Process Monitoring of Abrasive Waterjet Formation
M. Putza,b, M. Dittricha*, M. Dixa
aTechnische Universität Chemnitz, Institute for Machine Tools and Production Processes (IWP), Reichenhainer Str. 70, 09126 Chemnitz, Germany.
bFraunhofer Institute for Machine Tools and Forming Technology IWU, Reichenhainer Str. 88, 09126 Chemnitz, Germany.
* Corresponding author. Tel.: +49-(0)371-531-32631; fax: +49-(0)371-531-832631. E-mail address: email@example.com
Difficult to machine materials require innovative processing solutions for a stable and high quality contouring process of complex forms. Abrasive waterjet cutting gains in importance due to the continuous development of novel high performance materials and multi-material components. A reliable process monitoring during the machining operation becomes essential to avoid waste production. However, the measurement of the process conditions during abrasive waterjet cutting is difficult based on the rough environment inside the machining zone. In this paper appropriate methods for in-process monitoring of the jet conditions, in particular the critical nozzle wear as well as other process output parameters are being tested, discussed and classified.
In the context of developing novel high-performance materials new challenges become apparent during their manufacturing, especially in machining. In particular, multimaterial
components are difficult to handle due to the fact that different types of materials usually require different tool geometries for the best machining result possible. Abrasive
waterjet machining is a process that overcomes this barrier with an almost unlimited processing variety of diverse materials .
Raising the process stability at a simultaneous increase of the operating time of the nozzles, which results in a reduction of the downtime, necessitates new methods for the in-process
monitoring of abrasive waterjet machining. However, due to the demanding contamination with reflecting splash water, water fog or abrasive particles, a measurement recording of the waterjet quality after the outlet of the cutting head is not useful during contouring or surface structuring .
2. Process Disturbances and their Influence on the Machining Result
During abrasive waterjet cutting based on the injection principle, first of all the pressure of pure water is increased above 3,000 bar by using a high pressure pump. The water runs through a so-called water nozzle, thus changing the hydraulic energy into kinetic energy. Here, water can achieve a speed of 800 m/s and more . Inside the mixing chamber an underpressure, which intakes the applied abrasive particles, arises due to the high velocity of the waterjet. In the focusing tube the abrasive material is being accelerated and aligned in its flow direction . In this context the water inside the abrasive waterjet only serves as an acceleration fluid. The
abrasive particles are responsible for the cutting action . Hashish et al.  describe the different influences on the waterjet inside the cutting head, impairing the machining
result. One of these factors is the condition of the water nozzle.
As a result of furring, blockages due to dirt and disruptions at the water nozzle, the kinetic energy of the pure waterjet is already being influenced . In this connection turbulences
occur as well as a consequential nonpoint fanning out of the waterjet inside the mixing chamber. So, the process typical recess of the focusing tube caused by continuous contact
between the abrasive particles and the inner surface of the nozzle is intensified. As expected, besides a change of the acoustics, there is a transformation of the kinetic energy of the
waterjet to thermal energy as a consequence of the friction between focusing tube and the abrasive waterjet. This factor is connected with an intense speed reduction of the cutting jet.
Thus, the acceleration of the abrasive material is insufficient for chipping as well.
Interruptions and accordingly blockages of the abrasive supply can be related to water contamination of the abrasive inlet or bad suction performance inside the cutting head. In
this case, no more abrasive material is added into the cutting jet anymore. Inevitably, a process interruption occurs.
Based on the knowledge of these facts and the difficult circumstances inside the workspace, it proves to be useful to interpret the process conditions inside the cutting head without monitoring the waterjet pattern itself.
3. Experimental Equipment and Setup
The investigations were executed using a test stand of Technische Universität Chemnitz for 5-axis simultaneous machining via abrasive water fine jet based on the injection principle (cf. Fig. 1).
A waterjet with a diameter of 0.8 up to 1.0 mm is used for conventional machining tasks . However, the diameter of the examined abrasive water fine jet being used for precise
machining is 0.3 mm. According to [9, 10, 11] the diameter ratio between the water nozzle and the focus nozzle is 1:3 for ideal flow conditions. Appropriate nozzle diameters were
utilized accordingly. The length of the focusing tube was set to be 24 mm.
Due to the adverse conditions inside the working space of a 5-axis abrasive waterjet machine, first of all measurement systems have to be preselected regarding their usability. The
utilized sensors and their positioning at the cutting head are illustrated in Figure 2. Under the presupposition that disturbances inside the cutting head cause changes in its thermal conditions, a thermocouple was positioned in direct contact with the water nozzle carrier. As a result of a changing friction rate, for example due to damages at the sapphire nozzle, changes of the temperature in this area can be detected without influencing the possibility to replace the
nozzle. In order to validate the results as well as to collect the temperature changes due to changed flow conditions, a resistance thermometer was attached at the outside casing of
the cutting head near the mixing chamber.
