Building the best motion system for your needs can be a real challenge. But there are a few key things to consider when designing a system that will ensure you have the best motion system for your application.
1. Stiffness
First thing’s first. How stiff is your system? The primary cause of poor motion control is that your system is not stiff enough. Most engineers learned to meet the desired technical specifications while using the minimum amount of material to keep costs down. This leads to a beautifully optimized design for commodity products, but if your production requires a hefty duty-cycle and high precision, you need to consider other things.
One thing our engineers have learned is that a stress analysis is hardly ever necessary when designing a robust motion control system. In our designs we’re not talking about a typical factor of safety of two… it’s more like TEN to get a “stiff” system. This may sound ridiculous, but we hardly ever look at failure due to structural issues because we are focused on extracting the most motion potential out of the system, which requires a very stiff system.
If you don’t know how to preload a system, then you don’t have the right mindset. If you want the system to be able to handle disturbances, you are going to need a very tight system.
2. Tolerance
The next most common motion control system design problem is tolerances. If you think you can go to any machine shop and get the tolerances you need… then think again. The tolerances you need exceed what an average machine shop can give you. You will need a very talented machinist and world-class machines to get the tight tolerances you need.
“Which tolerances?” you ask.
Well, there are hole dimensions and straight and flat surfaces. Even the tiniest bit of slop in your system will create major problems later on. While it may seem expensive to get the precision your application calls for, it will be far more expensive to waste time and money on parts that won’t ultimately meet your specs.
3. Dynamics
Did you have a professor that jumped up and down during Dynamics 101 to emphasize a point? Well, I did, and guess what I remember? Dynamic forces are always greater than static forces. When bridges fail, the reason is not that the static load is exceeded, but rather that the dynamic load is exceeded and the system isn’t designed to handle a frequency that it eventually faces, like wind exciting the bridge at a resonant frequency (or people – like my professor – jumping up and down on the bridge).
4. Repeatability
We’ve all heard the adage, “If you want to succeed, try, try again.” But if a machine or motion system is trying and trying again to get to the same point repeatedly and failing… then you have a problem. You need to consider a system that will move back to the same location again and again and again every time.
Consider the things that cause differences in different directions. This could be the bearing design or errors in the surface, among other things. If your system deviates in some way, and you are not using precision bearings like recirculating ball bearings or air bearings, then your machine will not get back to the same location each time.
5. Thermal Expansion
It’s getting HOT in here. Did you consider the environment in which you will be using this system? Or did you consider how the machine will heat up with a high duty-cycle?
Thermal expansion plays a role in designing a motion control system. If you designed the system for room temperature, but your machine consistently goes beyond that, you could end up with more error than you thought you would. Remember that heat generally makes things expand, so it is important to consider the environment in your design.
6. Controls
Lastly and most importantly is how you handle errors when you get them. You may be able to design a tolerably stiff, dynamic, repeatable, and thermally stable system, but you will inevitably fail to get your system to move the way you want if you aren’t using the correct controls. If you have a feedback control like PID, a very popular algorithm, you will be able to move the system to a point and the system will automatically adjust itself until it actually reaches this point.
Maybe you can get away with designing a PID control for your system, but if you are trying to achieve the lowest total cost of ownership, you will fail without advanced controls. By the time you manage to develop these tools, you will have equivalently earned your PhD in Control Design. On top of that, you will still need to learn how to tune your system.
A finely tuned system will give you the ultimate advantage. Easy-to-use, yet powerful control tools can help you get there.
Some of Aerotech’s control tools will make your mouth water. Like tuning with Autotune – just press a button and the system determines the correct gains for you. Or optimize even more with a bode plot, a phase and gain plot that allows you to use simple up and down buttons for more phase and gain margin (more stability in the system and more ability to reject disturbances), or automatically fit filters wherever you want.
Even more tempting are advanced controls like Harmonic Cancellation, Command Shaping, Enhanced Tracking Control, and Iterative Learning Control that will take you from novice to (looking like) an expert in tuning in under an hour. Look for controls that will make YOU look good without all of the effort.