CMM touch trigger performance verification using a probe test apparatus

CMM touch trigger performance verification using a probe test apparatus

P. A. Cauchick Miguel^I; T. King^II; A. J. Abackerli^III

^INúcleo de Gestão da Qualidade & Metrologia, Faculdade de Engenharia Mecânica e de Produção – FEMP, Universidade Metodista de Piracicaba, Rodovia SP 306, km 1, 13450-000 Santa Bárbara d’Oeste SP. Brazil. [e-mail]

^IISchool of Mechanical Engineering, The University of Leeds. [e-mail]

^IIINúcleo de Gestão da Qualidade & Metrologia, Faculdade de Engenharia Mecânica e de Produção – FEMP, Universidade Metodista de Piracicaba, Rodovia SP 306, km 1, 13450-000 Santa Bárbara d’Oeste SP. Brazil. [e-mail]

ABSTRACT

A probe test device has been developed to assess CMM touch trigger probe errors. A precise single axis translational table triggers the probe through the use of gauge blocks fixed to the table carriage. The position of the carriage is monitored by a laser interferometer. The gauge block surface is brought into contact with the probe ball tip until the probe is triggered at which point the coordinate position is displayed by the laser. The probe is rotated 360° using the motorized probe head in 7.5° intervals and the coordinate of each position is computed. Repeatability errors (i.e. the ability of the probe to trigger at the same point each time with a 95% confidence level) and pre-travel variation (i.e. the distance traveled by a probe between the actual touching of a surface and the trigger event) are then calculated. This paper reports the measurement strategy and the design of the prototype rig. Results are also provided to demonstrate the test apparatus performance.

Keywords: CMM probe, touch trigger probe, probe verification, probe test rig

Introduction

CMMs have proved to be very useful devices for dimensional inspection in manufacturing systems. There is scarcely a workpiece whose dimensions cannot be measured by them. CMMs have, however, inherent measurement uncertainty which is the result of error interaction of the individual machine components. Although it is of prime importance to the user to have an accurate estimate of the CMM performance, it is also of key importance that the manufacturer knows how the individual CMM components contribute to its overall performance. As any probe system is also a CMM error source, inaccuracy in the probe mechanism will affect measurement results. Consequently, the nature of probing related errors must be known.

There has been much research towards establishing the probe performance over the past 15 years (Lotze, 1981; Nawara and Kowalski, 1984; Traylor and Jarman 1990; Butler, 1991; Deni, 1992; Phillips, Borchardt and Caskey, 1993; Overs, 1994; Davis et al., 1995). A survey of methods for checking probe performance is provided by Cauchick-Miguel et al. (1998). Probe tests are usually performed by measuring a reference sphere of known size and sphericity. Because the machine is, of course, driven to contact points distributed over the surface of the sphere, results from these tests include the contribution from machine errors (Butler, 1991). As it is difficult to isolate the probe system from the rest of the measuring system, probe tests cannot be regarded as fully representative of the probe performance alone (Peggs, 1991). Therefore, in order to have a through understanding about probing errors, it is essential to characterize the probe independent of machine errors. None of the current methods consider checking the probe system using a 'CMM-independent' method. In fact, there is a probe test device for calibrating analogue probes at the NPL (National Physical Laboratory) but it is not suitable for touch trigger probes at present (Leach, 1995). Although it would be possible, in principle, to adopt the same idea for checking touch trigger probes, the implementation would be much more difficult (Peggs, 1991). Therefore, any advance in this direction would be very useful as touch trigger probes are the most widely used in the field of coordinate metrology. This paper reports the development of a test apparatus to be used for checking touch trigger probes, described in the following sections. The introduction should contain information intended for all readers of the journal, not just specialists in its area. It should describe the problem statement, its relevance, significant results and conclusions from prior work and objectives of the present work.

Design of the Probing Test Apparatus

Owing to the fact that probe testing methods involving artefacts, whilst useful, are not always indicative of the probe performance as a completely independent functional unit, a probe test rig has been developed. The probe tested in the experiments was an omni-directional touch trigger probe. The full measurement strategy and design considerations are discussed in the following sections, and results are provided to demonstrate its feasibility.

The required hardware of the probe test rig consists of a motorized traversing table coupled with a laser interferometer which enables the laser to be employed as the table measuring scale. A gauge block, is then fixed on the traversing table providing a reference surface for triggering the probe. Figure 1 shows a photograph of the first prototype of the test apparatus, while Figure 2 illustrates the schematic arrangement of the probe test rig including all components.

