Authors: Lawrence Jongi, Sandeep Raju, Jakob Petri
Introduction
High accuracy and high cutting speed are critical factors for the economic success of a metal processing company. Both elements can influence the level of customer satisfaction tremendously, on the one hand by determining the workpiece quality and surface finish, on the other hand by influencing production costs and delivery times. To achieve an optimum of both, partially competing goals, a sophisticated machine tool design is needed. This paper aims to contribute to the solution of this issue, by focussing on the design of a specific, but decisive part of the tool, the feed drive system, and the selection of its components.
State of the art – Design of a feed drive system
This paragraph states out the status quo of machine tool designing, by considering the different machine elements, components, engineering principles and theories, which have already been in use in an industrial environment. The following figure 1 illustrates a typical 2 Degree of Freedom (2DOF) linear speed drive system, on which this paper is going to elaborate. Specific considerations in the design process will be explained and supported by examples.
Figure 1: Biaxial Feed Drive System (Dr. Eng. Toyohashi 2013)
Stress/Strain/Forces
There are many factors that can help or hinder a parts performance (temperature, moisture, dust/debris, impact, range of motion, etc.) A machine element that works great in the snow could fail in the desert, and these varying conditions will need to be accounted for. (Ryan 2019). During cutting temperatures can be as high as 1100 0C of which the cutting tool draws away 10%, the discarded chip carries away 60%-80%, the work piece act as a heat sink drawing 10% - 20% of the heat away (S 2016). The tangential cutting force will cause bending moments on the lead screw, the lead screw also experiences tensile and compressive stresses simultaneously due to the weight and the drive forces. However, the vibrations experienced from the tool will cause some dynamic loading on the lead screw and the bearings. The bearings experience pronounced compressive stresses and some shocks, pitting and all these coupled with temperature change will cause creep, pitting, abrasion, and fracture.
Know your Materials
With a keen understanding of the stresses at play in a particular environment, the next decision in the machine design process is what material to use in manufacturing a part. Certain metals are more resistant to corrosion and warping. Stainless steel 316 provides high tensile strength at elevated temperatures. It is more resistant to corrosion and pitting than either grade 304. Due to its composition, grade 317 stainless steel is a more expensive alloy than most other 300-level grades. This makes it the best fit for the lead screw, and the table can be made from lower grades resistant to abrasion e.g., Attractive, and malleable, grade 301 features high strength and corrosion-resistant composition. Since it is highly malleable and abrasion resistant. This makes the machine tool more durable since this part experience abrasive movements. We propose the ball bearings be made from 316 because of its resistance to corrosion and the female of the lead screw and the bearings to be made of 400 because of its higher strength and wear resistant though weak on corrosion, which will be counteracted by lubrication of the moving parts (Spira 2021).
Quality and Standardization
As much as customers’ satisfaction is concerned, they want value to be derived from their product which they have purchased with specifications like durability, market price, energy efficiency, speed of the spindle drive, speed of the lead screw. The overall ISO
standard 9001 for quality will also be handy.
Design for Accuracy
The accuracy of the machine depends on the capabilities of the servo controller. Factors such as friction force; ripples; spikes; machining force, position of the workpiece. Therefore, in order to have an acceptable transient response and disturbance rejection properties, a (2DOF) proportional integral- derivative (PID) controller will be employed for each axis. To compensate disturbances and machining contour errors, the utilization of filter observers to remove disturbances or noise, neural networks (NN), cross-coupled controllers (CCC) can be employed.
Specific Engineering Parameters
Maximum speeds of each axis, maximum acceleration of each axis, maximum weight each axis and individual weights of each axis, continuous force per motor, peak force per motor, continuous current, peak current, inductance, maximum bus voltage. It is critical to understand that all these parameters need to be analyses when the machine is static and dynamic. However, there other parameters which are at play, that is, operating parameters (parameters which are affected by other parameters), these include: temperature, vibration, and noise. These parameters are very important for condition monitoring and prognosis and prescriptive maintenance. The image below is a schematic diagram on the condition monitoring setup.
