The previous three parts of this blog series focused on 2-wire 4-20mA sensor transmitter designs composed completely of analog components. While analog signal conditioning is practical for linear sensors, many sensors have nonlinear outputs that can only be modestly corrected with analog compensation. Examples of non-linear sensors include pressure sensors, gas and chemical sensors and many temperature sensors. Therefore, high-accuracy designs require advanced math or lookup tables to convert the sensor output into an accurate analog linear control signal for the 4-20mA transmitter. In these systems, the sensor output is acquired using an analog-to-digital converter (ADC) and then processed in a microcontroller (MCU) before being converted back to an analog signal to control the transmitter.
Figures 1 and 2 show two simplified 2-wire transmitter designs that use an MCU with an integrated ADC for sensor acquisition and digital processing. Figure 1 displays a high-accuracy design using a 16-bit digital-to-analog converter (DAC), such as the DAC8411, to convert the processed sensor signal into an accurate analog control signal. Figure 2 uses the MCU pulse-width modulated (PWM) output and an external low-pass filter to create a cost-effective, but less accurate DAC.
Figure 1: 2-wire sensor transmitter with MCU+ADC and DAC
Figure 2: 2-wire sensor transmitter with MCU+ADC and PWM DAC
As I explained in my previous posts, the MCU and DAC must be fully powered between the VREG and IRET pins of the XTR116, and the IRET pin must not be connected to the VLOOP GND or to any other external voltage potentials. Also, the circuitry powered from the loop must consume less than 3.8mA so that the XTR116 can properly output the 4mA zero-scale level.
While the system designs shown in Figures 1 and 2 are common, you will occasionally have to design a system using an MCU powered from a voltage source referenced to another external potential, possibly VLOOP GND, to control the output of a 2-wire transmitter. Figure 3 shows an example of this type of system.
Figure 3: Issues with controlling a 2-wire transmitter with an externally powered MCU
In this type of system, the sensor and MCU can’t be directly connected to the DAC or PWM filter circuitry. If the sensor and MCU are directly connected, the 2-wire transmitter GND pin (IRET) will be pulled down to VLOOP GND, or another voltage potential, preventing the transmitter from functioning properly. Even if the GND pins aren’t connected, trying to connect the externally powered MCU output signals to the transmitter will cause issues because the sensor GND currents will try to find a return path through the 2-wire transmitter.
Therefore, as shown in Figure 4, you must use a digital isolator between the sensor and MCU circuitry and the 2-wire transmitter circuitry. Integrated isolation solutions such as the ISO7310 can accomplish the digital isolation. However, be sure that the digital isolator doesn’t consume more than the allowed current budget for the 2-wire transmitter.
Figure 4: Isolating an externally powered MCU from the 2-wire transmitter with the ISO7310 digital isolator
In summary, if you’re trying to control the output of a 2-wire transmitter from an externally powered control source, be sure to include a digital isolator between the controller and the 2-wire transmitter circuitry to prevent the grounding issues I described in detail in Part 2 of this series.
In Part 5 of this series, I’ll discuss the design of 2-wire sensor transmitters that require the sensor circuitry to be powered from, but also be completely isolated from, the rest of the 2-wire transmitter. These systems are commonly called “input isolated” or “sensor isolated” 2-wire transmitters.
Related TI Designs reference designs for 2-wire transmitters:
Related 3-wire blog posts from my colleague Kevin Duke:
Find commonly used analog design formulas in a new pocket reference put together by Art Kay and Tim Green.
On Figure 4, why does the Sensor Supply need to go to the Isolator and then connect to VREG? I'm trying to drive a uC into a discrete SPI DAC into the XTR117, and I'm somewhat confused about where to place the Isolator and when it is absolutely necessary.
In figure 4 the sensor supply powers the sensor, MCU+ADC, and 1/2 of the digital isolator. The XTR116 VREG connection powers the other 1/2 of the digital isolator.
The only cases where isolation is not required are those where the sensor, MCU, ADC, and DAC are all powered solely between the VREG/IRET pins of the XTR117. If the MCU is trying to communicate with multiple DAC devices+XTR devices, or its power source has a potential relationship to the VLoop supply then you must use isolation. Take a look at the next blog in the series for more information about designing input isolated 2-wire transmitters.
In Figure 4, why does the Sensor GND and Vloop GND connects through Chassis ? Is there any problem for connecting both grounds in PCB itself ?
In our case, we have an analog voltage (DAC o/p). We need to convert this into 4-20mA loop current. The power for MCU+DAC+all other circuitry were derived from 24V Vloop supply. So GND of MCU+DAC will be the same GND of Vloop. Then our plan is to use a linear opto coupler like IL300 (Vishay) for isolation b/n MCU+DAC and XTR input side. The secondary side of the opto shall be powered by Vreg & Iret. Please give me your valuable suggestion.
Thanks in advance
Figure 4 illustrates that isolation is required if the sensor/MCU GND has a voltage potential relationship to the VLoop GND (or is directly connected to Vloop GND). If it does, then shorting the sensor/MCU GND to the IRET pin will short IRET to VLoop GND which as described in the 2nd blog in the series will prevent proper operation of the 4-20mA transmitter.
Your system design sounds appropriate because you're including isolation between the DAC+MCU and the XTR input because the DAC+MCU GND is the same as the VLoop GND. As long as the isolator doesn't require more than ~3.5mA of current then the system should work well.
One more query. I just found XTR111, in which all grounds can be same (from my initial understanding). Then I think DAC o/p can be directly connected to XTR11 and linear opto coupler can be removed. So I am confused which is the best approach. My concern is usage of linear opto coupler may leads to loss of accuracy. I am eager to hear your valuable suggestion :)
This depends on if your application is to design a 2-wire or 3-wire transmitter. The XTR111 is used in 3-wire transmitter designs which are not discussed in this blog series but are discussed in some related blogs by my colleague Kevin Duke. You are correct that you can remove the isolation if you design with a 3-wire transmitter.
All content and materials on this site are provided "as is". TI and its respective suppliers and providers of content make no representations about the suitability of these materials for any purpose and disclaim all warranties and conditions with regard to these materials, including but not limited to all implied warranties and conditions of merchantability, fitness for a particular purpose, title and non-infringement of any third party intellectual property right. No license, either express or implied, by estoppel or otherwise, is granted by TI. Use of the information on this site may require a license from a third party, or a license from TI.
TI is a global semiconductor design and manufacturing company. Innovate with 100,000+ analog ICs andembedded processors, along with software, tools and the industry’s largest sales/support staff.