Vol.:(0123456789) Production Engineering (2025) 19:89–99 https://doi.org/10.1007/s11740-024-01296-w ORIGINAL PAPER Active runout compensation using the guide elements on metal band saws for a longer tool life and reduced material loss Tobias Tandler1  · Hans‑Christian Möhring1 Received: 9 January 2024 / Accepted: 17 June 2024 / Published online: 5 July 2024 © The Author(s) 2024, corrected publication 2024 Abstract At a time of rising energy and material costs, manufacturing process efficiency is becoming increasingly important. This is one reason why cutting and especially sawing processes, usually the first step in most manufacturing chains in discrete part production, have to be investigated more intensively. Due to problems with runout and poor surface finishes, raw material is conventionally cut to oversize by either circular sawing or band sawing. This oversize has to be removed by following processes, costing extra energy and wasting material. Since the problem of runout increases with increasing tool wear due to higher deflections of the thin and compliant tools, an even larger oversize is required. This paper describes an approach to reduce the need of oversize even with increasing tool wear in band sawing by tilting the saw band in order to compensate for tool deflections during cutting. To achieve this, it is necessary to measure and understand the saw band runout. The next step is to present a system design and controller for tilting the saw band. Finally, tests are carried out to analyse the effectiveness of the system. In addition, the approach allows the use of increased process parameters to the end of the tool life without losing more material. The tool can therefore be used productively for a longer time period. Keywords Band sawing · Adaptive process control · Tool life · Metal cutting 1 Introduction In today's manufacturing process development, which is characterized by digitization and automation increasingly, sawing processes play only a subordinate role, although they often are the first work step in the manufacturing pro- cess chain in discrete part production. In the course of ris- ing material costs, however, the sawing process is becom- ing more important. In circular sawing and band sawing, research and optimization focus on approaches to increase the economic efficiency of the machining process by means of a resource-saving production method [1]. A high potential for reducing material consumption can be seen in band saw- ing. Here, significant deviations between the actual and ideal cutting direction occur (Fig. 1). These deviations are caused by deflections of the thin tools and a resulting runout of the saw band, which increases with increasing saw band wear. This runout predominantly occurs in the direction of the passive force, i.e. transverse to the feed and cutting direction [2, 3]. As a consequence, the effective length of the sawn and separated material is affected. Therefore, an appropriate oversize of the separated material must be provided in pro- cess planning. This oversize has to be removed and the part length must be corrected in subsequent production steps. This conventional method causes an additional amount of energy consumption and wasted material. Considering the larger deflections and higher deviations with progressing tool wear, an even larger oversize is required. In order to avoid this, in practice an early tool change (e.g. at the begin- ning of each working shift) is usually applied though the limit of the useful lifetime of the tools is not reached. This leads to unnecessarily high tool re-grinding efforts and an inacceptable consumption of saw bands. Thus, a reduction of the tool deflections and tool runout provides a high potential regarding energy and resource efficiency. Many efforts have been made to review and improve the efficiency of the machining process. One method employed involves the determination of energy consumption during machining, which is contingent upon the tool condition and shape [4]. * Tobias Tandler tobias.tandler@ifw.uni-stuttgart.de 1 Institute of Machine Tools, University of Stuttgart, 70174 Stuttgart, Germany http://crossmark.crossref.org/dialog/?doi=10.1007/s11740-024-01296-w&domain=pdf http://orcid.org/0000-0002-8070-1047 90 Production Engineering (2025) 19:89–99 Vibrations are widely acknowledged as a significant concern, with a considerable portion of earlier literature dedicated to this subject [5, 6]. Recent studies adopt con- temporary sensors and methodologies for the analysis and detection of vibrations, particularly through acoustic features [7–9]. These investigations have yielded insights into chat- ter and have led to the development of methodologies for its analysis and evaluation [10]. In the field of circular saws, the influence of guides and their design as well as the effect of damping slots and stresses introduced in the steel blade with regard to cut devi- ation and vibrations in the