Fundamental Concepts for Precision & Throughput in Laser Processing Machines

Most everyone familiar with large-scale industrial laser processing has seen a high-throughput laser CNC machine whirling around, cutting out large sheets and tubes of steel or aluminum with dizzying speed. Those of us involved in laser microprocessing, where part quality depends on micrometers of processing accuracy, wonder if we can achieve such feats of machine throughput and still produce high precision parts. The answer is yes – and the question then becomes “how?” This article explores fundamental considerations in machine design and control that one must be familiar with to get the most throughput possible out of a precision laser microprocessing machine.

The criteria for determining an acceptable part is generally non-negotiable in a manufacturing process. Part tolerances are established by the requirements for a part to function properly or safely. They dictate the allowable error budget of the manufacturing process. The error budget is then “used up” by different error sources stemming from a machine’s design, its controller capabilities and the laser material interactions during processing. The key to achieving maximum throughput while manufacturing high-precision parts is to leave as much of this error budget as possible for dynamic tracking errors. Following sound system and structural design principles and choosing a highly capable motion controller – one that takes the most advantage of the dynamic tracking error budget – will maximize the throughput and therefore the economic justification for a laser micromachining system.

A manufacturing system’s structural design is foundational to increasing its ability to perform in high-throughput operations. For a control system to reject and minimize errors, the sensors it uses to “see” movement within the system must be able to observe relative motions between the tool and the part. In most systems, these sensors are not directly observing the tool tip’s motions, i.e. the laser spot, and instead their information comes from an optical read head looking at an encoder scale (in effect a ruler) embedded within the motion system mechanics. Therefore, to save as much error budget as possible for the dynamic tracking budget in the controller, designers must minimize unobservable errors resulting from bending or fibration within the machine frame. The key to minimizing unobservable errors is to maximize the structure’s stiffness. One way to achieve maximum stiffness is to minimize the length of the machine’s structural loop. A structural loop is the path through which forces produced by machine movement get to ground or to their equal and opposite forces produced in corresponding structural elements. Imagine the material making up the machine’s structural elements as an amalgamation of thousands of microscopic springs linked in a serial chain. Adding more springs to the serial chain actually lowers the chain’s stiffness. Therefore, a designer should shorten the structural “chain” of spring elements to stiffen the machine. Moreover, adding spring elements in parallel makes the chain stiffer. For maximum stiffness, a designer should add structural elements that redundantly support inertial forces to the machine’s frame. The stiffer the machine, the more energy that can be injected into its structures without causing unwanted movement. This allows a user to push the motion control elements faster, with more acceleration and energy, while minimizing unobservable processing errors. Figure 1, below, depicts machine structural loops and spring elements in series versus in parallel.

A stiff machine that allows for injecting more energy into it without bending as much, saving more error budget for other places, is a straightforward improvement. This paves the way to the next area of focus for improving throughput: the principles of machine dynamics. As the motion platform and machine frame become stiffer, their natural frequency also increases. As their natural frequency increases, so will their controllability and general ability to produce stuff faster.

Every motion trajectory – the path needed to be taken by a laser spot to manufacture a part – has spectral content for each axis involved in producing the motion. The commands to each axis have a certain frequency bandwidth of sine waves required to represent it with a mathematical series or summation. Figure 2, below, shows an example of a step function and its approximation using limited bandwidths of sine waves.

In this step function example, an infinite bandwidth is required to perfectly approximate the step, which makes it impossible to actualize in a real machine. This is one primary reason motion programmers do their best to avoid discontinuities in the commands they send to the machine. The principle demonstrated in Figure 2 holds true for every command signal. When motion profiles are multidimensional, involving more than one axis of motion, the speed with which the machine traverses that profile changes the bandwidth of commands sent to each axis involved. A simple example of this relationship is when two axes are used to create a circle. As learned in basic trigonometry, two axes traversing a circle experience a sine wave in position, velocity and acceleration. The frequency of the sine wave each axis is asked to perform is proportional to the speed with which the circle is traversed. The faster the machine needs to traverse a circle, the higher the frequency of sine wave each axis involved must be able to perform in position, velocity and acceleration. For any axis of motion to do the provided command profile, the bandwidth of that profile must be inside the motion system’s bandwidth. That’s right – every motion system has a bandwidth, too.

Control systems rely on feedback signals, servo control loops and powerful motors to react to commands and make real outcomes match desired outcomes. The speed with which a control system can react is based on the rate at which the controller can make decisions and effect change when actual motions don’t perfectly match the commanded motion. This “control system reaction rate” is almost entirely a function of the specifications and design of the control products used. Specifications like trajectory generation rate, the current loop closure rate (speed at which the current produced by a given motor drive can change) and the peak force producible by the equipment’s motors will determine the control system’s reaction rate. It therefore benefits designers to select capable control products and powerful motors, a somewhat obvious conclusion. However, the control system’s reaction rate is only part of the overall motion system’s ability to respond to commands, i.e. the motion system bandwidth. The combination of the motion platform’s physical stiffness and the control system’s bandwidth determines the dynamic capability of the overall system. The higher the mechanical system’s natural frequency is, i.e. the stiffer it is, given the same control system and motor, the larger the bandwidth of frequencies to which the system can successfully respond.

Generally, the signal of prime importance in motion control is the acceleration command. Acceleration is the primary signal of interest to machine operators because it is most closely related to what the machine controller is actually controlling, which is electrical current to the motors. The current sent to the motors of each axis is proportional to the force produced by each motor. The force produced by each motor is proportional to the acceleration experienced by that degree of freedom in machine movement. Tracking error, or error injected into the production process by the motion system’s inability to follow the commanded trajectory perfectly, is proportional to the portion of the commanded acceleration’s bandwidth that is above and beyond the motion system’s bandwidth. A car that is only capable of traversing a race track so fast, based on its suspension and engine and driver, will go off the road if forced to go faster around the turns than its limits allow. This is the same for a laser processing machine. By understanding the bandwidth of acceleration commands being sent to the machine in the motion profile as well as the bandwidth of the machine’s reaction capability, or dynamics, we have a firm footing for ensuring quality part production at maximum throughput rates. Some advanced motion controllers actually provide features that allow programmers to automatically take into account the motion system’s bandwidth and self-limit the acceleration commands being sent to the machine elements to prevent excess errors from ever occurring.

Combining these concepts creates a meaningful message for machine designers. The stiffer the machine frame’s structure, the less the process results will be affected by machine bending and vibration, leaving more error budget on the table for dynamic tracking errors. The stiffer the motion system’s mechanical design, the higher the motion system’s bandwidth will be. The higher the capability of the control products used, the higher the motion system’s bandwidth will be. The higher the motion system’s bandwidth, the larger the bandwidth of acceleration commands it can respond to while producing the same level of part errors. The higher the bandwidth of allowable acceleration commands without making bad parts, the faster a machine can be commanded to traverse a desired profile during part production. Thus, machine designers should look at every possible way to maximize machine stiffness and control system bandwidth to maximize process throughput without impacting part quality.

An expert motion control partner can help estimate the bandwidth capabilities of different motion systems and provide guidance to designers for selection of control products and mechanical stages alike. Contact Aerotech to learn more about the impact machine design decisions will have on throughput in your next laser processing machine.