Hang around any one of our industry trade shows for long, and you’re going to hear the term VFD. Of course, a lot of you are installing VFDs to deliver constant pressure and already know that a VFD is a Variable Frequency Drive.
Hang around or read about VFDs a little longer, especially on the commercial side, and you’re going to see or hear, “our VFD uses a PID controller.” But, as a rule, no tells you what a PID controller actually is, or even what it stands for. That’s probably because PID stands for Proportional, Integral, and Derivative. That right there probably explains why no one goes any further.
But, like many things, it’s not as intimidating as it sounds. A PID controller isn’t a physical device, but a piece of software inside the VFD. PID controllers are used in tons of applications beyond VFDs, and your brain has a pretty good one built right in. You use it for just about everything that requires physical action.
For example, you’re coming up on a stoplight that just went from green to yellow. Without you consciously thinking about it, your brain determines 3 things: How far am I from the light? How long has it been yellow? And, how fast am I approaching it? These get integrated into a decision that results in the correct (hopefully) physical action.
Thinking in terms of a pump now, the job of the PID controller is simple: “How fast should I tell the VFD hardware to run the pump at any given moment?” And, like your brain, it takes the answers to 3 questions (P, I, and D) to come up with the right answer under all the different circumstances and installations.
The proportional part of PID answers the question of, “How far are we off?” That is, “what’s the difference between the target pressure and the actual pressure coming from the sensor?” On one hand, it seems like that’s all we need to know. However, as it turns out, if we only tell the pump how fast to turn based on this question, there’s a tendency to chase and constantly overshoot our target. We no longer have constant pressure.
That’s where the I and D of PID come in. The integral part answers a completely different question: “For how long has our actual pressure been different from our target pressure?” This keeps the controller from overreacting. For example, if the difference is only momentary, the integral part will say, “look, we’ve only been off for a second here; let’s see what happens before we make a change”.
The D in PID stands for derivative and looks at things from the perspective of “how fast are we closing in the target?” For example, D may say, “we’re getting closer, but it’s taking forever; we need to make more progress here”. Or, D may scream, “We’re closing in too fast! Slow things down!”
All of this gets continually weighed into a final answer that delivers constant pressure over a wide variety of changing scenarios and demands. So, just think of a PID controller as simply a feedback loop that makes adjustments based on how far am I off, for how long, and how fast am I closing in on where I need to be? Think of that the next time the light turns yellow.