The components of the proposed test rig can be basically divided into; a translational stage with controller, a laser interferometer, and a probe interface. A commercial motorized stage with a travel range of 50 mm was used as the moving element for triggering the probe. The design of the carriage utilizes linear ball bearings with twin bearings track disposed laterally on each side of the carriage slides. This arrangement helps to control dynamic stability and run out. It has a dc servo motor to propel the carriage along the table base. The table is interfaced to a controller providing forward and reverse traversing movement. The system also provides home position with automatic origin search.

The laser interferometer is linked to the probe system to provide a signal for reading the laser display, as shown in Fig. 2. When the surface of the gauge block touches the probe ball tip, the signal from the probe is sent to the computer via a PCMCIA card. The system employed the laser facility which allows data to be captured by the laser system upon receipt of a trigger signal initiated from the machine under test (called 'TPin'). This may be provided, for example, from a relay within the CMM controller, via the probe interface module. The relay contacts are wired to the 'probe connection' of the interface module and the TPin input of the laser software should be connected to the 'output' connection. Data capture may be triggered either asynchronously or synchronously. In both modes, the trigger signal must be a clean debounced TTL, CMOS, or solid state relay signal. The mode used was TPin synchronous data capture. This mode provides a faster hardware trigger facility than the asynchronous mode, which greatly reduces the delay between the leading edge of the TPin pulse and the instant that the laser reading is recorded. The degree of synchronisation depends upon the specific laser used. In the experiment the data were recorded within ± 0.5 ms of the falling edge of the TPin pulse using a 1m/s laser system. The synchronous TPin mode is enabled from the menu in the laser software.

The motorized probe head is used for rotating the probe to successive angular positions. In order to retain the function of the head while the probe is triggered, it is necessary to link the probe interface unit to the probe head controller (a controller housed in the CMM cabinet). The interface unit used is capable of automatic recognition and interfacing touch trigger probes.

Figure 3 shows the necessary links between the laser interferometer, probe head controller, and probe interface unit. The main advantage of this configuration is the good compatibility between all components since they are produced by the same manufacturer.

Experimental Set up of the Test Rig on the CMM

The set up of the experiment comprises the alignment of the stage, retroreflector, laser head, and the gauge block. First, the table was set up on the CMM table by aligning it parallel to the X-axis using two reference points measured by the CMM. Second, the reflector was also aligned with the table and consequently with the X-axis direction of travel. Third, the laser head was properly adjusted in relation to the table carriage (to which the retroreflector was fixed). Then, the linear interferometer (the beam splitter bolted to another reflector) was placed in a fixed position between the laser head and the moving retroreflector. After fine adjustments, a maximum laser strength signal was obtained. The translational table could then be moved incrementally over the whole traverse range while readings were taken by the laser. Although the experiment has been set up in a controlled environment (temperature: 20° C + 1° C and relative humidity: 50% ± 5%), air and material temperature sensors were placed at the test area. The temperature variation was not likely to be significant since the temperature stayed constant during the experiments (0.31° C over a period of 2 hours and 0.79° C over a longer period of 7 hours). Pressure and relative humidity were also taken from the environment unit in order to automatically compensate variation in the refractive index of the air. Finally, the gauge block was aligned (using the CMM) perpendicular to the table's traverse movement. In all cases, the alignment was considered good enough when the CMM coordinates of the reference points presented the same value. The test sequence is presented in Fig. 4. It is worth mentioning that the procedure is repeated 10 times (probing-rotating) in order to gather the error in the head as well.

Error Sources in the Probe Test Rig

There are several sources of errors associated with the proposed test rig. Some of them are related to the motorized table itself and others to the laser system. Some of these error sources have been described, in detail, in other publications (Steinmetz, 1990; Cauchick-Miguel et al., 1996; ISO 1993). The result of the analysis of the total error budget associated with the probe test rig is the combined uncertainty. This uncertainty, denoted by Uc, can be thought of as representing the measuring uncertainty resulting from combining all known sources of uncertainty in a root sum of squares (RSS) manner. This approach is based on the CIPM method (ISO, 1993).

Results of Probe Performance Verification

The proposed probe test rig must be capable of verifying the main probe error sources, i.e. repeatability and pre-travel variation. Experiments have been carried out to quantify both. The results are presented in the following topics.

Repeatability

For the probe repeatability test, a grade '0' 20 mm gauge block, was fixed to the surface of the mounting support of the reflector (as illustrated in Fig. 5). The test consisted of measuring the distance from the table origin to the point where the probe was triggered. The table was moved forward until the gauge block touched and triggered the probe. The laser display reading was taken and the table was moved back to its home position. The probe head was then rotated through a 7.5° interval and the stage moved forward to trigger the probe again. This cycle was repeated until the probe head has been rotated by 360° of indexing angle (in fact, + 180° ) making a total of 49 points (readings at both -180° and +180° provide redundancy for double checking). The collected data were used to calculate the standard deviation for each indexing angle.

Figure 6 presents the results from probe repeatability of a touch trigger probe with a 20 mm stylus length using the test rig. As can be seen in the figure, the repeatability varies for different approach directions but is 0.24 mm on average. Each bar in the graph in this figure represents a standard deviation for ten runs. To establish the number of repeated runs required, a test was carried out to observe the data variation with an increasing number of measurements. Based on the results, the number of repeated runs was selected to be ten, since no significant difference existed between 10, 20, 30, or 40 measurements.

Due to the fact that the probe system is being verified using a method which is independent of the CMM, it was expected that the results obtained from conventional (on CMM) probe repeatability tests would be bigger that those obtained using the probe test rig. This indeed occurred as can be seen in Fig. 7. The same gauge block in the same position was touched ten times but this time by driving the CMM using a part program. The experiment was performed by probing the gauge block at different indexing angles aided by rotating the motorised probe head. It is quite clear that the repeatability error is smaller when applying the proposed method than that when 'measuring' through the CMM. In fact, the repeatability at all indexing angles was smaller when checking the probe using the test rig. This suggests that machine errors (and possibly other sources) are affecting the results of the probe system itself. In the case of using the test rig it is obviously possible to ensure that CMM errors are not influencing the results of the probe test.

Under the same conditions, the repeatability error increases dramatically when longer styli are used. Figure 8 shows that the repeatability is adversely affected by increased stylus length. Once again, the tests were conducted at different approach directions (indexing angles) using the probe test rig aided by the motorised probe head. These results are in accordance with McMurtry (1994) who stated that the probe repeatability should be within 0.35, 1, and 2.5 mm for 10, 50, and 100 mm stylus length, respectively. However, caution should be taken when comparing the results because the test conditions are probably different. Repeatability defined by the probe manufacturer (Renishaw, 1995a) is equal to two times the standard deviation. Professor David McMurtry is the inventor of the touch trigger probe (see US patent in McMurtry, 1979). The results of probe repeatability stated by him (and other sources associated with the probe manufacturer; e.g. Taylor and Jarman, 1990) were probably obtained with some sort of probe test rig.

The reason for it, is that the hysteresis effects, mainly caused by friction at the seating points and the rubber diaphragm (McMurtry, 1994), are magnified by elastic deflection of the stylus. An interesting aspect of the results indicated in Fig. 8 is that the trends for 50 and 100 mm stylus length are quite similar, suggesting a regular pattern as the indexing angle is varied. There are obviously some dissimilarities which could be attributed to distinct probe bending effects, strongly dependent on the length of the stem. The amount of probe bending depends on the stem material, dimensions (diameter and length) and magnitude of the force applied.

Pre-travel Variation

Another important source of error in touch trigger probes is the pre-travel variation. In some cases it can be a major source, e.g. when the machine does not have the capability for probe error compensation (which in fact, the vast majority of CMMs do not, according to McMurtry, 1994). Basically, the pre-travel tests use the same procedure as for the repeatability tests. The test is performed using all indexing angles so that the laser readings for each angle can be used to obtain the pre-travel variation. As described before, the probe is fixed to the motorized probe head for controlling the spatial orientation, i.e. for triggering the probe at different indexing angles. However, the position of the probe ball tip can vary when rotating the head, as illustrated in Fig. 9. This can be caused by an error in the head itself and due the fact that the probe-stylus configuration might not be straight (the error in the head is quoted by Renishaw, 1995a, as being ± 0.5 mm). To standardize the nomenclature of errors in that rotary axis, angular errors will be termed tilt motions (in accordance with Bryan and Vanherck, 1975). Previous test results with the probe test rig have shown that there is a tilting error when the probe is indexed at the required positions (Cauchick-Miguel, King and Abackerli, 1996). Therefore it was necessary to verify these errors and possibly to correct the probe pattern results which are affected by them.

In order to obtain the pre-travel results, the tilting errors had to be eliminated so that an alternative procedure was developed to do this. The test was based on 'measuring' a gap between gauge blocks. Three gauge blocks, grade 0, were wrung so that the reference faces of the block on each end protruded from the central block by about 5 mm. This made a 'U'-shaped gap to be 'measured' by the laser. The central gauge block was a 9 mm nominal length with an error of +30 nm. The other two gauge blocks used had nominal dimensions of 3.5 and 5.5 mm. This set up is shown in Fig. 10. By moving the stage, the probe will touch one of the faces of the gauge blocks. When this occurs, the probe interrupts its internal circuit and a pulse is generated from the probe interface to the laser computer.

Figure 11 shows a schematic arrangement of the gauge blocks where u and v are the laser readings on both sides, pd is the probe diameter and L is the length to be measured. It is worth recalling that the gauge blocks were aligned perpendicularly to the table movement using the CMM. This alignment is considered good enough for the test. The alignment can be regarded suitable when the CMM readings are the same. However assuming a deviation of 2 µm (e.g. due to the probe and CMM errors) for a length of 25 mm between the alignment points, an error of less than 1 nm would be transferred to the length of the gap, which is in fact negligible.

By moving the table back and forward, the probe will be touched by one of the faces of the gauge blocks at each indexing angle so that the laser display can be read. The cycle is repeated until the desired number of readings has been taken after reading the pre-established angles. The data collected for a particular indexing angle are subtracted (positive and negative laser display) and the probe nominal diameter is summed up, according to equation (1) presented in Fig. 11. Then, a difference between the calculated value (found by the gap 'measurement') and the true value of the gauge block was computed. When these differences are plotted the lobbing pattern will appear, as shown in Fig. 12.

Actually, Fig. 12 shows a six lobed pattern. This happened due to the fact that the gauge blocks were touched on both (opposite) sides of the ball tip. Therefore, for each indexing angle readings at 180° were obtained (which could be considered as a sort of 'complement' to each other). The maximum value is where the kinematic points are located. At that position, maximum values as well as minimum values 180° apart were obtained. A force diagram shown in Fig. 12 (dotted line), helps to identify the position of the kinematics points. The force diagram can be obtained by using a gram gauge (see Renishaw, 1995a). It can be seen that the maximum value is obtained in the highest value of force and the minimum in the lowest one. This demonstrates that the procedure is able to isolate the lobbing effects from the tilting errors in the head. In that case, the magnitude of pre-travel variation is 2.9 mm.

There is a sensitivity to stylus length in pre-travel variation. Figure 13 shows an increase of pre-travel when the stylus length is enlarged. It also depicts pre-travel tests using the CMM in comparison with those performed with the probe test rig. When comparing such results, higher pre-travel variation was found in the probe tests using the CMM. These results suggests that the probe errors are being superimposed by those of the CMM.

Discussion

The performance of touch trigger probes is characterized by their repeatability and pre-travel variation. The latter is the most relevant and includes the former as most CMMs today do not have the ability to error map a probe. However, there are some new CMMs that have the ability to error map probes, known as vector qualification, in which case the former is the most relevant as lobbing errors are handled within the CMM software, leaving repeatability as the major source of uncertainty associated with the probe system. However, whichever one is the most important, both must be assessed either for checking whether the probe is within manufacturer's specification or to use probe test results for error compensation (in case of pre-travel variation).

For some CMM measurement tasks, the repeatability can be the only error source. For example, take the measurement of the centre-to-centre distance between two bores. The repeatability is supposed to be the only measurement error because the pre-travel variation is present while determining the centre of each bore. However, it must be assumed that both bores are measured using the same measuring strategy so that the calculated centres will be displaced by the same amount in the same direction. Consequently, the relative distance between the centres will solely be influenced by probe repeatability.

The results of repeatability tests for a 10 mm and 50 mm stylus length were similar to those found by Traylor and Jarman (1990). Although not clearly stated in their publication, the findings suggest the use of some sort of probe test device. The main reasons for this assumption is due to the magnitude of errors assessed in the probe in addition to the fact that the authors (Traylor and Jarman, 1990) are from the probe manufacturer in the US and in the UK, respectively. The other error present in touch trigger probes is hysteresis. This error has not been checked since its magnitude tends to be of the same order as repeatability (Bryan and Vanherck, 1975; Oakes, 1993).