Furthermore, a monitoring microphone capsule was positioned in a borehole inside the cutting head carrier, directed towards the mixing chamber. Thus, surrounding noises shall be faded out and changes of the flow conditions shall be detected to the greatest extent possible.
By conducting structure-borne sound measurements it is possible to detect vibrations of the cutting head. The vibrations can be a result of the start of oscillations, which occur, for example, due to changing flow conditions inside the mixing chamber and the focusing tube. The monitoring principle was executed by the application of a single-axis accelerometer directly attached to the cutting head. The accelerometer was positioned in a way to record radial
amplifications of the head. Reference values were delivered by a second tri-axial accelerometer attached to the machine frame near the cutting head.
The observation of the air volume flow inside the abrasive particle supply offers valuable clues concerning the wear status of the water nozzle. Ideally it is possible to detect blockages of the abrasive particle supply as well. The measurement took place by using a single-pipe flow
measuring system that had been attached inside the abrasive supply hose.
4. Experimental Procedure and Results
The investigations were carried out using the parameter combinations listed in Table 1. During data acquisition, the abrasive waterjet was switched on as usual for a machining operation. After that, constant process conditions without feed motion were recorded in a time range of 15 seconds. Finally, the turning off of the jet was recorded, as well.
The experiments took place under different water nozzle conditions. Figure 3 shows the wear states of the two utilized water nozzles and its effects.
The novel nozzle shows no wear or disruptions. In consequence, this nozzle shapes an ideally focused pure waterjet. The disruptions at the die orifice of the defect water nozzle brick cause a turbulent flow during forming of the waterjet. As a consequence, the pure waterjet is fanning out and its kinetic energy gets lost.
By taking measurements using the resistance thermometer and the thermocouple, after starting the processing differences between the temperature drifting without and with adding
abrasive particles were detected at the cutting head. Obviously, the friction at the inside of the focusing tube and the mixing chamber causes heating up of the machine components during
processing. However, a change of the abrasive mass flow rate did not have an influence on the temperature profile, but the water pressure and the resulting waterjet velocity did.
By utilizing the water orifices with variable wear states, a higher temperature gradation was determined with the worn nozzle after turning on the water valve. Due to the premature
fanning out of the waterjet directly at the outlet of the water nozzle the water impinges the inner walls of the cutting head. The abrasive particles increasingly collide with the mixing
chamber and the focusing tube as well (cf. Fig. 3, right). Besides these factors, the worn sharp-edged water nozzle with microcracks causes major dissipations of water. In consequence, the cutting head heats up. So, by measuring the thermal behaviour of the cutting head, it is generally possible to detect the wearing condition of the nozzles. However, the heat conduction causes a time mismatch which varies, depending on the configuration of the cutting head and the positioning of the sensors.
In addition to the thermal measurement, the noise level was recorded by using a measuring microphone capsule inside the cutting head. After filtering the signals, characteristic sectors
inside the curve progression were assigned to switching of the water and abrasive supply on and off. The blue curve in Figure 4 shows such a typical measuring signal. At point 1, the water valve is being opened. Up to point 2, the pure waterjet stabilizes. From this point on the measuring microphone records a homogenous sound level, up to point 3, where the abrasive medium is finally applied. Between point 4 and point 5, the effective machining can take place with the abrasive waterjet. Then the abrasive supply and, with a time offset, the water is switched off again at point 5 and point 7. The highlighted zone shows a blockage inside the abrasive supply. This blockage is characterised by the declination of the noise level not exceeding the noise of a pure waterjet. Hence, a permanent reduction of the noise level during
machining would be an indicator for a process interruption causing a damage of the workpiece material.
During the investigations the change of the mass flow of abrasive material per time unit was registered by observing the noise level. A reduction of the noise level with increasing
amount of abrasive material was noticeable. The reason for this phenomenon is the dedicated kinetic energy of the waterjet for accelerating the abrasive material. Due to the fact
that a reduction of the water pressure also causes a decreased noise level, it can be assumed that the waterjet velocity has a major influence on the measuring result.
During the waterjet development the erratic fluctuations, which are caused by the defect orifice, can also be monitored inside the noise level graph (cf. Fig. 4, red curve). The sound level of the fanned out waterjet inside the mixing chamber is influenced only negligibly by adding abrasive particles. This is based on the increasing Venturi effect which causes higher suction of the abrasive material as well as smaller fluctuations in speed between the fluid and the solid particles. Also, the abrasive grains only marginally change the shape of the already expanded waterjet. The reduced jet velocity as a result of the adverse flow conditions becomes noticeable due to the reduced noise level when apart from that using the same test parameters.
