Nuts and Volts - September 2008

standalone software. This means you might have multiple programmers connected to your computer's USB ports, along with the MPLAB IDE, and the PICkit 2 stand-alone applications all running simultaneously; so this could cause confusion on which tool is being used by the MPLAB IDE and which one is used by the GUI standalone application. When you open the stand-alone software, the screen will ask which PICkit 2 you want to use, as seen on the right side of Figure 6. I have two PICkit 2s connected to my PC. Notice also in Figure 6 that they have unique names. You can give each PICkit 2 a unique name through the "Calibrate VDD and Set Unit ID…" selection under the Tools menu. This is the same menu where you select the UART or Logic Tool. This Unit ID name will then get stored in the internal PICkit 2 memory so it will stay with the unit. I also put a small piece of tape with the number on the programmers case, so that I can keep them straight.

To use multiple PICkit 2s with the MPLAB IDE, you need to update to the MPLAB software version 8. 15, which was recently released. You may also have to download the latest control software, depending upon how old your current version is. The MPLAB IDE will ask you which PICkit 2 it should work with. That window is also shown on the left in Figure 6. Therefore, I can easily have three PICkit 2s for many projects. One is used for programming and debugging from the MPLAB IDE, one for UART communication, and one for a logic analyzer. You can buy three of these for far less than the cost of an oscilloscope. If you need to monitor more than three signals, then you can use even more of them in Logic Tool mode.

The Logic Tool has cursors built in, so you can measure the pulse width. As you can see in Figure 5, the X cursor is blue and the Y cursor is purple. You use the left mouse button to set the X position and the right mouse button to set the Y position. The screen shows the time difference between them. The time measurement in Figure 5 is the pulse width of our PWM signal, which measures 20. 8 milliseconds — not 20, as the PAUSE command created. The extra 0.8 is from the delay in the HIGH and LOW commands. I used Channel 3 on the PICkit 2 connector and the trigger was set to a rising edge on Channel 3 in the Trigger portion of the Logic Tool window. The red trigger line shows that the signal was actually captured on the rising edge. As you may have noticed, the Logic Tool is a great addition to the PICkit 2.

HARDWARE PWM

Now, back to the PWM explanation. Driving a PWM signal is easy, but most of the time you’ll want to do other things in your program, in addition to creating a PWM signal. As soon as you try to do something else, you will be affecting the signal timing. This is why many PIC MCUs have a hardware PWM built in. Some have more than one. The hardware PWM runs on its own in the background once it is set up, allowing you to write code to run separately in your main loop. You can modify the setup for a particular duty cycle or frequency in your main loop, but the hardware PWM will continue running even when your main loop is running a PAUSE command or some other delay section of code.

The PIC16F690 included with Microchip’s PICkit 2 Starter Package has a hardware PWM port on-chip. It’s called the ECCP or Enhanced Capture Compare PWM. The datasheet has a detailed explanation but, based upon reader feedback, the datasheets tend to confuse the beginner. I’ll try to simplify it here. Figure 7 shows the key diagrams I captured from the datasheet to help me explain the hardware PWM setup.

The block in Figure 7 labeled FIGURE 11-3 is the block diagram of the hardware PWM circuitry. Though the details are small, you can hopefully see that the operation is based on the TMR2 (or Timer 2) within the PIC16F690 MCU. I added the block labeled FIGURE 7-1 because this shows a key component left off of the PWM block diagram — the TMR2 prescaler. This divides the oscillator down, prior to feeding Timer 2 with its heartbeat.

The block labeled FIGURE 11-4 shows the definition of the PWM output. You can see the diagram definitions are similar to Figure 2, described earlier. The details to note in this block are the equations for the end of the pulse width and the end of the period. When the TMR2 value equals the PR2 register value, the signal starts over. Therefore, the PR2 value determines the frequency of the PWM signal. The block labeled EQUATION 11-1 shows the formula for calculating the PWM period. It’s a straightforward equation:

PWM Period = (PR2 + 1) * 4 * Tosc * (TMR2 Prescale Value)

SETTING THE PWM FREQUENCY

■ FIGURE 5. PICkit 2 Logic Tool.

■ FIGURE 6. PICkit 2 selector windows.

Tosc is the period of the oscillator you are using to run the PIC16F690 MCU. Tosc is equal to 1/Fosc, where Fosc is the oscillator frequency at which the PIC is running. PWM period is also 1/PWM frequency. You will typically know the frequency you want, but not the period. Therefore, reworking the formula to solve for PR2 is more useful: