Albrecht Praezisions-Spannfutter

86 87 Production Engineering in order to calculate estimated surface roughness values. Experiments and calculations both delivered the best sur- face for the shortest A-length. Fleischer et al. [13] compared the influence of tool holder clamping systems on dynamic stability, productivity and tool wear. By means of modal and operational analysis, resonance frequencies and related damping properties were identified. Weck, Hennes and Krell [14] investigated possibilities to improve the low dynamic stability limits of processes with slender overhanging tools (length/diameter-ratio, l ∕ d > 6 ), i.a. by adding a dampen- ing fluid film inside a thermal shrink fit tool holder, thus creating a spring-damper element with computable, con- figurable properties. The developed tool holder exhibited lower dynamic compliance compared to a hydraulic expan- sion and a thermal shrink tool holder. Regarding dynamic stability of milling processes and its prediction, Gradisek et al. [15] investigated the milling with long overhanging tools ( l ∕ d = 12 ) at small radial immersions using thermal shrink tool holders in high-speed milling operations. For stability prediction two approaches, Zero Order Approxi- mation (ZOA) and Semi-Discretization (SD) method were applied. In this context, dynamic tool deflections for vari- ous types of chatter-related tool movements were studied. Agapiou and Rivin [16] on the opposite investigated two concepts to enhance drawbacks caused by the spindle-side interface of 7/24 taper tool holders compared to interfaces such as HSK (standardised in DIN 69893 [17]) or Capto (standardised in ISO 26623-1 [18]). One method suggested by Rivin [19] uses cages with balls out of variable material which are attached to the taper surface and is capable to enhance radial stiffness and radial/axial positioning without limiting aspects like denting surfaces in use. Brecher et al. [20] investigated the influence of machine-spindle interfaces of the tool holder on the (radial) static stiffness depending on the cutting forces in machining experiments. A model based on different combinations of mechanic springs was devel- oped to describe the components of the power train in order to predict the overall stiffness. Nebeling [21] analysed the extent, to which machine tool, spindle and tool–tool holder 3 Experimental preparation 3.1 Investigated tool holders Fig. 1 Examples of the investigated tool holder systems Table 1 Main properties of the investigated tool holders a Full-periphery clamping b Point-line clamping Tool holder A mass m D 1 b m Design [mm] [g] [mm] [mm] APC 92 1457 40 18 Modular ER 100 1274 42 12.5 Modular HE 90 1123 32 9 Monolithic a TS 90 918 24 8 Monolithic a Weldon 80 1120 42 15 Monolithic b values. best sur- ompared dynamic f modal d related nd Krell dynamic ng tools dampen- der, thus ble, con- xhibited c expan- dynamic Gradisek hanging thermal ons. For Approxi- od were for vari- studied. ated two dle-side nterfaces or Capto uggested material pable to without Fig. 1 Examples of the investigated tool holder systems Table 1 Main properties of the investigated tool holders Tool holder A mass m D 1 b m Design [mm] [g] [mm] [mm] APC 92 1457 40 18 Modular ER 100 1274 42 12.5 Modular HE 90 1123 32 9 Monolithic a TS 90 918 24 8 Monolithic a Bewertung der Werkzeughalterleistung beim Schruppfräsen mit Schaftfräsern Vol.:(0123456789) 13 Assessment of tool holder performance in roughing with end mills Process dynamics and surface formation Oliver Rosenthal 1 · Wolfgang Hintze 1 · Carsten Möller 1 Received: 21 May 2019 / Accepted: 13 December 2019 © German Academic Society for Production Engineering (WGP) 2020 Abstract Tool holders are an important power train component of milling with shaft tools having a high impact on dynamic process behaviour and process results. This leads to a significant demand among industrial companies for scientifically proven methods to analyse tool holders which are easy to use as well. Different approaches were applied in order to meet these demands. The dynamic vibrational behaviour of different clamping mechanisms is investigated using tool holders of similar dimensions. First, the resonance frequencies of the tool holders are identified from dynamic compliance measurements in the machine tool. Subsequently, the dynamic process behaviour is investigated by peripheral milling tests in which vibrations of tool and tool holder are detected by acceleration sensors and microphones. Analysis of the sensor data and optical analysis of the manufactured surface reveal a significant influence of the particular clamping mechanism, superimposed by certain geometrical variations of the investigated tool holders. Chatter frequencies occurring during milling tests seem independent of the clamping mechanism respectively the particular tool holder and are caused by the tool or machine-tool components. They correspond roughly with the natural frequencies identified before. Chatter intensity and spindle