Advertisement

Blog Signal Chain Basics

Signal Chain Basics #128: Optimizing solenoid control with precision current measurement

Beyond simple on/off use cases, you’ll find linear solenoids in many applications requiring precise linear motion and the regulation of pressure, fluid or air. Automotive applications include automatic transmission controls, electronic fuel injectors, or electronic stability control actuators. Non-automotive examples include critical medical applications requiring precise airflow control or industrial applications that redirect and control fluid.

Solenoid operation and control

In linear applications, the current flowing through the coil is directly proportional to the magnetic field generated; see Figure 1. This magnetic field is responsible for exerting the appropriate force on the rod or armature and causing it to move to the desired position. By knowing the current that’s flowing, you can know the position of the armature in the solenoid.

Figure 1

Typical solenoid configuration showing current flow versus armature positioning.

Typical solenoid configuration showing current flow versus armature positioning.

A typical circuit for solenoid activation is to use a high-side driver like the one shown in Figure 2. A pulse-width modulated (PWM) control signal turns the high-side switch on and off, controlling the current flow through the coil, which in turn controls the magnitude of the generated magnetic field. The control of the PWM frequency and duty cycle determines the resulting average current in the solenoid, which in turn controls the force applied to the actuator. The current is measured between the high-side switch and the solenoid. The higher the duty cycle, the higher the average current flow to the solenoid and the further the armature moves “outward.” Conversely, a lower duty cycle drives less average current, which results in the armature moving in the opposite direction.

Figure 2

Typical high-side driver with high-side current-sense solenoid control and feedback circuit.

Typical high-side driver with high-side current-sense solenoid control and feedback circuit.

Optimizing current measurement

Measuring the high-side current in this configuration presents these challenges when selecting a current-sense amplifier:

  • The PWM control of the switch generates Δ V/ Δ t “noise.” Accurate measurements of the current flowing to the solenoid require noise rejection at the inputs of the current-sense amplifier.
  • The common-mode voltage at the inputs will essentially be VBATTERY , which in some newer vehicles could be 48V.
  • The common-mode voltage could be negative (below ground) due to the inductive nature of the solenoid and the kickback created when the switch turns off and current flows in the opposite direction through the diode.
  • A high slew-rate output is required to fully swing to the rail in the case of very high or very low PWM duty cycles.
  • The solenoid has a large change in characteristic impedance caused by temperature changes. The current measurement must minimize any additional temperature-related error.

A device such as the Texas Instruments INA240 is designed for linear solenoid applications and specifically eliminates the Δ V/ Δ t noise with its enhanced PWM rejection circuitry. With a common-mode range from -4V to +80V, the device supports both high-voltage applications as well as the negative voltage generated by inductive kickback. The 2V/µs slew rate enables a 5V rail-to-rail swing in 2.5µs. This will support a 90%/10% duty cycle of a 40KHz PWM signal.

The device’s worst-case room-temperature offset of 25µV and temperature drift of 250nV/o C, combined with the initial 0.2% gain error and 2.5ppm/o C gain error drift, enable precise measurement across both a wide dynamic range of current and the full temperature range, as you can see in Figure 3.

Figure 3

Error curve for the Texas Instruments INA240 over the typical solenoid current range at both 25oC and +125oC when using an 50m
	Ω
 shunt resistor.

Error curve for the Texas Instruments INA240 over the typical solenoid current range at both 25o C and +125o C when using an 50m Ω shunt resistor.

Automatic transmission example

Let’s look at a specific application: changing gears in an automatic transmission. An electronically controlled automatic transmission still uses hydraulic control to change gears. Multiple linear solenoids control the hydraulic system by changing the amount of pressure applied to the actuators attached to the clutches. Smooth operation requires precise and repeatable control of the solenoid movement, which in turn enables precise amounts of hydraulic fluid for accurate and repeatable gearshifts.

Summary

The ability to precisely monitor the current flowing through a solenoid control circuit enables repeatable and accurate solenoid positioning. A stable overtemperature performance can save the cost of performing multitemperature calibrations during the manufacturing process. A current-sense amplifier with precise DC performance as well as excellent common-mode rejection is critical to optimizing your solenoid control circuitry.

References

  1. Prakash, Arjun. High-Side Drive, High-Side Solenoid Monitor with PWM Rejection, TI TechNotes (SBOA166A), October 2016.
  2. Texas Instruments. Getting Started with Current Sense Amplifiers video training series.
  3. Texas Instruments INA240 data sheet.

About the author

Dan Harmon is the automotive marketing manager for the Current and Magnetic Sensing product line at Texas Instruments. In his 31-year career at TI, he has supported a wide variety of technologies and products including interface products, imaging analog front ends (AFEs) and charge-coupled device (CCD) sensors. He has also served as TI’s USB Implementers Forum representative and TI’s USB 3.0 Promoter’s Group chairman.

Dan earned a bachelor’s degree in electrical engineering from the University of Dayton and a master’s degree in electrical engineering from the University of Texas at Arlington.

6 comments on “Signal Chain Basics #128: Optimizing solenoid control with precision current measurement

  1. HughVartanian
    September 13, 2017

    The author suggests that by controlling a solenoid's current one can directly infer and control its armature's position. This is something of a misrepresentation. The current through a solenoid coil is directly proportional to the magnetic field generated which is generally related to the force on the armature and is only loosely related to the position of the armature. In fact in all of the solenoids, that I have experience with, the relationship between the force and the current (& induced magnetic field) depends on the size, position and magnetic properties of the armature relative to the coil. If one has a force applied against the armature with, say, a spring, then one could do a manageable job plotting the position vs. current (& magnetic field). It certainly isn't going to be linear, particularly if the force is not constant vs position. The relationship is certainly subject to any number of variables (spring constant, mechanical and physical property variations, etc.) This may all add up to make the open loop control of the position vs current suitable for a given application, but eyes need to be wide open between the requirements and the capability of the use of the current to control position. Certainly, to make a control system that controls position with a position and velocity feedback control system, one requires a good current control loop actually driving the solenoid (or motor, whatever electromagnetic actuator). To implement a current control loop, one does need fast (10x the velocity loop bandwidth is a good place to start) and accurate feedback of the load current. That way it can respond to current commands generated by the outer control loops to get that solenoid to where it needs to go. (if one uses a permanent magnet instead of a plain iron bar, and bidirectional current feedback/control, one can move the armature in both directions!!!) (different chip needed, I believe) -Hugh

  2. forthprgrmr
    September 13, 2017

    I did this 18 years ago. See my (now expired) patent: US6249418 – easy to look up but PlanetAnalog won't let me include a url to the page. Yes you can get position by measuring the current and modelling the current/flux relationship. Not perfect, but very good.

  3. Steve Taranovich
    September 13, 2017

    @HughVartanian–thanks for you excellent addition to this topic

  4. Steve Taranovich
    September 13, 2017

    Readers can enter this link into their browser for the patent page https://www.google.com/patents/US6249418

  5. DHarmon
    September 15, 2017

    Thanks Hugh for clarifying and adding to my article. Regards, Dan

  6. DHarmon
    September 15, 2017

    Thanks for reading and highlighting your patent. Regards, Dan

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.