process are being investigates as part of various research projects [11–14]. Other research works in the field of band saws leveraged machining forces to assess the process. In this pursuit, a diverse array of measurement systems has been subjected to examination and evaluation, culminating in the recogni- tion that the process forces provide the most informative data [15]. Nonetheless, achieving a square and precise cut is regarded as one of the primary challenges when using bime- tallic band saws [3]. Minimizing deviations is pivotal for achieving favourable outcomes and reducing losses, thereby directly influencing the productivity of the sawing process [16]. In this context, the tension of the saw band exhibits an inversely proportional relationship with the cutting profile [2]. Additionally, the side clearance also exerts an influence on the machining process [17]. This clearance characterizes the extent to which the cutting channel exceeds the width of the saw band's base material and is achieved through specific setting techniques and diverse tooth profiles. Tool displacement during the process also plays a role in other processes, so that there are already various studies on simulation, decimation and compensation. These are mainly concerned with surface qualities and positions of features. In the field of deep drilling, there are various studies on the topic of the centre course, how it can be measured and compensated for. For example, tools with automatic com- pensation are already being developed in the BTA (Boring and Trepanning Association) deep drilling process. The cen- tre course is measured by ultrasonic sensors [18]. In the field of milling, particularly in finishing pro- cesses, research is being carried out to predict tool deflec- tion in order to achieve better surfaces and precision parts. To this end, a number of studies have been carried out to analyse the deflection of milling tools in process, the influence of cutting parameters and to validate measure- ment [19–21]. With information like that a model has been developed to calculate the deflection correlated to the actual tool position. It also takes into account the tool holder and the machine structure [22, 23]. Tool displacement when working with a robot is also an important area of research due to the low rigidity of the robot compared to a standard milling machine [24]. There is already a publication on the subject of band saw blade wear and the influence of parameters on this [25]. In addition, various methods for measuring wear on band saw tools were presented and discussed in a poster presentation [26]. When used in horizontal band sawing machines, the band saw blade is clamped, propelled and guided by two wheels and, with the help of guiding devices, twisted by up to 90° in the feed direction. In this way, the saw band circulates outside the processing area of the sawing machine without affecting the usable workpiece length. In addition, the guid- ing devices also serve to stabilize the saw band. For this, they are located as close as possible to the workpiece. Fig- ure 2 shows the principle of band sawing with its typical kinematics in the feed and cutting direction as well as the guides of the saw band. In order to measure the course of the saw cut and the behaviour of the saw band, a measuring method using eddy current sensors has been developed in previous works [27, 28]. Several contactless displacement eddy current sensors are arranged one above the other in order to measure the lateral displacement of the saw band in different heights and, thus, a tilting of the saw band can be measured and determined. In order to compensate for the tool deflection during the cutting process, initial tests have already been carried out with a change in the saw band tension [16]. However, this puts more stress on the weld and the machine. Therefore, a mechanical construction for adjusting the saw band tilting will be presented and investigated in this paper. Hydraulic actuators are used, as most band sawing machines already have hydraulic units and the costs for an extension are there- fore low. Fig. 1 Runout of sawing cut Ideal cut Cut Devia�on WorkpieceCu�ng surfaceFeed direc�on De fle c� on 91Production Engineering (2025) 19:89–99 2 Methods and material For the experiments, a Klaeger Pharos 300 band sawing machine is used. The saw band size is 34 × 1.1 × 3950 mm. Tools with high-speed steel (HSS) and carbide (HM) saw teeth are used, respectively. The sawing machine uses a ball screw for the feed drive. The cutting speed is electri- cally controlled. The maximum cutting speed is set at vc = 100 m/min and the feed rate is limited to vf = 200 mm/min. Three eddy current sensors on each side of the cut (infeed and outfeed) are used to measure the saw band runout. These sensors are positioned perpendicular to the saw band body. Figure 3 shows the sensor positions relative to the saw band teeth and back. Figure 3a shows