Принципы нанометрологии (1027623), страница 62
Текст из файла (страница 62)
The vibrating end forms a ‘virtual’ tip that willdetect contact with the measurement surface while imparting very little force [41].279280C H A P T ER 9 : Coordinate metrologyThis probing method also addresses the problem caused by the surfaceforces.
After extensive investigation into the strength of the contact forces,the detection sensitivity can be tuned so that only a true surface contactregisters as a measurement, rather than probe interaction with surface forces.As such, vibrating probes have displayed an ability to repeatedly resolvesurface features of less than 5 nm [41]. However, even though currentvibrating probes exert low measurement forces and address the problem ofmeasurement errors due to surface interactions, they do not have the abilityto have the same probing interaction force (or compliance) in each axis.With the requirements of a high-accuracy micro-scale probe clearlydefined, developments have been made that address all of the main problemsfaced and extensive work is being carried out on integrating all theserequirements into one single system. Current research includes the development of a fully 3D vibrating microprobe at NPL [42].Extensive work has also been carried out to alter existing surfacemeasuring systems so that they can cope with the high-aspect-ratio featuresthat this new generation of micro-scale probes hope to address [43].
A devicedeveloped at PTB consists of an AFM tip attached vertically onto the end ofa standard AFM cantilever, shown in Figure 9.10. The device was able to takeinternal measurements of sidewalls on MEMS devices [44].FIGURE 9.10 Vertical AFM probe for MEMS sidewall investigation [44].Calibration of miniature CMMsA simple procedure for calibrating the form of the probe tip, which isrelatively easy for millimetre-sized probes, becomes a significant challengewhen dealing with probes of a few tens of micrometres diameter. Reversalmethods have been developed [25] and a stand-alone optical instrument hasbeen developed [45].
However, to date, accuracies of several tens of nanometres have been achieved – this will need to be improved for future, smallerprobes. The manufacture of ever-smaller probes is also a very difficult task.Drawn optical fibres [38] and micro-electro-discharge machining [46] techniques have been used to produce miniature balls on stems for probes but arefast approaching their manufacturing and materials limits [47,48].Optical methods have also been used in conjunction with miniatureCMMs. Optical probes are usually based on the same principles as those forsurface topography measurement (see section 6.7) and are summarized in[49]. Optical probes have also been developed that operate in a similarmanner to a tactile probe but with an interferometric output [50].9.6 Calibration of miniature CMMsMiniature CMMs suffer from geometric errors in the same way as large-scaleCMMs.
With accuracy goals being higher for the miniature CMMs,the importance of a proper calibration of the instrument increases. Thepurpose of the calibration is to map the systematic errors of the miniatureCMM, so that they can be compensated for. Some effects will perhaps notbe compensated, but they still have to be measured in order to assign anuncertainty contribution to them. If care is taken to ensure that all steps inthe calibration are traceable to the standard of length, this forms the basis forthe traceability of the miniature CMM as a whole.For large CMMs, it is customary for the manufacturer to performanceverify a CMM with gauge blocks according to ISO 10360 part 2 [11].
Theadvantages of gauge blocks as performance verification artefacts are that theycan be calibrated with low uncertainty (around 25 nm), and that their use inperformance verification is well established for large-scale CMMs. Because ofthe short shaft of miniature CMM stylus systems, it is typically not possibleto use the central length of the gauge block. Probing will be close to theedge of the gauge block, which should be taken into account in the initialgauge block calibration. If the gauge block is rotated out of the horizontalplane, the CMM probe can no longer reach the bottom face of the gaugeblock, and an additional surface has to be wrung onto the gauge block.Some specialized artefacts have also been developed for performance verification of miniature CMMs.
For one-dimensional verification281282C H A P T ER 9 : Coordinate metrologymeasurements, METAS (the NMI in Switzerland) has developed miniatureball bars [51] (see Figure 9.11a), consisting of ruby spheres connected bya ZerodurÔ rod. Spheres are widely used in artefacts for performanceverification, because measuring the relative position of spheres eliminateseffects from the probe diameter, shape and sensitivity, thereby allowing verification of the guidance error correction only.
However, the probe relatedeffects have to be verified in an additional test.Two-dimensional artefacts in the form of regular arrays of balls or holeshave been developed by PTB (see Figure 9.11b) [52]. PTB has also developeda 2D calotte plate and a 3D calotte cube (see Figure 9.11c and Figure 9.11d)[52]. As an option with the F25 miniature CMM, Carl Zeiss supplies a halfsphere plate with seven half spheres on a Zerodur plate (Figure 9.11e). Theuse of half spheres instead of full spheres gives better contrast in opticalmeasurements with a vision system.
By measuring a ball or hole plate indifferent orientations and using error separation techniques, it is possible toobtain the remaining errors of the CMM but not scale, without externalcalibration of the ball or hole positions.acbdeFIGURE 9.11 Miniature CMM performance verification artefacts. (a) METAS miniatureball bar, (b) PTB ball plate, (c) PTB calotte plate, (d) PTB calotte cube, (e) Zeiss halfsphere plate.Calibration of miniature CMMs9.6.1 Calibration of laser interferometer-based miniatureCMMsWith the calibration of the laser interferometers on a miniature CMM, thelength scale is established.
The following geometrical errors have to becharacterized in order to establish traceability:-cosine errors;-Abbe errors;-mirror shape deviations;-squareness errors.The cosine error is directly related to the quality of the laser alignmentrelative to the mirror normal (see section 5.2.8.3).
Abbe errors result fromparasitic rotations in combination with an offset between the probed positionon the object and the position where the measurement is taken. Abbe errorscan be minimized by moving the sample instead of the probe and having thevirtual intersection of the laser beams coincide with the probe centre (as on theNMM in section 9.4.1.2). The maximum Abbe offset that remains has to beestimated, in order to quantify the maximum residual Abbe error.
The rotational errors can be measured with an autocollimator or a laser interferometerwith angular optics (see section 5.2.9). The NMM (see section 9.4.1.2) usesdouble and triple interferometers to measure the angular deviations duringoperation and actively correct for them – this greatly reduces the Abbe errors.The mirror flatness can be measured on a Fizeau interferometer (seesection 4.4.2). The angle between the orthogonal mirrors can be measured byremoving the mirror block from the instrument and using optical techniques(for example, by comparison with a calibrated optical square).
It is alsopossible to calibrate the orthogonal mirror block directly, by extending it withtwo additional mirrors and calibrating it as if it were a four-sided polygon[53]. Alternatively, the squareness can be determined using a suitable calibration artefact on the miniature CMM (see below).9.6.2 Calibration of linescale-based miniature CMMsFor a linescale-based miniature CMM, such as the Zeiss F25 (see section9.4.1.1), the traceability is indirect via the linescales. The linescales areperiodically compared to a laser interferometer in a calibration. The calibrated aspects are the linearity, straightness and rotational errors. Thesquareness between the axes is determined separately, by a CMMmeasurement on a dedicated artefact.283284C H A P T ER 9 : Coordinate metrologyFor the linearity determination, a cube-corner retro-reflector is mountedin place of, or next to, the probe.
![](https://s.studizba.com/z.php?f=/templates/si/i/noavatar.png&w=100&h=100&t=1"){kind=avatar}
