In part 5 of this blog series, I discussed the construction of an input-isolated 2-wire sensor transmitter. Input-isolated 2-wire transmitters require isolating the sensor power and data signals from the 2-wire loop supply and 2-wire loop return (RTN). The main challenge when creating input-isolated or other complex 2-wire sensor transmitters is keeping the total current consumption from the loop to less than 4mA, a requirement I explained in part 3. One way to increase the available current in a 2-wire transmitter is to use a switching step-down DC/DC converter. The step-down converter must be optimized for low-output current applications so that it can operate at high-enough efficiencies to be useful in the system. Figure 1 is a simplified block diagram of such a system.

Figure 1: Simplified circuit of 2-wire sensor transmitter with step-down DC/DC converter to increase current available for the sensor


With a step-down power converter in the system, the maximum sensor current, ISENSOR, can be increased based on the converter efficiency, η, the output voltage, V­­OUT, the input voltage, VIN, and the maximum IRET current of 3.8mA, as defined in Equation 1:

Assuming an efficiency of 80%, an output voltage of 3.3V and an input voltage of 8V, the maximum sensor current can be increased to 7.36mA, an increase of almost 100% compared to the original maximum current of 3.8mA.

You must take special care when designing 2-wire transmitters with switching converters because the high-impedance loop power supply can negatively interact with the incremental-input impedance of the switching converter, causing loop stability and compliance issues (as explained in this DC/DC converter reference design. Therefore, isolated and nonisolated step-down converters for 2-wire transmitters are commonly designed with unregulated fixed-frequency charge pumps (such as the design shown in part 5 and explained in detail in these two reference designs:

1)      Uniquely efficient isolated DC/DC converter for ultra-low power and low power applications reference design (TIDA-00349).

2)      Isolated ultra-low power reference design for 4 to 20 mA loop-powered transmitters (TIDA-00167).

So to summarize, you can add switching step-down power converters to 2-wire 4-20mA sensor transmitters to increase the available sensor current. As I mentioned, you should construct the design carefully to verify that the operation of the step-down converter does not negatively interact with the input protection circuitry or impact the stability of the 4-20mA control loop.

This is the final post in the 2-wire transmitter series, but in future posts I’ll discuss 3- and 4-wire transmitters, expanding on this series as well as the information in fellow blogger Kevin Duke’s posts.

Additional resources:
    • Related TI Designs reference designs for 2-wire transmitters:
      • Bridge Sensor Signal Conditioner with Current Loop Output, EMC Protection (TIPD126).
      • Low Cost Loop-Powered 4-20mA Transmitter EMC/EMI Tested Reference Design (TIPD158).
      • Isolated Loop Powered Thermocouple Transmitter Reference Design (TIDA-00189).

    • See all the posts in our 2-wire transmitters series.
  • Related 3-wire blog posts from my colleague Kevin Duke:

    • See an overview of analog outputs and architectures.

    • Read about the evolution of 3-wire analog outputs.

    • Find commonly used analog design formulas in a new pocket reference by Art Kay and Tim Green

  • Hi Benjamin,

    Your comments don't appear to be related to this blog.  I saw you also posted similar questions on the forum.  I've answered your questions there.

  • Also, does Rload and Rshunt the same thing?

    In my understand, Rshunt is used as sensing the current,  Rload can be either sensing or rising up the potential so it will not the same as Vloop(-), so, can I put Rshunt any series place of the loop, as long as there is a Rload connect to the Vloop(-)?

  • Hello Colin.

    I found all the  4-20mA receiver's Ref circuits is taking the method of 'Low-Side Measurements', which connects R_shunt  to Vloop(-).

    (Ref to page 3,   )

    While in my design, I use the 'High-Side' method to ensure the load is shorted to GND, so R_shunt is connected to Vloop(+). I am using a 24.9ohm resistor  and an AD629 to sense the loop current.

    Now I'm uncertain about  my design, is it fit or unfit to most of the industry's  4-20mA transmitter? Is there any fatal defects or disadvantages? To design a transmitter, what factor(s) shall be considered?


    Benjamin Gao

  • Hi Rob,

    Thanks for reading!  The TIDA-00349 reference design includes more information about the incremental input impedance differences between a typical inductor based buck converter versus a simple charge-pump buck.  This the main reason that inductor based buck converters typically have stability issues in these systems if there is a current limiter or other protection circuitry in the transmitter loop.  

    We know that applications like the one described in this post need a higher voltage solution than the solution presented in the TIDA-00349 or TIDA-00167 reference designs. I checked with our charge pump team, and our current portfolio of charge pumps are focused on standard single battery sizes and not the higher voltages required for this application.  However, there a few products available on the market which are specifically meant for this application.  While many sensor transmitters are input-isolated (as described in Part 5 of this series) this application does not require isolation and can be achieved with a non-isolated buck charge-pump.  For isolated designs, the two reference designs I linked to include a recommended transformer in the BOM.

  • Hi Colin

    This is definitely of interest to me. I want to make a 2 wire transmitter with a sensor that doesn't need isolation, and needs 5.5mA @ 5VDC regulated, using XTR11X. I would also have a low power micro, A/D, D/A, total budget 500uA.

    I have failed twice using regulated inductor buck devices, stability issues just as you say. So an unregulated buck charge pump, post linear sounds great. The problem is all the charge pumps I see are low input volts, no help at 28VDC, many so-called bucks are rail splitters with limited drive for all source, no sink.

    Your note is largely isolated apps, if that's how it must be, do you have a suggested SMT transformer.


    Rob Taylor