schematically how the band moves between the guides during cutting. Initially, the position of the band between the guides is undefined. At the beginning of the cut, the saw band is shifted to one side of the guides. As the cut- ting depth increases, the saw band begins to tilt between the guides. The correlation between measured and actual displace- ment can be determined in a simplified manner using the bending line equation. In reality, the curve will be somewhat Driving wheel Saw band tension device Entrance Exit Saw band- guides Workpiece with Vise Carrier band Saw teeth made of High Speed Steel (HSS) Or Carbide (HM) Fig. 2 Process schema (a) (b) Work piece Rotation t=0 t=1 t=2 Translation Side guide with air gap Guide saw band back Saw band Eddy 1 Eddy 2 Eddy 3 0 20 40 60 80 Time t /s -0.15 -0.1 -0.05 0 0.05 0.1 Di sp la ce m en ts /m m Eddy 1 Eddy 2 Eddy 3 Fig. 3 Path measurement and saw band course (a) and displacement behaviour of the saw band (b) [32] 92 Production Engineering (2025) 19:89–99 smaller, as the saw blade tension acts against the bending. There are a number of publications on the exact analysis [29–31], but it is beyond the scope of this article to go into this in detail and for the planned project the simplified approach is sufficient and offers the advantage that it can be calculated parallel in real time without requiring many resources. Figure 3b shows a sample measurement of the infeed eddy current sensors during a cut. At the beginning, the saw stands still, then accelerates to cutting speed before entering the material after 10 s. The saw band then starts to deflect and run away, increasing the distance from the sensors until the end of the cut. At the end of the cut, the band snaps back into position because there is no counterforce from the material. Figure 4 shows the actual relationship between the eddy current sensors (here using the example of the lower sen- sor) and the actual workpiece surface, measured in the feed direction. Initially, the plane runs in the opposite direction to the eddy current sensors. This is because the saw blade is not guided through the sawing gap at the start of the cut, but the back of the saw blade can move freely. As a result of the feed force, the saw blade initially twists in one of the direc- tions. As soon as it is slightly lower and guided, it runs in the same direction as measured by the eddy current sensors. In order to manipulate the angle of tilt of the saw band, a new design of saw band guides has been developed. From the initial findings on the behaviour of the saw band in the process, it can be concluded that the band tilting device must be able to influence both the translational and rota- tional movement of the blade. Due to the limited space in the area of the band guide system in the test machine, only one hydraulic cylinder can be used per guide element. This means that the translational and rotational movements must be coupled. The concept chosen for this is shown schemati- cally in Fig. 5, which also shows how the position of the pivot point of the lever mechanism affects the overall move- ment of the system. Only one side of the guides is shown in Fig. 5. On the left side of Fig. 5 the pivot point is at the level of the saw tooth tips. It can be seen that the blade is able to rotate but cannot make any translational movement. In the middle of Fig. 5, the centre of rotation is shifted upwards by the value a, measured from the tips of the saw teeth. This causes the saw blade to rotate by the angle β, while the teeth are deflected laterally by the value b, following an arc. On the right side of Fig. 5 the real system is shown. One development goal was to be able to set the pivot point for the four adjustment units in the same way in a simple and defined manner, and to ensure that the control characteristics were as similar as possible. To achieve this, the relevant components were fitted with a simple set of teeth, allowing the pivot point to be adjusted in 1 mm increments. Figure 6 shows the realisation. For the tests in this paper, a distance “a” of 2 mm was selected. The influence of the distance was analysed in a separate series of tests. The required measurement, control and regulation tasks are implemented using a rapid prototyping system of the type Performance-Target-Machine from the system provider Speedgoat GmbH. This allows controllers designed in Mat- lab/Simulink to be tested on the real machine without major adaptations and conversions from a hardware-in-the-loop simulation, thus creating a match between simulation and reality. In order to test different setups and controllers, a process model was developed and implemented in Matlab/Simulink. It was extended to include the newly developed construction for saw band tilting and the hydraulic actuators. Results from 0 10 20 30 40 50 60 Posi�on on part l /mm 0 0.2 0.4 0.6 0.8 Di sp la ce m en ts /m m 0 0.2 0.4 0.6 0.8 De fle c� on d /m m Cut deflec�on