CMM inspection fundamentals This presentation outlines the key choices facing a CMM user who needs to specify a probing system. Key questions: Do my measurement applications require a scanning solution? If so, what is the scanning performance of a system? Scanning accuracy at high speeds Total measurement cycle time, including stylus changes If I also need to measure discrete points, how fast can I do this? Will I benefit from the flexibility of an articulating head Access to the component Sensor and stylus changing What are the lifetime costs? Purchase price What are the likely failure modes and what protection is provided? Repair / replacement costs and speed of service The factors that affect CMM measurement performance and your choice of probing solution Issue 2
Which inspection solution will suit your application? Probing applications Touch-trigger or scanning? Dynamic effects on scanning performance Articulation or fixed sensors? The unique needs of your business will drive your choice of dimensional measurement solution. There is no single solution that will meet the needs of every measurement application. Your choice will depend on a number of factors, including speed, accuracy, flexibility and cost. Stylus changing or sensor changing? Active or passive scanning?
Probing applications - factors Manufacturers need a range of measurement solutions. Why? Machining processes have different levels of stability: Stable form : therefore control size and position Discrete point measurement Form variation significant : therefore form must be measured and controlled Scanning How stable are your manufacturing processes? In general, you should measure your components only as often as required to ensure the stability of your manufacturing processes. In reality, this involves focussing on key features on your components - those that are critical to its function - to work out the best process control strategy. Your choice of manufacturing technique for these features will be a critical factor in the choice of process control method: If, for example, you machine a critical bore with a process that reliably produces features with good form, it's size or position may vary. In this case, control of the size and position will be important, but not control of the roundness or cylindricity. By contrast, if you use a machining process that produces features with significant form variation (i.e. the variability of the form is a significant proportion of the form tolerance), then understanding form errors will become important.
Probing applications - factors Manufacturers need a range of measurement solutions. Why? Features have different functions: for clearance or location form is not important Discrete point measurement for functional fits form is critical and must be controlled Scanning Features that have functional fits Some features on your components will need to mate with other parts for your product to work correctly. In many cases, the form or profile of these features will be critical to the functional fit. This is where scanning is the ideal measurement technology. Other features Most of the features on your parts will not have such exacting tolerances. Many will be clearance holes, location features or drilled holes. In these cases, measuring discrete points might be the best approach, since it can be faster and simpler than scanning such features, and it avoids excess stylus wear. The pragmatic approach Typically you will want to use a combination of scanning and discrete point measurement to control your production processes. On most parts, there will be some features that you want to scan, whilst many others can be controlled by discrete point measurement. This practical approach will give you the right amount of information for control, without incurring unnecessary measurement time and expense. Measured values Best fit circle Maximum inscribed (functional fit) circle
Discrete point measurement Ideal for controlling the position or size of clearance and location features Data capture rates of 1 or 2 points per second Avoids stylus wear Touch-trigger probes are ideal lower cost, small size and great versatility Scanning probes can also be used passive probes can probe quickly active probes are slower because the probe must settle at a target force to take the reading The majority of features on a part are likely not to require tight tolerances for the proper function of the component. Where a feature has a clearance fit or where it is used just to locate the part, then only simple parameters such as size or position need to be controlled. Typical features that fall into this category are drilled or tapped holes, holes that carry fluids (e.g. coolant holes), clearance holes and non-mating surfaces. Discrete point measurement is an effective way to measure these types of features. This involves locating a small number of surface points quickly, which define each feature (typically four points for a bore, for instance). It is fast, simple and flexible, and provides an accurate assessment of feature size and location. Taking discrete points also minimises stylus wear, since the stylus usually approaches the surface on a normal vector, rather than being dragged over the surface during scanning. Both of the two main types of CMM probes can be used to measure discrete points: Touch-trigger probes are the simplest form of sensor for CMMs. They are ideally suited for discrete point measurements and have the advantages of low cost, small size and great versatility. However, they are not the best choice for high point density scanning or digitising applications. Scanning probes can also be used to acquire discrete points at a similar rate to touch-trigger probes if extrapolate to zero routines are used. Furthermore, scanning sensors provide the extra flexibility of scanning and digitising. If a high proportion of the features will be measured with discrete points, then the speed of single point data capture is critical.