speeds, at which chatter occurs, are influenced by the tool holders. The occurrence of chatter vibrations correlates with a significant drop in the surface quality of the workpiece. High resonance frequencies of the tool holder excited by chatter and low-frequency waviness observed on the machined surface are related. The well-known phenomenon can be explained under rough milling conditions by a 2D-model-based superposition of subsequent cutting edge engagements. This in turn may enable dynamic analysis and optimisation of rough milling operations by easy to use shop floor equipment in future. Keywords Tool holders · Dynamic behaviour · Surface formation · Frequency analysis · Shaft tools · Steel milling 1 Introduction Tool holders for end mills are an important component of the power train in milling. They affect the manufacturing pro- cess itself significantly, e.g. the dynamic stability and thus the productivity. The selection of a suitable tool holder can require numerous experimental tests due to the complexity of the system as well as lack of knowledge and information. Especially for small-lot orders, this can cause time-consum- ing and expensive production preparation. Tool holder selection gets more complicated due to an increasing variety. During the last years, manufacturers came up with a high number of modifications and inven- tions within this field. In practice, tool holder selection is often based on experience of single workers. Scientifically reliable information about tool holder per- formance can only be obtained by elaborate series of tests. These often require equipment and methods not available in industry, es ecially in smaller compani s. A limiting factor for rough machining is the dynamic stability of the overall process which is, amongst other aspects, also influenced by the clamping mechanism and the dimensions of the tool holder. Due to the high impact on productivity and costs of the manufacturing process, dynamic stability is important when designing manufactur- ing processes. The limiting influence becomes obvious when it comes to negative impacts on productivity and tool wear * Oliver Rosenthal oliver.rosenthal@tuhh.de 1 Institute of Production Management and Technology, Hamburg University of Technology, Denickestr. 17, Hamburg, Germany Test-Aufbau: Maschine: MC12, Gebr. Heller Werkzeug: Torusfräser, D=12, Zäh ez hl= 4 Mat ria : Stahl, 42CrMo4 (1.7225) Parameter: n = 3979 min -1 f z = 0,04 mm a e = 7 mm a p = 18 mm Ta elle 1 Abmessungen der Untersuchten Werkzeughaltersysteme Unt rsuchte Werkzeughaltersysteme: Zusammenhang zwischen prozessinternen Schwingungen und Oberflächenqualität: Rosenthal O. / Hintze W. Möller C. (Veröffentlicht 08.01.2020 https://doi.org/10.1007/s11740-019-00944-w. Assessment of tool holder performance in roughing with end mills. Institut für Produktionsmanagement und -technik, Technische Universität Hamburg. Springer Nature) APC mit den besten Ergebnissen im Vergleich zu anderen Werkzeughaltern: Production Engineering 13 5 Substitution of acceleration sensors by audio‑recording equipment for detection of chatter frequencies Proven by experience, audible sound indicates the machine operator whether a milling process is stable or not. Sound generated by chatter can be captured by m processed by freeware (e. g. application and mobile devices) and communicatio able on the consumer market. This can for SMEs to get closer insight into their In order to investigate this aspect, sou Fig. 6 Visualisation of in-process vibrations using Campbell diagrams based on acceleration sensor data. ( n = f z = 0.04 mm/rev/tooth , a e = 7 mm , a p = 18 mm ) Zusammenfassung Werkzeughalter sind ein wichtiger Bestandteil des Antriebsstrangs beim Fräsen mit Schaftwerkzeugen und haben einen hohen Einfluss auf d s dyna ische Prozessverhalten und die Prozessergebnis e. In der Industrie besteht daher ein großer Bedarf an wissenschaftlich fundierten und g ichzeitig einfach zu handhabende Methoden zur An lyse v n W rkzeugaufnahmen. Um die en A ford rungen gerecht zu werden, wurden verschiedene Ansä ze verfolgt. Das dynamische Schwingungsverhal- ten verschied ner Spannmecha ismen wird anhand vo Werkzeu haltern ähnlicher Abmessungen un ersucht. Zunächs we - en die Resonanzfrequenze d r Werkzeughal er aus dynamisch n Nachgiebigkeitsmessungen in der W rkzeugma chine ermittelt. Anschließend wird das dy a ische Prozessverhalten durch Umfangsfräsversuche untersucht, bei denen die Schwin - gungen von Werkzeug und Werkzeughalter durch Beschleunigungssensoren und Mikr fone erfasst werden. Die Auswertung der Sensordaten und die optische Analyse der gebildeten Oberfläche zeigen einen signifikanten Einfluss des jeweiligen Spann- mechanismus, überlagert von bestimmten geometrischen Ausprägungen der untersuchten Werkzeughalter. Die bei den Fräs - tests auftretenden Ratterfrequenzen scheinen unabhängig vom Spannmechanismus bzw. der jeweiligen Werkzeugaufnahme zu sein und werden durch das Werkzeug bzw. die Werkzeugmaschinenkomponenten