Eddy 3 Surface Fig. 4 Comparison between eddy current sensors and actual sawn workpiece surface measured in the feed direction 93Production Engineering (2025) 19:89–99 analogue tests were used for force prediction [33]. In the test, band saw blade segments with individual teeth are used on a lathe like a grooving tool with an interrupted cut. The process forces and impulses are measured. The next step is to use the information to implement a new control system for the machine. To do this, the three eddy current sensors are used to calculate the actual dis- placement of the tool, taking into account that the saw band is also curved upwards and the actual position of the sen- sors relative to the saw band varies depending on the feed force. A line laser from Keyence (type LJ-V7060) is used to measure this displacement. Together with a set target angle, the error information is fed into a controller which controls the hydraulics with proportional valves. With the outcome from the simulation and the actual tool wear and workpiece information (e.g. size, shape, mate- rial…) some kind of feed forward control could be added. The schematic of this control design is shown in Fig. 7. With prior knowledge of the workpiece geometry and the tool, it is possible to determine how many teeth are in which position in the material. With the help of the above- mentioned single-tooth tests/force prediction, it is possible to determine which forces are currently acting on the saw blade and, taking into account the beam theory, it is pos- sible to calculate in a simplified way how the saw blade will behave, whereby it should be mentioned here that it works well at the beginning, but the more the blade runs, the more boundary conditions occur (saw blade rests on the cutting channel or is supported…) and can therefore only be used as a rough estimate. If these boundary conditions are also to be taken into account, some assumptions of the first-order Euler–Bernoulli beam theory no longer apply and the second order would have to be used, whereby there is also a need for discussion here as to what still applies and what no longer does. In most cases, the step to FEM is the logical one for these questions. However, these can no longer be easily inte- grated into the control system. In the stand presented here, initially only a simple imple- mentation of the feed forward controller was realised. A con- stant value was added based on the classification before the Hydraulic cylinder One adjustment unit Fig. 5 Selected functional principle for the compensation of the saw band tilting and its constructive implementation Fig. 6 Detailed picture of the adjustment unit Guiding element serration 94 Production Engineering (2025) 19:89–99 start of the cut and thus acts as a kind of offset. This offset is relatively small (± 1V) compared to the parameter window (± 10V) for the controller. This was chosen because the tool evaluation only worked with limited reliability in the process environment with dirty saw bands etc. due to insufficient training data. Figure 8 shows a schematic of the experimental setup with all components. The diameter of the workpiece is 60 mm and the material used here was 1.7225 (42CrMo4). The space between the guides was approximately 310 mm. The Figure also shows the direction of the cutting speed vc and feed rate vf. The corresponding positive forces act against these directions. All tests are carried out with flood cooling. Initially, various parameters were tested to determine the limits of the system. Comparative measurements were then made with the conventional set-up and the increase in tool life was considered. The cancellation criterion chosen was a saw blade path displacement of approximately s = 0.3 mm at the eddy current sensors. If the system is viewed in sim- plified terms as a bending beam, it corresponds to a saw blade displacement of approx. 0.9–1.0 mm in the centre. In reality, it is slightly less due to the saw blade tension. Figure 4 shows the comparison between the sawn surface Eddy current sensor 1 Eddy current sensor 2 Eddy current sensor 3 Calcula�on of the �lt Angle and the displacement Shi�ing of band saw back (Line laser) + - Target Angle (0°) + + Knowledge of workpiece User input Force measurment Predicted displacement Process parameters Measurement User input Control plant (adjustment unit/ hydraulic cylinder) Tool wear Simulink Controller Fig. 7 Advanced control design Fig. 8 Experimental setup with components 1: Hydraulic actuators and Band guides; 2: Saw band back guide; 3: Clamping device with workpiece; 4: Eddy current sensors; 5: Line laser sensor; 6: Force sensors 5 1 1 2 2 4 4 3 6 60 mm 310 mm 90 mm 35 mm 95Production Engineering (2025) 19:89–99 and the measurement with the eddy current sensors for this test setup. It can also be seen that the actual course on the workpiece surface is approximately a factor of 2.6–2.7 from the eddy current sensors. The used cutting