Volumetric verification, a real solution in the verification of machine-tool of big dimensions

The high existent competitiveness in a sector like the one of the machine-tool (MH) the precision and the possibility of compensation of the errors is a competitive difference between manufacturers and users of the same, purchasing special importance in the mechanised of precision.

Of the distinct sources of error that affect to the precision of a machine-tool [1,2], are the geometrical errors those that greater contribution realizar to the total error of the machine with a 60-70% of the total error [3]. These can divide in errors cuasi-static, caused by the structure of the machine, and dynamic, caused by the movement of the husillos or vibrations of the structure. His characterisation and back compensation will come determined by the structural configuration, volume of work and software of control of each MH.

Traditionally, the geometrical verification has been widely used in the verification of MH and machines to measure by coordinates (MMC). This bases in the characterisation of the errors of each axis separately by means of the direct measurement of the errors to compensate in some determinate positions of the machine, independently of the cinematic model of the machine and of his sequence of movements [4]. The extrapolation of the functions of characterisation obtained by means of this method to the rest of the volume of work is a complex process. The need to use systems of dimensions known of high rank together with the considerable time of necessary manufacturing, does that this complexity was especially notable in machines of long able route to mechanise pieces of big dimensions; doing of the geometrical verification a slow and costly process. It is in this type machine, widely used in sectors like the wind, rail or nautics, where the volumetric verification by means of laser tracker provides big advantages in front of the geometrical verification.

The volumetric verification works from the measurement of the conjoint effect of the geometrical errors in each point from the cinematic model of the machine and the movement multi-axis of the same [5-7]. What translates a significant reduction of the necessary time in the capture of information in front of the geometrical verification. At the same time, the strong development experienced by the controls of CN of open architecture provides a favourable frame for the correction homogénea of the geometrical errors in all the volume of work of the MH, from the parametric functions of compensation obtained by means of the process of basic identification of the volumetric verification.

The strong demand of mechanised of pieces of big dimensions with big value added, experienced by the peak of sectors like the wind, rail or nautics, do of the volumetric verification by means of laser tracker the ideal technician of verification and improvement of precision in MH of long route. This presents a reduction of time and costs in front of other methods of verification.

The volumetric verification is a process of intensive identification of parameters from a model of optimisation no linear by means of the indirect measurement of the errors of the machine in his volume of work. This consists in the minimisation, by means of the software developed, of the difference between pairs of theoretical points entered by means of numerical control (CN) and real captured by the system of measure, in function of the possibilities and need of the technician to employ.

In machines of long visited the volume of work of the same will be a factor limitante of the utilisation of artefacts of traditional measure, reducing the possibilities of technicians to employ with which realizar the capture of points. The systems of measure that better adecuan to the route of the axles of this type of machine are the systems of long rank like laser tracker and laser tracers. The volumetric verification by means of LT bases in the characterisation of the geometrical errors of each machine through the mathematical model of his cinematic chain (figure 1).

From the configuration of the MH and his sequence of movements generates the equation of movement that allows to obtain the position of the tip of the tool regarding the nominal position programmed.

The difference between theoretical coordinates of the machine and the coordinates obtained by the laser tracker shows the conjoint influence of the geometrical errors of the machine in his volume of work, volumetric error (ev). The minimisation of east by means of different strategies of identification, functions of regression and criteria of convergence employed the software provides the functions of characterisation of each one of the geometrical errors (figure 2).

Notice that it treats of a mathematical compensation no physical of the errors, which compensates the conjoint effect of the geometrical errors in all the volume of work and no each one of them of independent way.

The characterisation of the errors from the indirect measurement of the same diminishes the necessary time in front of the direct measurement of the errors (figure 3). For this, the positioning of the machine realizar by means of a program CN configuring properly the time of stop in each position, and by means of the detection of this by part of the laser tracker takes a number of measurements predefinidas providing the LT the half value of all they.

The space disposal of the laser tracker with regard to the volume of work to verify comes determined by the structural configuration of the machine to verify. The laser tracker occupies the corresponding place to the piece inside the sequence of movements of the machine while the retro-reflector occupies the position of the tool during the mechanised.