Regarding the pre-travel variation, the test results are not very consistent with other publications (e.g. Traylor and Jarman, 1990; Overs, 1994; Renishaw, 1995a). However, the values for 10 and 50 mm stylus length were found to be similar, but slightly higher, than those suggested by the probe manufacturer (Renishaw, 1995a). The results were 0.4 mm bigger on both stylus lengths. This discrepancy may be caused to specific performance of the probe tested, since the probe performance is supposed to deteriorate with its life. According to McMurtry (1979) even though great improvements have been made in probe life over the years, kinematic resistive probes do not have an infinite life. In fact, the life expectancy is not much greater than 10⁶ triggers. Nevertheless, a number of other published works have reported different results in pre-travel tests (Traylor and Jarman, 1990; Deni, 1992; Overs, 1994; Renishaw, 1994; 1995a; 1995b; Krejci, 1990).

Another achievement of this work is related to the tilting errors in the motorized probe head. Because the probe head had to be used for providing rotational motion of the probe, further errors were introduced. These errors are related to the accuracy of probe reallocation and articulation (called tilting errors). Besides, many CMMs include a motorized probe head so that it is worth quantifying tilting errors and understanding how they affect some CMM measuring tasks. As previously mentioned, the indexable probe head is capable of finite increment rotations about two axes. In each indexing position used, the probe must be qualified which determines the size and location of the stylus ball. Supposing the CMM task is to measure the diameter and concentricity of two opposite ball bearing seats (see Fig. 14). The result of the diameter measurement will depend upon establishing the stylus size (effective ball tip diameter) whereas the centre-to-centre distance will be dependent on the errors in the offset vector, as the probe head will have to be indexed in different positions (in this particular case using both axes of the probe head).

The probe qualification procedure effectively calibrates out the errors in the probe head rotary axes provided the errors are repeatable. The head manufacturer quotes such errors in the range of ± 0.5 mm (Renishaw, 1995a) which can be regarded as small enough concerning current CMM measurement tasks. Nevertheless, because tilting errors in the head are systematic, if those were quantified and used for probe error mapping, the qualification procedure for the required indexing position could be eliminated. This is particularly important when considering the use of CMMs in a highly automated environment, e.g. when the CMM is part of a FIS (Flexible Inspection System). Therefore knowing tilting errors in the probe head is not only important regarding the experiments using the test rig but also beneficial for further CMM applications (e.g. probe error compensation).

Finally, analyzing the way in which potential error sources within a probe may combine and manifest themselves in a measurement is not a trivial task. The final judgement will be made by the user in order to establish if the error that the probe is contributing would be a minor part of the measurement tolerance required. This can be considered as a difficult decision since the measurement tasks have been becoming more complex and the part tolerances even tighter. One might argue that probe errors are small when compared to typical part tolerances inspected by CMM. However, under certain circumstances, such as when using long styli, the probe can contribute significantly to the uncertainty of measurement. Therefore, in addition to the fact that current CMM designs have the ability to error map parametric errors, probe uncertainties must be understood and quantified.

Conclusions

The probe test rig described in this paper has served the purpose of touch trigger probe verification within a required accuracy (uncertainty of one order of magnitude relative to the expected results). Probe repeatability and pre-travel variation have been assessed and found to be smaller than the results from CMM tests. The systematic error result, i.e. the pre-travel variation, can now be used as a basis for compensating probe errors by the CMM' software. Further work will involve the application of the obtained data for probe error correction. The probe test apparatus has been proved very useful to the CMM users for checking their probes in the acceptance or in interim checking.

Acknowledgements

The authors would like to thank Professor Ken Stout, a former head of the Centre for Metrology at the University of Birmingham, UK, for providing CMM facilities and Mr Bob Aston and Mr Paul Saunders for their technical support. Additional thanks to Dr Jim Davis who helped the authors with some suggestions in this research. Thanks are also due to the CNPq Brazilian Government Agency for funding two of the authors.