Discrete point measurement Speed comparison The movie clips show discrete point measurement using both touch-trigger and scanning probes. With appropriate discrete point measurement routines, scanning probes can measure points at a similar rate to touch-trigger probes, governed mainly by the dynamic capabilities of the CMM. Active sensors tend to be slower at discrete point measurement since they have to settle at the calibrated force and ‘static-average’ to take the reading. Of course, scanning probes can provide far higher data capture rates than touch-trigger probes when scanning. Touch-trigger probes are ideal for high speed discrete point measurement Scanning probes can also measure discrete points quickly, and provide higher data capture rates when scanning
Scanning Ideal for controlling the form or profile of known features that form functional fits with other parts Data capture speeds of up to 500 points per second Incurs wear on the stylus Scanning allows you to: Determine the feature position Accurately measure the feature size Identify errors in the form or shape of the feature Scanning allows you to gather a lot of data very quickly. This data can be used to determine not just the size and position of features on your components, but their form as well. With a scanning inspection system you can acquire several hundred points per second, rapidly gaining an accurate understanding of the surface of your components. This is ideally suited to features where the tolerance of form is a significant proportion of the total tolerance. Features which must form a functional fit with another component benefit most from scanning. Where form is not important (on clearance features, for instance), discrete point measurement is generally preferred.
Scanning Scanning a cylinder block Typical scanning routine, measuring precision features where form is critical to performance Scanning should be used where measurement of form is required. This example shows inspection of an F1 cylinder block, in which precision surfaces are critical to the performance of the race engine. Scanning provides much more information about the form of a feature than discrete point measurement
Digitising Ideal for capturing large amounts of data about an unknown surface Uses many of the same techniques as scanning Deflection vector of the probe is used to determine the motion vector in which the machine moves next Digitised surface data can be: Exported to CAD for reverse engineering Used to generate a machining program for re-manufacture Like scanning, digitising is best performed using a scanning probe since the amount of data required is very high. Whilst discrete point measurement techniques can be used, these are very much slower. Digitising uses many of the techniques needed for scanning, except that the motion of the CMM is controlled in a different manner. On described parts, the probe can move in a pre-defined path, accommodating any surface deviations. By contrast, on an unknown part the probe is moved within a pre-defined area and the probe deflection vector is used to determine which way to move the CMM to keep the probe stylus in constant contact with the surface. Digitising is used for re-manufacture and reverse engineering where a master part must either be replicated or converted into digital form.
Digitising Re-manufacture and reverse engineering Digitising a master part to acquire an accurate description of the surface Scanning cycle and data analysis handled by Tracecut software Digitising can be performed on CMMs, machine tools or dedicated platforms like Cyclone This digitising routine shows a typical master component being measured as a series of scan lines across the surface, with a small step-over between each line. The motion of the machine is driven by the deflection of the scanning probe as it responds to the shape of the surface. Using Renishaw’s Tracecut software, a series of grids can be set-up to cover the component. As data is captured it is filtered and a picture of the surface is built up. Tracecut can export surface data to CAD, plus it can produce a machining program to produce a mould or stamping tool to reproduce the master part. Digitising provides large amounts of data to define unknown contoured surfaces
Which inspection solution will suit your application? Probing applications Touch-trigger or scanning? Dynamic effects on scanning performance Articulation or fixed sensors? Scanning pros: Ideal for profile and form measurement Very high speed data collection High data point density ensures better repeatability and reproducibility (R&R) of measured features and stability of workpiece datums More comprehensive metrology analysis - form as well as size and position Scanning cons: More complex that touch-trigger sensors Slower discrete point measurement than touch-trigger probes Dynamic effects due to accelerations during measurement must be compensated if high speed scans are to produce accurate measurement results Wear on the stylus Touch trigger pros: Ideal for inspection of 3D prismatic parts and known surfaces Highly versatile - suitable for a diverse range of applications Wide range of probes and accessories available Lower cost of purchase and lower running costs than scanning systems Simple to use Ideal for size / process control type applications Robust Smaller than scanning probes, aiding access to deep features Machine dynamics have less effect on measurement accuracy Touch trigger cons: Limited data collection rate Stylus changing or sensor changing? Active or passive scanning?