The space distribution of the points to measure determines in function of the ranks of work more used by each machine during the mechanised. If the mechanised realizar in all the volume of work of the machine, realizar a distribution homogénea of points from which provides a global correction giving the same importance to all the zone of work. However, if the mechanised realizar fundamentally in a determinate zone, the space distribution of the points to measure centres in this zone which translates in a better compensation of the errors in this zone with regard to the rest (figure 4)

In any case, the measurement of the points has to realizar without that break the ray between LT and reflecting being necessary to move away the laser tracker of the volume of work to verify avoiding the no allowed twists by the cabezal of the LT, well simply moving away the tripod on which plants or by means of a profile in voladizo associated to the movement of the piece. Once checked that the LT is able to address the ray to any point of the volume of work, is necessary to check that it does not produce lost of the do by part of the reflector in east. For this has of reflectors with different diameters and different angles of entrance of the do of the laser (figure 5). A retro-reflecting conventional SMR of 2,5 inches has an angle of incidence of ±30º, whereas a reflector cateye presents angles of next incidence to the ±90º. Special quotation deserves the activate target of Api, retro-motorised reflector which orients the reflector automatically so that the do of the ray incida in the centre of the same allowing the capture of points in volumes of greater work.

During the phase of acquisition of data, the environingingmental conditions have to be the most stable possible avoiding thermal sensors variations during the measurement. These can produce like consequence of gradients of temperature because of currents of air, installations like heating or systems of lighting. It is necessary to procure have a temperature homogénea along the measurement so much in the structural components that define the cinematic chain of the machine as in the volume of work of the same. In the case to produce significant variations of temperature the structural elements modify his dimensions doing that the machine do not position in the same point at the beginning and at the end of the taking of data in front of a same coordinated nominal. Of the same way, if it produces a significant variation along the way that visits the do laser of the LT this influences in his wavelength and therefore in the coordinates of the point captured. The monitoring of the temperature during the taking of data in different points of the machine by means of sensors of temperature allows to know the changes of this with the aim to determine his influence in the coordinates of the points measured, but does not allow us compensate his effect (figure 6).

One of the most influential factors in the capture of data is the uncertainty of measurement associated to the system of measure employed.

Point Measured = Real Point ± Uncertainty measurement

The uncertainty of measurement is formed by the conjoint influence of systematic errors of constructive or environingingmental character, easily compensables, and random errors like the noise of measure of the LT. Consequence of the errors entered by the two encoders angular and the interferometer that provide the position of the point measured in spherical coordinates. The influence of the noise of measurement with the coordinates obtained is directly related with the distance between reflector and laser tracker. In front of a same error of the encoder angular the difference between the coordinates of a point with and without angular noise increases when increasing the distance of measurement. A lower contribution realizar the radial uncertainty determined by a linear relation between the reflecting distance-laser tracker but a random independent term (figure 7).

The influence of the noise of measure in the coordinates of the points captured, as well as the process of car-calibration of the LT do that the volumetric error of the machine was not formed only by the conjoint influence of the geometrical errors of the machine, hampering the characterisation of these when including in her a behaviour no achacable to the geometrical errors. The volumetric verification results espacialmente useful in machines of long route, where the contribution of the geometrical errors to the volumetric error is a lot of elder that the contribution of uncertainty of measurement. By means of a correct positioning of the laser tracker with regard to the points to measure reduces the influence of the uncertainty of measurement, being possible to apply this technician in machines of lower size when deleting the contribution of the noise of angular measure by means of technicians like the multilateración. From the information from three or more laser trackers (figure 8).

In this point present the results obtained in the volumetric verification of two machining centres with dimensional characteristics entirely different. Both realizar by means of the utilisation of an only laser tracker in an only position splitting of the same strategy of optimisation.

In the first place present the results of a fresadora of standard numerical control, to which realizar a compensation homogénea of his geometrical errors in all his volume of work. For this realizar a mallado homogéneo of his volume of work with 0 mm ≤ X ≤ 500 mm, 0 mm ≤ And ≤ 400 mm 0 mm ≤ Z ≤ 200 mm.

The results of the figure 9 show as it has produced a compensation homogénea of the conjoint effect of the geometrical errors along all the volume of work. It has reduced the half volumetric error of 63.9µm to 14,9µm, what equivale to a reduction of the error of 76.7%. Of the same way, the maximum error has reduced of 101.9 µm to 32.3 µm, what equivale to a reduction of 68.3%.

In this MH the volumetric error end is of the same order that the error of measurement achieving reduce the initial volumetric error in 76.7%. In front of this type of MH, in which the volumetric error end is of the same order that the error of measurement can arise problems of overfitting. The application of technicians of multilateración in suitable conditions can reduce the limit of optimisation improving the characterisation of the errors [8, 9].