Figure 2: Connections within the set up (Heydarzadeh, M. S 2016)
Testing
This includes training data sets, to model friction and force ripples, static neural networks with tapped delay lines in their inputs will be utilized in the design. Since both friction and force ripples are functions of the displacement and velocity, time series of the translator displacement will be considered as inputs to the network. The training inputs should cover all the range of working conditions. Cross coupling controlling of both axis is an important strategy to reduce contour errors instead of tracking errors of each axis. (Heydarzadeh, Rezaei et al. 2016).
Figure 3: Cross Coupled Controller (Heydarzadeh, M. S 2016)
To achieve the common criteria in machine tools controller design, a 2DOF modified PID controller should be designed. The benefit of such a controller is to achieve both appropriate transient response and excellent disturbance rejection at the same time. The schematic of such a controller is shown in figure 4.
Figure 4: Proposed Controller (Heydarzadeh, M. S 2016).
The zero- placement method can be used to design this controller and resulted in a PID controller in the feedforward path and a derivative in the feedback path resulting in a PIDD controller.
Discussions and proposals
A holistic data analytics approach (Juergen Lenz)
With the advent of data analytics in manufacturing and machine tools, data-related technologies like the Industrial Internet of Things (IIoT), sensor networks, and cyber-physical systems have supplemented enormously the available information in this field. But this diverse analytical landscape exists among silos between individuals and experts, leading to redundant processes and inefficient exchanges between the different domains. Another observation within a typical manufacturing organization is the current information and decision-making exchange setups between intra-departments. This restriction limits the decision-making capabilities and the bundling up of analytics objectives across different departments and functions at the production line, factory, or even the supply chain level. The approach toward data and analytics should be holistic to tackle the above-discussed gaps.
Figure 5: Holistic data analytics approach (J.Lenz et al., 2018)
Real-time prescriptive analysis
The characteristics of machine tool data are the nonpermanent volatile nature of each variable (or parameter) value, continuously overwritten with each cycle of the controller, only valid at one specific snapshot of the process, and generally accurate. This data from these sensors can be monitored in real-time to increase the accuracy of tool positioning and deflections.
Figure 6: Multi-Domain Machine Tool Data Analytics (J. Lenz et al., 2018)
Conclusions
We have seen various approaches to increase the tool-feed drive system accuracy and speed and covered aspects of design, forces, materials, and engineering parameters. Also covered are two proposals in data analytics that can be incorporated to increase the tool speed and the overall accuracy. We can conclude that using known skills like retrofitting and restructuring processes, the accuracy and speed of a machine tool can be substantially increased.
References
Heydarzadeh, M. S., S. M. Rezaei, N. A. Mardi and A. Kamali E (2016). "Motion control of a two-axis linear motor-driven stage in the micro-milling process." Proceedings of the Institution of Mechanical Engineers, Part E: Journal of Process Mechanical Engineering 232(1): 65-76. (last accessed: 09/17/2022)
Juergen Lenz, Thorsten Wuest, Engelbert Westkämper, (2018), "Hollistics approach to machine tool data analysis”, https://doi.org/10.1016/j.jmsy.2018.03.003, 180-191, (last accessed: 09/19/2022)
Ryan (2019). "Types of Machine Design and Design Basics." https://randrmanufacturing.com/blog/types-of-machine-design-design-basics/.S, A. C. (2016)., “<Measuring of Cutting Temperature During Machining.pdf>." (last accessed: 09/16/2022)
Spira, N. (2021). "A SHORT GUIDE TO GRADES OF STAINLESS STEEL." https://www.kloecknermetals.com/blog/a-short-guide-to-grades-of-stainless-steel/ (last accessed: 09/17/2022)
Y.Altintas, A. Verl, C. Brecher, L. Uriarte, G. Pritschow, "Machine tool feed drives”, (2011), https://www.sciencedirect.com/science/article/abs/pii/S0007850611002125?via%3Dihub, 779-796. (last accessed: 09/19/2022)