Ideal applications Scanning Touch-trigger Measurement of size, position and form of precision geometric features Measurement of profiles of complex surfaces Inspection of 3D prismatic parts and known surfaces Size and position process control applications where form variation is not significant Scanning pros: Ideal for profile and form measurement Very high speed data collection High data point density ensures better repeatability and reproducibility (R&R) of measured features and stability of workpiece datums More comprehensive metrology analysis - form as well as size and position Scanning cons: More complex that touch-trigger sensors Slower discrete point measurement than touch-trigger probes Dynamic effects due to accelerations during measurement must be compensated if high speed scans are to produce accurate measurement results Wear on the stylus Touch trigger pros: Ideal for inspection of 3D prismatic parts and known surfaces Highly versatile - suitable for a diverse range of applications Wide range of probes and accessories available Lower cost of purchase and lower running costs than scanning systems Simple to use Ideal for size / process control type applications Robust Smaller than scanning probes, aiding access to deep features Machine dynamics have less effect on measurement accuracy Touch trigger cons: Limited data collection rate
Speed and accuracy Scanning Touch-trigger High speed data capture - up to 500 points per second Large volume of data gives an understanding of form High point density gives greater datum stability Dynamic effects due to accelerations during measurement must be compensated if high speed scans are to produce accurate measurement results Slower data capture rate Less information about the surface Simple calibration of probe and machine yields accurate point data Dynamic performance of the machine has little impact on measurement accuracy since probing is performed at constant velocity Scanning pros: Ideal for profile and form measurement Very high speed data collection High data point density ensures better repeatability and reproducibility (R&R) of measured features and stability of workpiece datums More comprehensive metrology analysis - form as well as size and position Scanning cons: More complex that touch-trigger sensors Slower discrete point measurement than touch-trigger probes Dynamic effects due to accelerations during measurement must be compensated if high speed scans are to produce accurate measurement results Wear on the stylus Touch trigger pros: Ideal for inspection of 3D prismatic parts and known surfaces Highly versatile - suitable for a diverse range of applications Wide range of probes and accessories available Lower cost of purchase and lower running costs than scanning systems Simple to use Ideal for size / process control type applications Robust Smaller than scanning probes, aiding access to deep features Machine dynamics have less effect on measurement accuracy Touch trigger cons: Limited data collection rate
Complexity and cost Scanning Touch-trigger More complex sensors, data analysis and motion control Higher costs than basic touch- trigger systems Conventional systems have higher purchase and maintenance costs Renishaw scanning systems are more cost-effective and robust Simple sensors with a wide range of application software Lower costs than scanning systems Robust sensors Easy programming Simple to maintain Cost-effective replacement for lower lifetime costs Scanning pros: Ideal for profile and form measurement Very high speed data collection High data point density ensures better repeatability and reproducibility (R&R) of measured features and stability of workpiece datums More comprehensive metrology analysis - form as well as size and position Scanning cons: More complex that touch-trigger sensors Slower discrete point measurement than touch-trigger probes Dynamic effects due to accelerations during measurement must be compensated if high speed scans are to produce accurate measurement results Wear on the stylus Touch trigger pros: Ideal for inspection of 3D prismatic parts and known surfaces Highly versatile - suitable for a diverse range of applications Wide range of probes and accessories available Lower cost of purchase and lower running costs than scanning systems Simple to use Ideal for size / process control type applications Robust Smaller than scanning probes, aiding access to deep features Machine dynamics have less effect on measurement accuracy Touch trigger cons: Limited data collection rate
Flexibility Scanning Touch-trigger Renishaw scanning probes are supported by a range of articulating heads, probe and stylus changers Head and quill-mounted sensor options Conventional scanning probes cannot be articulated and suffer restricted part access Renishaw touch-trigger probes are supported by a wide range of heads and accessories long extension bars for easy part access wide range of touch-trigger sensors Scanning pros: Ideal for profile and form measurement Very high speed data collection High data point density ensures better repeatability and reproducibility (R&R) of measured features and stability of workpiece datums More comprehensive metrology analysis - form as well as size and position Scanning cons: More complex that touch-trigger sensors Slower discrete point measurement than touch-trigger probes Dynamic effects due to accelerations during measurement must be compensated if high speed scans are to produce accurate measurement results Wear on the stylus Touch trigger pros: Ideal for inspection of 3D prismatic parts and known surfaces Highly versatile - suitable for a diverse range of applications Wide range of probes and accessories available Lower cost of purchase and lower running costs than scanning systems Simple to use Ideal for size / process control type applications Robust Smaller than scanning probes, aiding access to deep features Machine dynamics have less effect on measurement accuracy Touch trigger cons: Limited data collection rate
The ideal scanning system Characteristics of the ideal scanning system: High speed, accurate scanning of the form of known and unknown parts Rapid discrete point measurement when measuring feature position Flexible access to the component to allow rapid measurement of all critical features on the part Easy interchange with other types of sensor, including touch-trigger probes and non-contact sensors. Allows the sensor choice for each measurement to be optimised Minimum stylus wear The typical manufacturer will require a combination of high speed, accurate scanning for critical features where form measurement is essential, plus discrete point measurement for features that require just size or position control. Scanning systems can do both of these things, although they are not the optimum solution for fast measurement of discrete points. The ideal scanning system therefore has the following characteristics. For more information on how Renishaw scanning systems are designed to meet these criteria, explore the Renishaw scanning topic in the Quick links below. Ideal scanning system characteristics High speed, accurate scanning of the form of known and unknown parts Rapid point measurement to capture discrete data points when measuring feature position Flexible access to the component to allow rapid measurement of all critical features on the part Easy interchange with other types of sensor, including touch-trigger probes and non-contact sensors. This allows users to optimise their sensor choice for each measurement application.