In the figure 10 observe the results of the volumetric verification of a machine with a route of greater movement that in the previous essay. The volume of work of this machine-tool 0 mm ≤ X ≤ 1500 mm, 0 mm ≤ And ≤ 1500 mm -1490 mm ≤ Z ≤ 0 mm discretizó forming a mallado uniform with eleven points of measurement in each one of his axles.

By means of the practice of a process of volumetric verification reduces the volumetric error of the machine of an initial value in the process of identification of 503.4 µm to 54.2, what equivale to a reduction of 89.3%. The volumetric error of the optimisation is not formed only by the effect of the geometrical errors but also by other random errors, like the noise of measurement, or influences of the thermal sensors variations in the structural elements of the machine to verify. Reason whereby the error after the compensation is not equal to the volumetric error obtained in the process of identification.

To be able to compensate the influence of the geometrical errors, once obtained the functions of approximation of the different geometrical errors, and depending of the software of control that have the machine is necessary to realizar a treatment of the same so as to compensate the errors in the CN of the machine. If the software of the machine is able to work with parametric functions of compensation of each one of the errors of the machine this treatment will not be necessary. If this does not occur, is necessary to apply a process of postprocesado of the CN to use in the machine; by means of the use of a software and from the cinematic model of the machine compensates the effect of the geometrical errors generating a new CNC (figure 11). The postprocesado can realizar so much from a PC as from the software of the machine through a routine devoted.

The volumetric verification by means of laser tracker diminishes the time employed in the acquisition of data when realizar an indirect measurement of the geometrical errors of the MH, being these treaties later in off-line way allowing in the meantime the use of the machine to verify. Of the same way, results the most adapted procedure in the verification of machines of long route, allowing a characterisation in function of the zones of interest of the machine and deleting the zones of concentration of errors.

By means of the data captured in three or more distinct positions by means of 1, 3 or more laser tracker, in function of the environingingmental conditions in which it realizar the taking of data, and applying technical of multilateración, reduces the effect of the noise of measure of the laser tracker improving the characterisation of the errors being able to employ this technician of verification in MH with lower rank of work.

Bibliography

[1] R.Reamersh, M.To. Mannan, To.N. Poo (2000). Error compensation in machine tools Part I: geometric, cutting-force induced and fixture- dependent errors. Machine Tools & Manufacture 4. Pages1235-1256.

[2] R.Reamersh, M.To. Mannan, To.N. Poo (2000). Error compensation in machine tools Part II: thermal sensors errors. Machine Tools & Manufacture 40. Pages 1257-1284.

[3] N.To.Barakat, M.To. Elbestawi, To.D. Spence (2000). Kinematic and geometric Error compensation of to coordinate measuring machine. Machine Tool Manufacture International Jurnal of Machine Tools and Manufacture 40. Pages 833-850.

[4] Jenq shyong Chen, Tzy Wei Kou, Shen Hwa Chiou (1999). Geometric Error calibration of multi-axis machines using an car-alignment laser interferometer. Precision Engineering 23. Pages 243-252.

[5] S.Aguado, D.Samper, J. Santolaria, J.J. Aguilar (2011). Identification strategy of Error parameter in volumetric error compensation of machine tool based on laser tracker measurements. International Journal of Machine tool & Manufacture. Volume 53. Pages 160-169

[6] H.Schwenke, W.Knapp, H. Haitjema, To. Weckenmann, R.Schmitt, F.Delbressine (2008). Geometric Error measurement and compensation of machines-An update. CIRP Annals-Manufacturing Technology. Pages 660-675.

[7] S.Aguado, D.Samper, J. Santolaria, J.J. Aguilar (2012). Towards an effective identification strategy in volumetric Error compensation of machine tools. Measurement Science and Technology, 23,12pp.

[8] S.Aguado, D.Samper, J. Santolaria, J.J. Aguilar (2013). Influence of measurement noise and laser arrangement on measurement uncertainty of laser tracker multilateration in machine tool volumetric verification. Precision Engineering. 37, 929-943.

[9] S.Aguado, D.Samper, J. Santolaria, J.J. Aguilar (2013). Study of self-calibration and multilateration in machine tool volumetric verification for laser tracker Error reductions.PartB: Journal of Engineering Manufacture. DOI: 10.1177/0954405413511074.