Paper accepted March, 2003

Technical Editor: José Roberto de França Arruda

• Bryan, J.B. and Vanherck, P., 1975 "Unification of Terminology Concerning the Error Motion of Axes of Rotation". Annals of the CIRP, Vol. 24/2, pp. 555-562.
• Butler, C., 1991 "An Investigation into the Performance of Probes on Coordinate Measuring Machines". Industrial Metrology, Vol. 2, No. 1, pp. 59-70.
• Cauchick Miguel, P.A. and King, T.G., 1997 "Aspects of Design and Uncertainty an Opto-mechanical Device for Checking CMM Touch Trigger Probes". Proceedings of the International Conference on Integrated and Sustainable Industrial Production, Lisbon, Portugal, pp. 389-396, May.
• Cauchick Miguel, P.A., King, T.G. and Abackerli, A.J., 1998 "A Review on Methods for Probe performance Verification". Measurement Journal, Vol. 23, pp. 15-33.
• Cauchick-Miguel, P.A., King, T.G. and Abackerli, A.J., 1996 "Error Assessment of a Motorised Traversing Table for Using in CMM Probe Performance Verification". Proceedings of the Second Annual Postgraduate Research Symposium, School of Man. & Mech. Eng., The University of Birmingham, Birmingham, UK, pp. 1-4, April.
• Davis, E.J. Liao, W.J., Scott, G.F. and Sout, K.J., 1995 "Sources of Measurement Uncertainty When Using Coordinate Measuring Machines: A Top-down Systematic Approach". 2nd Laser Metrology and Machine Performance, Southampton, pp. 185-201.
• Deni, M., 1992 "Tough Trigger Probe Accuracy Verification". CMM News, NPL- National Physical Laboratory, Issue 1, Autumn.
• ISO Guide to the Expression of Uncertainty in Measurement. International Organization for Standarization, Switzerland, 1993.
• Krejci, J.V., 1990 "CMM Measurement Enhancement Using Probe Compensation Algorithms". SME Technical Paper MS90-09, Society of Manufacturing Engineers.
• Leach, R.K., 1995 "A Novel 3-axis CMM Probe Calibration Facility". NPL Report MOM 126, National Physical Laboratory, Teddington, April.
• Lotze, W., 1981 "Prufkorper fur Mehrkoordinaten-Messgerate". Feingeratetechnik, No. 4, pp. 154-156, January.
• McMurtry, D.R., 1994 "Probing Present and Future". Two Days Seminar, University of Hudderfield, Hudderfield, UK, 18-19 April.
• McMurtry, D.R., 1979 "Touch Trigger Probe". United States Patent 4,153,988, May.
• Nawara, L. and Kowalski, M., 1984 "The Investigations on Selected Dynamical Phenomena in the Heads of Multi-coordinate Measuring Devices". Annals of the CIRP, Vol. 33, No. 1.
• Oakes, S., 1993 "CMM Probing Technology". Renishaw Internal Report, Renishaw plc, May.
• Overs, A., 1994 "Probe Performance - Optimise or Compromise". Quality Today, pp. s1-s4, January.
• Peggs, G.H., 1991 "A Review of the Methods for the Accurate Metrology of Complex Three-dimensional Components and Artefacts". NPL Report MOM 101, National Physical Laboratory, Teddington, March.
• Phillips, S.D., Borchardt, B. and Caskey, G., 1993 "Measurement Uncertainty Considerations for Coordinate Measuring Machines". NISTIR-5170, National Institute of Standards and Technology, Gaithersburg, April.
• Renishaw, 1994 "Data Sheet Part No. H-1000-2104-01-A. G F High Performance Stylus Range". Renishaw plc.
• Renishaw Technical Specifications Part No. H-1000-5050-07-A., 1995a "Probing Systems for Coordinate Measuring Machines - Technical Specifications", Renishaw plc.
• Renishaw, 1995b "TP 200 Probe System - A New Concept in Probing Technology". Renishaw plc.
• Steinmetz, C.R., 1990 "Performance Evaluation of Laser Displacement Interferometry on a Precision Coordinate Measuring Machine". Industrial Metrology, Vol. 1, No. 3, pp. 165-191.
• Traylor, A. and Jarman, T., 1990 "Performance Characteristics of Touch Trigger Probes". SME Technical Paper MS90-266, Society of Manufacturing Engineers.

Publication Dates

• Publication in this collection
  18 Mar 2004
• Date of issue
  Apr 2003

History

• Accepted
  Mar 2003