Ron White has famously said, “You Can’t Fix Stupid”. And he’s right, of course. We all do stuff that’s a little crazy and generally, we live long enough experience our own stupidity, and hopefully correct it and go on to do much better things.
The motion control industry has demonstrated “doing stupid” at a variety of levels. I will mention a few. For those readers who create products for the industry, I humbly apologize in advance if you find anything offensive in my comments, but honestly, some of what goes on defy common sense consideration. My hope is that some of my rantings will inspire people to make changes that will benefit us, and our customers, after all.
Motion control and machine control are based on the discipline of more than two centuries of mechanical innovation. Modern motion control and machine control seeks to emulate electronically all the various mechanisms of the Industrial Revolution. So it should come as no surprise that all of the fundamental rules of controlling machinery are mechanical in nature. For the most part, these rules are simple mechanics, higher order analysis is helpful but rarely required.
What makes this “stupid” is that few motion controllers are designed by mechanical engineers. Of course, this is entirely plausible. ME’s don’t write software and don’t deal in power electronics. Software engineers are largely mathematicians. Good software creation is about describing a system in mathematical terms, writing the code for a description on a certain hardware platform, and then verifying that the description is valid to a high degree. So the question is “How much motion control software is created by ME’s who are expert in motion?” Answer; practically none. That’s Stupid.
If the software is not created by a domain expert, you are going to have problems. I can attest by direct experience that this is the case. Better still, a friend of mine is the guy that created the original PID code for Rockwell PLCs. He agrees with my perspective: PID is not particularly well suited to motion control, particularly position control. The model is incapable of describing the problem adequately and always leads to shortcomings that have to be programmed around. That’s Stupid.
How about the notion that motor designers are ME’s and have little or no background in electronics. But an electric motor requires electricity to operate, and an embedded processor and power semiconductors to control it. The state of today’s motor market is that we do not have subject matter experts in magnetics, electronics, software, and mechanics collaborating to make the world’s best motors. That’s stupid.
There are much better ways to describe motion control that will simplify the user interface and make the functionality of software-based control much more effective for motion control users. Let’s get all the subject matter experts together in a room when we work on next-generation motor/drive and machinery projects.
Let’s fix Stupid, and make everyone more productive and profitable in the process.