speed was vc = 60 m/min and the feed rate was vf = 48.5 mm/min. 3 Results and discussion The modelled system has the voltage of the proportional hydraulic valves as input and the angle being set on the saw blade as output. The maximum voltage (U = 10 V) is given as the excitation step and the angle is measured. The step starts at t = 0 s. The used transfer function of the model shows three poles and two roots. Figure 9 shows the approxi- mated model and the real system response to the step signal. It can be seen that there is a good fit (98.6% goodness of fit). This model was used to test controllers in Matlab/Sim- ulink. One of the results shown in Fig. 10 is that a simple PID controller is already good enough for this application and does not require a lot of computer power and time to cal- culate. It is also robust to disturbances such as some incor- rect measurements due to chips. Although there is already an integrator in the control system with hydraulic actuators, better results could be achieved with an integrator and a derivative part in the controller design, even if the derivative part is a small number. The controller parameters used can be seen in the Table 1 The system reacts quickly to the incoming change in tar- get angle, overshooting only slightly and quickly reducing the error to zero. After some testing and confirmation with the real machine and this controller, cutting tests were carried out. Figure 11 0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04 0.045 0.05 Time t /s -0.5 0 0.5 1 1.5 2 2.5 3 An gl e α / ° Reaction of the real system and the model to the step excitation Model real System 0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04 0.045 0.05 Time t /s -0.5 0 0.5 1 1.5 2 2.5 3 An gl e α / ° Reaction of the real system and the model to the step excitation Model real System Fig. 9 Model fitting Fig. 10 PID controller Table 1 Controller parameters Control parameter Value Proportional (P) 5 Integral (I) 10 Derivation (D) 0.01 Filter coefficient (N) 150 96 Production Engineering (2025) 19:89–99 shows a saw cut with the developed actuator system with a partially worn saw blade. In the first run, the cutting parameters were increased with each cut to determine the capabilities of the construc- tion and the controller. Figure 12 shows the measurement of the eddy current sensors on the exit side of the final cut with the maximum parameters of a cutting speed vc = 100 m/ min and a feed rate vf = 200 mm/min. It also shows the input values for the saw band guide valve on the same side. It can be seen that the system kept the saw band in a straight line for about 50% of the cutting time ((B) and (C); t = 24 s to 32.5 s). Then the construction reached its mechanical limits and the cut started to run away (D). At t = 40 s, the machine had to be stopped because the forces were too high. As an additional note, it should also be pointed out again that a round material was machined here. As a result, even in the "normal" process, the forces increase up to half of the cut and then drop. In the test, the maximum force was reached Fig. 11 Stable saw cut with partially worn saw blade 20 25 30 35 40 45 Time t /s -0.5 0 0.5 Di sp la ce m en ts /m m -5 0 5 Vo lta ge U /V stable sawing cut Eddy 4 Eddy 5 Eddy 6 Ven�l 2 Fig. 12 Measurement results of maximum possible parameters on the test machine, vc = 100 m/ min and vf = 200 mm/min 20 25 30 35 40 45 Time t /s -1 0 1 2 Di sp la ce m en ts /m m -15 -10 -5 0 5 Vo lta ge U /V Aborted process Displacement and reac�on Eddy 4 Eddy 5 Eddy 6 Ven�l 2 20 25 30 35 40 45 Time t /s -2000 0 2000 4000 Fo rc es F /N Processforce Fc Fp Ff (A) (E)(D)(C)(B) 40 97Production Engineering (2025) 19:89–99 in the last third of the cut. The run-out started at about the halfway point. The forces were measured on the workpiece side using the force measuring platform. Figure 13 shows the lateral paths of the saw blade on the entry side in a series of experiments on the service life of the blade. In this series, the conventional construction with fixed jaws was used up to a cutting area of approximately A = 0.96  m2. When the saw blade began to deviate from its path and the end of its service life was approaching (left- hand path in Figure), the machine was converted to the new design described to investigate how many more cuts could be made without the blade deviating. The middle path in Fig. 13 shows the measurements after A = 1.0  m2. The first downward deflection is caused by the first contact with the workpiece and the pushing away of the worn tool. This curve is compensated for by the actuators in the first few millime- tres and is kept constant until shortly before the end of the process. Towards the end of the process, there is another jump in the curve, which is caused by the slightly slower controller and the changing engagement ratio when sawing a round material (the engagement length decreases rapidly) and the resulting overdrive of the controller. After a total cut area of A = 1.08  m2 the compensating construction reached its mechanical limit and the blade once again deviated analogously to the old construction (right- hand path). With the new design, the service life of the blade was increased by above 10% on average. When analysing the cutting edges of the tools, it can be seen that with the conventional design, one side is particu- larly worn and the saw blade also runs in this direction. The growth of wear marks and the rounding of the edge with increasing cutting area are visible in Fig. 14. By switching to the compensating design, the cutting edge is worn more evenly and therefore used more efficiently. (Fig. 14). 