Which inspection solution will suit your application? Probing applications Touch-trigger or scanning? Dynamic effects on scanning performance Articulation or fixed sensors? This section highlights how dynamic errors induced in the machine structure impinge on measurement accuracy as scanning speeds rise. Stylus changing or sensor changing? Active or passive scanning?
Dynamic effects on scanning performance The scanning paradox… Modern CMMs can move quickly, yet conventional scanning is typically performed at low speeds less than 15 mm sec (0.6 in/sec) Why? Modern CMMs are able to move very quickly - often faster than 500 mm/sec - and can generate high accelerations - sometimes more than 0.5g. Yet scanning on conventional scanning systems is typically performed at a tiny fraction (less than 5%) of this potential. These slow scanning moves offset most of the gains in cycle time made by having a fast measuring machine. Clearly there is scope for significant cycle time improvements if some of this untapped potential can be released.
Dynamic effects on scanning performance Scanning induces dynamic forces in the structure of the CMM and the probe itself, which can affect measurement accuracy Dynamic errors are related to acceleration of the machine and probe as the stylus is moved over the surface of the component Scanning is different to touch trigger probing in that the machine is under inertial load throughout and this leads to deflections in the structure that are very difficult to predict. As scanning speeds rise, this effect becomes increasingly significant. At high scanning speeds, the dynamic performance of the machine becomes much more important than its static performance. Conventional scanning systems minimise this by moving slowly. However, this compromises inspection productivity.
How do machine dynamic errors arise? Discrete point measurement is done at constant velocity - acceleration is zero at the point of contact with critical damping Consequently there are no inertial forces on the machine or probe Touch-trigger probing does not suffer from dynamic errors to the same degree as scanning, since most inspection moves are performed at a constant velocity, so that the machine is not accelerating at the point of contact. Critical damping is vital to ensure that oscillations are not set up as the machine moves onto the surface. The diagram shows how the machine is not accelerating at the point of contact with the surface in a typical discrete point inspection move.
How do machine dynamic errors arise? Scanning requires continuously changing velocity vectors as the stylus moves across a curved surface Varying inertial forces are induced, which cause the machine to deflect Vibration is also a factor when scanning During a scanning measurement cycle, the sensor must be moved over the surface of the component, generally describing a path that is not straight. For instance, as a probe is moved in a circular path (when measuring a bore, for example) the CMM on which it is mounted is undergoing constantly changing velocity. This diagram shows how accelerations arise during scanning. As the probe picks up speed on its circular path, the magnitude of the acceleration increases until it reaches a magnitude of Vel2/R (where R is the radius of the bore). These accelerations induce changing inertial forces in the CMM's structure, causing it to deflect. Although not visible to the eye, these deflections can become significant when measuring most features. The size of these dynamic errors induced by acceleration of the machine depend upon the mechanical design of the machine structure and the stiffness of the bearings. It is this dynamic performance that limits measurement speeds in conventional scanning systems. Whilst machine and servo performance can be improved through mechanical and control system design, there is a limit to how far this can be taken. At some point, a barrier is reached beyond which it is not possible to scan with sufficient accuracy. This is frustrating since modern CMMs can move at high speeds. The speed at which a CMM can scan accurately using conventional techniques is a fraction of its potential performance.
