufacturability and reliability are improved because discrete components such as amplifiers, switches, and resistors are not required. Sometimes a voltage reference is also integrated.

• Measurement of systematic error and total unadjusted error (TUE). Linearity, noise, offset, and drift are guaranteed; by summing the various error sources within the DAC, it is easy to calculate the total systematic error or TUE. TUE is sometimes specified in the data sheet.

• Endpoint errors. In some cases, bipolar DACs include calibration features that can adjust system offset and gain errors over time.

Main disadvantages:

• Limited flexibility. An integrated high-voltage amplifier may not be optimal for the application. Output amplifiers are often optimized for specific load and noise requirements. While the data sheet ranges may match the actual load in the system, other parameters such as settling time or power consumption may not meet the system requirements.

• Cost and board area. Bipolar DACs are typically designed on larger geometry processes, resulting in larger die and package sizes and higher costs. Using a low-voltage DAC with external signal conditioning is another way to generate the high-voltage output swing and range required for industrial applications. Again, it has important trade-offs to consider.

Main advantages:

• LVSS DACs have a high logic integration and high-speed logic interface , which frees up the microcontroller to handle more tasks.

• The output may have to source high currents or drive large capacitive loads that the bipolar DAC's on-chip amplifier cannot handle. A discrete solution allows the best stand-alone amplifier to be selected to meet the application needs.

• An overrange feature (10.8 V output for a 10 V nominal range) is easily implemented, providing the end user with greater application flexibility, such as in applications where worn valves need to be opened or closed.

• Cost. LVSS DACs are generally less expensive than bipolar DACs, resulting in a lower overall bill of materials cost.

• Reduced board area. LVSS DACs are designed using low-voltage submicron or deep micron processes to provide small footprint packages.

Main disadvantages:

• It takes more time to optimize the board and design the endpoint adjustment circuits.

• Calculation of the total error or TUE becomes more difficult because more error sources must be considered.

• The increase in the number of discrete components leads to reduced manufacturability and reliability.

• Applications A low voltage power supply (5 V or 3 V) must be available.

In summary, there are many factors to consider in the design of precision 10 V industrial applications. Obviously, you must have a clear understanding of the output load requirements and the total error that can be accepted in the system. In addition, board area and cost are also important considerations in selecting the best solution. For applications that must drive large capacitive loads (1 μF) while requiring low noise and fast settling (20 V range less than 10 s), discrete solutions almost always win; although bipolar DACs are not as flexible as discrete solutions, their simple design and easy TUE calculation make them attractive for a wide range of industrial and instrumentation applications.

The following discussion shows how to achieve a precision 10 V output using a dual-supply bipolar output DAC and a low voltage single-supply DAC with external signal conditioning .

The main components of a bipolar output DAC are shown in the functional block diagram above (Figure 1). It consists of a precision DAC, a voltage reference, a reference buffer, offset and gain adjustments, and an output amplifier.

Integrating a precision voltage reference to accommodate 16-bit applications is difficult, but recent process advances and design techniques allow voltage references with excellent drift and thermal characteristics to be designed and integrated on-chip. Fault protection modes such as thermal shutdown, short-circuit protection, and output control during power-up/power-down conditions are important features that are often integrated into bipolar DACs to simplify system design. The DAC provides a digital code to convert the output voltage relative to the reference voltage. The adjustment block provides the ability to offset and adjust the DAC transfer function.

*The AD5764 is a quad-channel, 16-bit serial input, voltage output DAC that operates from 12 V to 15 V. It has a nominal full-scale output range of 10 V and includes an output amplifier, reference buffer, precision reference, and proprietary power-on/power-down control circuitry. The AD5764 is designed using Analog Devices' industrial CMOS (iCMOS®) manufacturing process technology, which combines high-voltage complementary bipolar transistors with submicron CMOS. It also has an analog temperature sensor and digital offset and gain adjustment registers for each channel.

Figure 2 shows how to use an LVSS DAC to generate the 10 V output range required for industrial applications. It consists of five different blocks: LVSS DAC, reference, offset adjustment, reference buffer, and output amplifier.

The DAC provides the digital code to convert the output voltage relative to the reference voltage. The offset adjustment block provides the function of shifting the DAC unipolar transfer function to produce a bipolar output, as well as calibrating the 0 V endpoint. The reference buffer provides load isolation for the reference voltage and offset adjustment block (multiple DACs can share this buffered output). The output amplifier provides the required gain to increase the output swing to the required level after accounting for the offset adjustment. In addition, the output amplifier provides the ability to drive large capacitive loads to the supply rails.

The circuit shown in Figure 3 illustrates how to amplify a precision LVSS 16-bit DAC to achieve a 10 V output swing . The DAC has a 0 to 2.5 V output range and is connected to the noninverting input of amplifier U3. The noninverting gain of this input is (1 + R2/R1), which is 8 in this case.

The inverting input of the op amp is connected to the 1.429 V voltage generated by the reference and resistor divider network U6. The inverting gain of this input is (–R2/R1), which is –7 in this case. Therefore, when the DAC is set to 0 code, 0000h, the output of this circuit is:

When the DAC is set to full-scale code FFFFh, the output is:

In general, the output voltage for any input code can be calculated as follows:

Where D represents the decimal input code (0 to 65535) to a precision 16-bit DAC (as in this example). VREF = 2.5 V, R1 = R, and R2 = 7 R. A digital potentiometer with nonvolatile memory is used to adjust the zero offset error of the system so that the offset value is retained even when power is removed. U7, U6, and R3 can be selected to form a resistor network to provide the required adjustment range for 0 V. Other output ranges required by the PLC analog output module, such as +5 V, 5 V, +10 V, or 10.8 V (for cases where overrange is important), can be easily configured.

*The 16-bit AD5062 is guaranteed monotonic with a maximum DNL and INL error of 1 LSB. Its unipolar output has a maximum offset error of 50 V and a maximum gain error of 0.02%. The high-speed serial interface supports clock rates up to 30 MHz. The device is available in a small SOT-23 package.

More and more industrial and instrumentation applications require the use of precision converters to achieve accurate control and measurement of various processes. In addition, these end applications also require higher flexibility, reliability and feature sets while reducing costs and board area. Component manufacturers are addressing these challenges and have launched a range of products to meet the requirements of system designers for current and future designs.

As shown in this article, there are multiple ways to select the right components for precision applications, each with its own advantages and disadvantages. As system accuracy increases, more attention needs to be paid to selecting the right components to meet the application requirements.