4 Conclusion It could be shown that with adjustable saw band guide ele- ments the saw band can be used longer with the same mate- rial loss. In addition, higher feed rates can be used towards the end without it running further. This reduces the process costs and increases the overall efficiency of the process. In addition to increasing the service life when sawing raw materials, the knowledge gained about the process and behaviour can be used to compensate for the cutting of inho- mogeneous materials and cross-sections. Especially in the field of metal 3D printing, where the component is sepa- rated from the printing platform using band saws, the find- ings can be of great benefit. With metal printers, different components with different cross-sectional areas and orienta- tions are often printed on one platform and must be cut off together. As a result, the engagement conditions and thus the forces on the saw blade vary, and so does the course of the cut. Active compensation could reduce the distance from the component to the platform and thus save material. It is also possible to cut closer to the component, which is a great advantage for downstream processes, especially in the area of difficult-to-machine metals such as titanium or tool steel. In the next step the hydraulic shall be replaced with elec- trical motors. By simplifying measurement and control, later the system can be used as a retrofit for existing machines that do not have a complex control system. For example, 0 50 100 -0.4 -0.2 0 0.2 Rigid structure, end of service life A = 0.96 m² 0 50 100 -0.4 -0.2 0 0.2 New structure, beginning A = 1.00 m² Eddy 1 Eddy 2 Eddy 3 0 50 100 -0.4 -0.2 0 0.2 New structure, end of service life A = 1.08 m² �me t /s Di sp la ce m en ts /m m Fig. 13 Lateral course of the saw band in the adaptive design compared to the conventional design 98 Production Engineering (2025) 19:89–99 with the new knowledge and simulation model the number of sensors can be reduced and the line laser can be replaced by a tactile sensor. Acknowledgements Funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) - 289598126. Author contributions Prof. Möhring provided the basic idea for the funding and was on hand to advise during the research. Mr Tandler carried out the elaboration, experiments and evaluations. The article was written in co-operation. Funding Open Access funding enabled and organized by Pro- jekt DEAL. Deutsche Forschungsgemeinschaft e.V., 289598126, 289598126. Data availability No datasets were generated or analysed during the current study. Declarations Conflict of interest The authors declare no conflict of interests. Open Access This article is licensed under a Creative Commons Attri- bution 4.0 International License, which permits use, sharing, adapta- tion, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. References 1. Albrecht D, Möhring H-C (2018) Potentials for the optimization of sawing processes using the example of bandsawing machines. Procedia Manuf 21(3):567–574 2. Reng D (1976) Das Trennen von Metallen durch Bandsägen unter besonderer Berücksichtigung des Verlaufens des Schnittes. Dis- sertation, München 3. Sarwar M, Haider J, Persson M (2011) Scientific evaluation of cutting-off process in Bandsawing. Sci Proceed Int Congr Mach Technol Mater 2:9–12 4. Sarwar M, Persson M, Hellbergh H, Haider J (2009) Measurement of specific cutting energy for evaluating the efficiency of band- sawing different workpiece materials. Int J Mach Tools Manuf 49(12–13):958–965 5. Gendraud P, Roux J-C, Bergheau J-M (2003) Vibrations and stresses in band saws. J Mater Process Technol 135(1):109–116 6. Wang KW, Mote CD (1987) Band/wheel system vibration under impulsive boundary excitation. J Sound Vib 115(2):203–216 7. Thaler T, Potočnik P, Bric I, Govekar E (2014) Chatter detec- tion in band sawing based on discriminant analysis of sound features. Appl Acoust 77:114–121 8. Thaler T, Potočnik P, Govekar E (2014) Experimental charac- terization of chatter in band sawing. In: Dalpiaz G, Rubini R, D’Elia G, Cocconcelli M, Chaari F, Zimroz R, Bartelmus W, Haddar M (eds) Advances in condition monitoring of machin- ery in non-stationary operations. Springer, Berlin Heidelberg, Berlin, Heidelberg, pp 701–709 9. Thaler T, Potočnik P, Mužič P, Bric I, Bric R, Govekar E (2012) Chatter recognition in band sawing based on feature extraction and discriminant analysis. In: Fakhfakh T, Bartelmus W, Chaari F, Zimroz R, Haddar M (eds) Condition monitoring of machin- ery in non-stationary operations. Springer, Berlin Heidelberg, Berlin, Heidelberg, pp 607–615 10. Thaler T, Krese B, Govekar E (2015) Stability diagrams and chatter avoidance in horizontal band sawing. CIRP Ann 64(1):81–84 11. Schajer GS (1986) Why are guided circular saws more stable than unguided saws? Holz als Roh Werkst 44(12):465–469 12. Singhania S, Law M (2024) Regenerative instabilities in guided metal circular sawing. Sādhanā. https:// doi. org/ 10. 