A problem during monitoring the sound level is the surrounding noise. Besides independent disturbances, the acoustic noise that develops inside the point catcher has a major influence on the measuring result. Investigations with foam material as sound absorber resulted in a displacement of the characteristic noise level curve to lower ranges. For this reason it can be assumed that different machining settings have an influence on the measuring result and, as a result the triggered numeric control could cause unwanted stopping of the process without prior calibration.
When evaluating the results of the accelerometers, it can be verified that there is a dependence on the water pressure, just as it is true for the acoustic noise measurement. While the frequency distribution remains constant, the amplitudes of the frequency spectrum increase with an increasing water pressure. Frequencies which distort the measurement, such as resonance frequencies of pumps, can easily be filtered out of the spectrum by preliminary inspection of the working environment.
Within the study of the different wear constitutions of the water orifice, when examining a pure waterjet, almost no changes between the new and the damaged orifice were detected in the frequency spectrum.
However, as shown in Figure 5, the input of abrasive particles inside the mixing chamber caused a monitoring of different oscillation amplitudes. Contrary to the assumption
that a defect water nozzle causes stronger collisions of abrasive particles with the inner walls of the cutting head and, as a consequence, stronger structure-borne sounds, the oscillation of the amplitudes actually decreases with higher wear of the orifice. The same behaviour was verified during the interpretation of the frequency spectrum of acoustic noises.
The airflow was measured by using different values of water pressure in steps of 500 bar without addition of abrasive particles and with both the novel water nozzle and the damaged one. Figure 6 shows the measurement results. The findings point out the obvious dependence of the air flow on the orifice condition. With this knowledge it is possible to attach a volume flow gauge with an evaluating processor unit to the systems engineering. This instrument can be linked with the numeric control and abort criteria can be defined. External influences that could have a negative effect on the measurement are not possible for this measurement principle.
The investigations give an insight to the different possibilities for an economic online-monitoring of the abrasive waterjet machining being independent from the operating conditions. Thereby the assets and drawbacks of each measurement method have been demonstrated.
Monitoring of the temperature changes of the cutting head is insofar not suited for the detection of irregularities inside the machining process as the process deviations can only be interpreted with a time offset.
A measurement of the acoustic noise and structure-borne sound is possible. However, the evaluation of the results is proportionally difficult. Moreover, perturbations of the surroundings can influence the measurement.
Air flow measurement is preferable due to the process security and economic feasibility. The suction behaviour inside the mixing chamber delivers sufficient information about the wear constitution of the nozzle components. In addition, blockages of the abrasive particle supply, which would inevitably cause an interruption of the machining process, can be detected by using the same measurement system.
 Sloan J, Kural C (2010) CFK-Bearbeitung ist eine riskante Sache. Maschinen Markt (2010) Nr. 35, S. 16-19.
 Shukla M (2013) Abrasive Water Jet Milling. In: Nontraditional Machining Processes: Research Advances, Davim JP, London: Springer pp. 177-204.
 Osman AH, Théry B (1998) Velocity Doppler Investigations of the Flow in Abrasive Water Jet (AWJ) Using a Fabry-Perot Interferometer. In: Proceedings of the 5th Pacific Rim International Conference on Water Jet Technology New Dehli, India.
 Westkämper E, Warnecke HJ (2002) Einführung in die Fertigungstechnik. Mit 9 Tabellen. 5. überarb. und aktualisierte Aufl. Stuttgart, Leipzig, Wiesbaden: Teubner (Teubner-Studienbücher: Technik), pp. 152-154.
 Awiszus B (2012) Grundlagen der Fertigungstechnik, 5., aktualisierte Aufl. Leipzig: Fachbuchverl. Leipzig im Hanser-Verlag, pp. 192-200.
 Hashish M (2003) Inside AWJ Nozzles. In: 2003 WJTA American Waterjet Conference. Houston, Texas.
 Knaupp M (1994) Erhöhung der Fertigungssicherheit und -qualität beim Hochdruckwasserstrahlen durch den Einsatz von Sensoren. Berlin [u.a.]: Springer (IPA-IAO-Forschung und -Praxis, Bd. 199).
 Neugebauer R (2012) Werkzeugmaschinen. Aufbau, Funktion und Anwendung von spanenden und abtragenden Werkzeugmaschinen. Berlin [u.a.]: Springer Vieweg (VDI-Buch), pp. 241-251.
 Hashish M (1991) Optimization factors in abrasive-waterjet machining. ASME Journal of Engineering for Industry 113, pp. 29-37.
 Singh PJ, Chen W, Munoz J (1991) Comprehensive evaluation of abrasive waterjet cut surface quality. Proceedings of Sixth American Waterjet Conference, Houston, Texas, pp. 139-161.
 Momber AW, Kovacevic R (1998) Principles of Abrasive Waterjet Machining. London: Springer.