Electronicdesign 4652 Xl ifd New 3

Maximize PWM Driver Resolution By Matching Output To Application

May 20, 2011
This article describes a simple technique for using a processors PWM output to produce an analog control voltage that matches the load’s input range, thus maximizing control resolution.

Most microcontrollers have a pulse-width modulator (PWM) output that can effectively drive variable output loads such as proportional solenoid valves, motors, or light dimmers. You control these loads by using the PWM basically as a digital-to-analog converter (DAC). You set the duty cycle in the PWM register (digital) and, with a couple of components on the PWM output pin acting as an integrator, you generate a dc voltage output (analog). The analog voltage is proportional to the value in the duty cycle register and ranges from the PWM’s logic low to logic high voltages.

 In many applications, however, the device this PWM driver is controlling doesn’t operate over the PWM’s full output voltage range. A motor, for instance, may not turn at all when the drive is below 2 V or a proportional valve may be fully open at 4 V. If your PWM’s logic outputs range from 0 to 5 V, some of the resolution in your PWM is wasted. With the valve, for instance, an 8-bit PWM provides 256 control settings but only about 200 would be useful.

To be able to control devices with the finest resolution possible, you can use some simple textbook analog circuitry and a little trial and error to match the PWM driver to whatever voltage range the device needs. The circuit shown in the figure has six stages, some of which you may not need and can simply leave out:

• Enabler: The switch U1 allows selection from among several devices needing control. In the applications that generated this idea, multiple valves and pumps all share one PWM line. The Enable line lets each device be turned on and off individually.

• Integrator: This stage of the circuit (R1 and C1) converts the square wave of the PWM to a dc level proportional to the PWM’s duty cycle. The type of capacitor used for C1 affects the speed at which it can charge and thus convert the PWM pulse train to a nice, flat dc level. My application found success with both electrolytic and ceramic capacitors.

• Divider: Resistors R2 and R3 form a simple voltage divider that determines the upper limit of the circuit’s output control voltage. For instance, my processor runs on +5 V and thus the PWM output switches 0 to +5 V. The valve I use is fully open when driven at about 4 V. Without this voltage divider stage, 20% of the PWM’s resolution capacity would be wasted as the PWM driver can achieve nothing useful when its output ranges from 4 to 5 V.

• Adder: There are many cases in which the low end of the PWM driver’s output isn’t useful. Motors, for instance, usually don’t start running until the drive voltage reaches some minimum threshold. To raise the starting point of the PWM driver’s output range, R4 and R5 form another voltage divider that should be set to a little bit below the desired minimum output. If your motor doesn’t start up until the driving voltage reaches 2 V, you might want to set this voltage divider so R7 sees about 1.8 V. With this scheme you would be only wasting about 0.2 V or 4% of your PWM’s control range rather than 2 V or 40%.

• Multiplier: Often the control voltage that the PWM driver produces (5 V in this circuit) is lower than the hardware’s drive needs (12 V here). A multiplier is a simple way to increase the driver’s output voltage to whatever range is required. The values of R8 and R9 determine the multiplication factor. In the figure, R8 and R9 are equal, creating a voltage doubler. For other multiplication factors, use the formula:

Vo = Vi (1 + R8/R9)

where Vo is the output of U2 and Vi is the voltage at the IN+ input of U2.

• Converter and driver: U3 here is used to convert the PWM driver output voltage to a current and to help reduce variance in the PWM’s control output due to temperature swings. Inductive loads like motors and valves, for instance, generally have a big coil of wire in them, and this wire’s resistance can change with temperature. The voltage follower U3 helps compensate for such changes. During testing of the circuit, the Texas Instruments OPA137 op amp shown proved to be very stable and to perform much better than anything else in its class. D1 is a freewheeling diode used to eliminate the sudden voltage spike seen across an inductive load when its supply voltage is suddenly reduced or removed. Both SK32 and LL4148 diodes have proven effective in various applications.

When using this circuit, optimizing the resistor values for the specific application may take a little trial and error. To simplify the process, use potentiometers so you can adjust them to find the best values, and then replace them with fixed resistors.

Once set up properly, a scheme like this can improve the fineness of control the PWM driver provides. Instead of having only 20% to 40% of the output range be useful, you can have something more like 90% for the cost of just a few inexpensive components.

Anoop’s Analysis
by Anoop Hegde

This is a good design idea that addresses the most common problems we face while using a PWM to drive load current. However, I have two concerns.

First, the IPS7091 in the “enabler” stage is an intelligent high-side switch that can carry up to 5 A. This does work. In my opinion, though, it’s overkill to be acting as a gate for the PWM signal. Also, it doesn’t have separate power pins, so it takes its power from input pins that are shown as “Vcc,” thus drawing about 3.5 mA from the PWM signal.

As you draw power from a microcontroller output (the PWM output in this case), its voltage drops because microcontroller output drivers aren’t really high-current capable. (They can supply at most 10 to 20 mA.) Using the IPS7091 will thus reduce the PWM output range even further because of this additional voltage drop due to the current drawn by this high-side switch IC. A plain AND gate (a single AND gate IC) would do the job of this IPS7091 IC and do it better.

Second, this analog method is only suitable for low-power loads of a few watts to tens of watts. Keep in mind that an analog amplifier is inefficient for driving motors, solenoid valves, and the like that have high power requirements. There will be too much power dissipation in Q1 when it acts as an analog current amplifier. It makes more sense to use MOSFETs on a PWM output so a digital waveform (PWM) can directly drive the load. An inductive load will itself act as a filter for the PWM.

Sponsored Recommendations

What are the Important Considerations when Assessing Cobot Safety?

April 16, 2024
A review of the requirements of ISO/TS 15066 and how they fit in with ISO 10218-1 and 10218-2 a consideration the complexities of collaboration.

Wire & Cable Cutting Digi-Spool® Service

April 16, 2024
Explore DigiKey’s Digi-Spool® professional cutting service for efficient and precise wire and cable management. Custom-cut to your exact specifications for a variety of cable ...

DigiKey Factory Tomorrow Season 3: Sustainable Manufacturing

April 16, 2024
Industry 4.0 is helping manufacturers develop and integrate technologies such as AI, edge computing and connectivity for the factories of tomorrow. Learn more at DigiKey today...

Connectivity – The Backbone of Sustainable Automation

April 16, 2024
Advanced interfaces for signals, data, and electrical power are essential. They help save resources and costs when networking production equipment.

Comments

To join the conversation, and become an exclusive member of Electronic Design, create an account today!