1007/ s12046- 024- 02442-x 13. Singhania S, Singh A, Law M (2022) Dynamics and stability of metal cutting circular saws with distributed and lubricated guides. J Vib Eng Technol 10(8):3119–3131 14. Szymani R, Mote CD (1977) Principal developments in thin circular saw vibration and control research. Holz als Roh- und Werkstoff 35(6):219–225 A = 0.025 m² A = 0.5 m² A = 0.96 m² A = 1.08 m² 1,1 mm Fig. 14 Wear patterns for the material 42CrMo4 at D = 60 mm, vc = 60 m/min and vf = 48.5 mm/min http://creativecommons.org/licenses/by/4.0/ https://doi.org/10.1007/s12046-024-02442-x https://doi.org/10.1007/s12046-024-02442-x 99Production Engineering (2025) 19:89–99 15. Thaler T, Potočnik P, Mužič P, Bric I, Bric R, Govekar E (2012) Characterization of band sawing based on cutting forces. J Mach Eng 12:41–54 16. Lund D (2020) Leistungssteigerung beim Metallbandsägen durch prozessabhängige Regelung der Sägebandspannung. Dis- sertation, Universität Stuttgart 17. Lehmann BF (1993) The cutting behavior of bandsaws. Dis- sertation, Kanada 18. Gerken JF, Klages N, Biermann D, Denkena B (2020) In-pro- cess compensation of straightness deviation in BTA deep hole drilling using experimental and simulative analysis. Procedia CIRP 93:1417–1422 19. Baek DK, Ko TJ, Kim HS (2001) Optimization of feedrate in a face milling operation using a surface roughness model. Int J Mach Tools Manuf 41(3):451–462 20. Pimenov DY, Guzeev VI, Mikolajczyk T, Patra K (2017) A study of the influence of processing parameters and tool wear on elastic displacements of the technological system under face milling. Int J Adv Manuf Technol 92(9–12):4473–4486 21. Tatar K, Gren P (2008) Measurement of milling tool vibrations during cutting using laser vibrometry. Int J Mach Tools Manuf 48(3–4):380–387 22. Wojciechowski S, Twardowski P, Pelic M, Maruda RW, Barrans S, Krolczyk GM (2016) Precision surface characterization for finish cylindrical milling with dynamic tool displacements model. Precis Eng 46:158–165 23. Wojciechowski S (2011) Machined surface roughness including cutter displacements in milling of hardened steel. Metrol Meas Syst. https:// doi. org/ 10. 2478/ v10178- 011- 0009-7 24. Abele E, Bauer J, Rothenbücher S, Stelzer M, Von Stryk O (2011) Prediction of the tool displacement by coupled models of the com- pliant industrial robot and the milling process 25. Diebold J, Wiesinger G, Bleicher F (2021) Cutting deviations in bandsawing of cylindrical metal parts. In: Holl H (ed) 37th Danubia-Adria symposium on advances in experimental mechan- ics, Linz, pp 186–187 26. Diebold J (2024) Tooth Wear Measurement Methods for Bandsaw- ing [Poster Presentation]. 1st Science Day – Faculty of Mechani- cal and Industrial Engineering, Wien, Austria. https:// doi. org/ 10. 34726/ 5819 27. Albrecht D, Stehle T (2016) Influence of saw blade tension on the stability of sawing processes. AMR 1140:189–196 28. Albrecht D, Stehle T (2016) Measuring the dynamic twisting behaviour of saw Blades in the kerf during the sawing process. J Mach Eng 2016(3):75–87 29. Andersson C, Andersson M, Stahl J-E (2001) Bandsawing. part I: cutting force model including effects of positional errors, tool dynamics and wear. Int J Mach Tools Manuf 41(2):227–236 30. Andersson C, Ståhl J-E, Hellbergh H (2001) Bandsawing. part II: detecting positional errors, tool dynamics and wear by cutting force measurement. Int J Mach Tools Manuf 41(2):237–253 31. Kremleva LV, Malygin VI, Lobanov NV (2016) Dynamic stability of band saw machines. Russ Engin Res 36(9):735–740 32. Wiedemann M, Melzer T (Eds.) 2022. Tagungsband 4SMARTS 2022: Smarte Strukturen und Systeme - Tagungsband des 4SMARTS-Symposiums 09.-10. März 2022, Braunschweig, pp. 125 33. Albrecht D, Stehle T A new approach for calculating the analytical resultant force of sawing processes considering the tooth impact momentum, In: , 7. WGP Jahreskongress, pp. 443–450 Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. https://doi.org/10.2478/v10178-011-0009-7 https://doi.org/10.34726/5819 https://doi.org/10.34726/5819 Active runout compensation using the guide elements on metal band saws for a longer tool life and reduced material loss Abstract 1 Introduction 2 Methods and material 3 Results and discussion 4 Conclusion Acknowledgements References