verursacht. Sie stimmen in etwa mit den zuvor ermittelten Eigenfrequenzen überein. Die Ratter-intensität und die Spindeldrehzahlen, bei denen Ratterschwingungen auftreten, werden durch die Werkzeughalter beeinflusst. Das Auftreten von Ratterschwingungen korreliert mit einer deutli - chen Verschlechterung der Oberflächenqualität des Werkstücks. Hohe Resonanzfrequenzen des durch Rattern angeregten Werkzeughalters und die an der bearbeiteten Oberfläche beobachtete niederfrequente Welligkeit stehen im Zusammenhang. Dieses bekannte Phänomen lässt sich unter Schruppfräsbedingungen durch eine 2D-Modell-basierte Überlagerung von aufei - nanderfolgenden Schneideneingriffen erklären. Dies wiederum könnte in Zukunft die dynamische Analyse und Optimierung von Schruppfräsvorgängen durch einfach zu bedienende Produktionsanlagen ermöglichen. Stichwörter Werkzeughalter - dynamisch s V hal en - Oberflächenbildung - Frequenzanalyse - Schaftwerkzeuge – Fräsen in Stahl Abb. 7 Zusammenhang zwischen prozessinternen Schwingungen und Oberflächenqualität. (Werkzeughalter: ER-Spannzange, n = 2708 – 5250 min -1 , ƒ z = 0,04 mm, a e = 7 mm, a p = 18 mm) Production Engineering 13 evaluated using a condenser microphone in combination with a linear audio-recorder and compared to the accel- eration signals obtained with scientific measurement equipment. In order to compare both methods, the microphone is mounted at the head of the spindle, next to the accelera- tion sensor. This setup ensures a constant distance between the tool contact zone and the location, where the sound is detected. The subsequent analysis of both signals is executed using the same algorithms programmed in MATLAB. The evaluation of both signals shows very similar results with different amplitudes due to the different measuring principles. The frequencies identified by using sound are the same as those obtained via the acceleration sensor. Frequency identification was possible for all investigated tool holders across the whole spindle speed range. Figure 8 shows an example. 6 Model for surface structure due to chatter vibrations Another approach to verifying chatter vibrations instead of measuring accelerations is the analysis of the generated workpiece surface. The analysis is carried out by using a simplified kinematic model. It considers subsequent cutting edge engagements at a constant z-coordinate, i.e. at a con- stant depth of cut superimposed by a chatter vibration of a defined amplitude A and frequency f chatter . The path of each cutting edge relative to the workpiece is first calculated sepa- rately for the x- and y-direction (see Fig. 9). In a shop floor environment the chatter frequency can be captured by the audio and communication equipment according to Sect. 5. The movement in x-direction consists of two components, a lateral part representing the translational motion derived from feed rate per tooth f z , tooth number z , spindle speed n and the time t and the x-part of the circular tool movement, represented by a cosine part and the individual cutting edge radius R i . The argument of the cosine describes the rota- tional angle depending on spindle speed n and time t plus a phase shift i representing the relative positions of the individual cutting edges to each other. The movement in y-direction is given by the sinus term of the circular tool movement (analogue to Eq. 1) and a second sinus term modelling the actual tool vibration perpendicular to the feed rate direction depending on the amplitude A and the frequency f chatter of the chatter vibration. Superposition of subsequent cutting edge movements pro- vides an envelope which represents the generated workpiece surface at the coordinate z. As shown in Fig. 10 by way of (1) x i = R i ⋅ cos − 2 ⋅ n ⋅ t + 2 i − n ⋅ f z ⋅ z ⋅ t (2) y i = R i ⋅ sin − 2 ⋅ n ⋅ t + 2 i + A ⋅ sin 2 f chatter ⋅ t Fig. 7 Correlation between in-process vibrations and surface quality. (Tool holder: ER-collet, n = 2708 − 5250 rev/min , f z = 0.04 mm/rev/tooth , a e = 7 mm , a p = 18 mm ) Spindelgeschwindigkeit in min -1 stabil n = 4661 min -1 n = 4165 min -1 unstabil Frequenz in Hz Beschleunigungsdaten Abb. 1 Beispiele er Untersuchten Werkzeughaltersysteme d related od were Examples of the investigated tool holder systems D ER-Spann- zange Thermisches Schrumpfen Weldon Hydrodehn APC Werkzeug- halter A (mm) Masse m (g) D 1 (mm) b m (mm) Ausführung APC 92 1457 40 18 Modular ER 100 1274 42 12,5 Modular HD 90 1123 32 9 Monolithisch a TS 90 918 24 8 Monolithisch a Weldon 80 1120 42 15 Monolithisch b a voll umfassende Spannung b punktuelle Spannung Frequenz in Hz Frequenz in Hz Freq enz in Hz Frequenz in Hz Spindelgeschwindigkeit in min -1 Spindelgeschwindigkeit in min -1 Spindelgeschwindigkeit in min -1 Spindelgeschwindigkeit in min -1 Spindelgeschwindigkeit in min -1 HD Abb. 6 Visualisierung von prozessinternen Schwingungen durch Campbell Diagramme, basierend auf Daten von Beschleunigungssensoren. (n = 2708 – 5250 min -1 , ƒ z = 0,04 mm, a e = 7 mm, a p = 18 mm)