What about scanning sensor dynamics? During scanning, the deflection of the probe varies due to the difference between the programmed path and the actual surface contour The probe must accommodate rapid changes in deflection, without loss of accuracy or leaving the surface The ideal scanning sensor can accommodate rapidly changing profile due to: a high natural frequency low suspended mass low overall weight At high scanning speeds, the dynamic performance of the CMM is the dominant source of potential measurement error. However, the scanning sensor also contributes to the dynamic performance, although generally this forms a small proportion of the total error. As speeds rise, the dynamic performance of the sensor becomes increasingly important. To be able to track rapidly changing surface contours, a scanning system must have tight servo control as well as a sensor that can handle high accelerations without losing contact with the surface. The ideal scanning sensor therefore presents a low suspended mass, with a lightweight and stiff mechanism that exhibits a high natural frequency. Whilst important, probe dynamics have a very small effect compared to machine dynamics
Dynamic errors in practice Example: measure a Ø50 mm (2 in) ring gauge at 10 mm/sec (0.4 in/sec) using a CMM with performance of 2.5 + L/250 Static errors dominate at low speed Form error 2m This chart shows the results of a scan around a ring gauge on a typical CMM. A range of readings are detected, some of which represent form errors in the gauge, whilst others are vibration errors caused by a combination of surface roughness and noise in the feedback system. At the speed selected, dynamic errors are small since accelerations are low. At these speeds, it is mostly the CMMs static performance characteristics (as described by its U3 specification) which contributes to the size of any measurement error.
Dynamic errors in practice Example: re-measure ring gauge at 100 mm/sec (4 in/sec) on the same CMM Dynamic errors dominate at high speed Form error 8m At higher speeds, dynamic errors increase. Within the error chart is a low frequency sinusoid caused by deflection of the machine axes under acceleration as the probe moves around the ring gauge. There are also higher frequency vibrations evident, mostly deriving from the servo performance.
The dynamic performance barrier Dynamic errors increase as speeds rise At higher scanning speeds, machine dynamics becomes the dominant source of measurement error Error It is this variability of accuracy with speed that normally acts as a barrier to measurement cycle times. Errors on bore measurements will increase with roughly the square of the speed. The inertial forces on the CMM at 100 mm/sec are 100 times greater than they are at 10 mm/sec. As a rule of thumb, when taking measurements on a CMM, the measurement error should be no more than 10% of the allowable tolerance. Speed
The dynamic performance barrier Scanning speeds have to be kept low if tight tolerance features are to be inspected Left uncorrected, machine dynamics present a dynamic performance barrier to accurate high speed scanning Error Modern CMMs are capable of moving at high speeds - several hundred mm/sec - but typically scanning needs to be done at much slower speeds if accuracy is required. To get acceptable accuracy on tight tolerance parts, conventional scanning systems take measurements at low speeds - typically less than 20 mm/sec (0.8 in/sec). Clearly there is potential to scan much faster than this, if only the dynamic errors induced by the deflection of the machine's structure could be overcome. Emax S1 Speed
The dynamic performance barrier We need a way to break through the dynamic performance barrier, making high speed scanning more accurate Error There are various possible ways to tackle the dynamic performance barrier: Improve the mechanical stiffness of the CMM so that it deflects less under inertial loads. Whilst this is possible, there are diminishing returns here since extra stiffness generally requires extra mass and hence greater dynamic forces. Tighten the motion control of the CMM to provide a 'stiff' servo loop. Close integration of the sensor data and machine position is needed to keep the stylus on the surface of the part at high scanning speeds. Ensure that the scanning sensor is designed to handle high speed scanning. This requires a fast dynamic response from the suspended stylus mechanism to accommodate surface changes. A low overall mass is also desirable since it reduces the dynamic loads on the CMM quill. Rather than fighting the laws of physics by trying to eliminate dynamic errors, it is better to find a way to quantify and then eliminate them from the measurement results. More details of this technology can be found in the Renishaw CMM motion control presentation. Emax S1 S2 Speed
Which inspection solution will suit your application? Probing applications Touch-trigger or scanning? Dynamic effects on scanning performance Articulation or fixed sensors? Articulating heads add flexibility to any probing system, and also reduce cycle times. They provide particular benefit to users with complex parts to measure, with features that are not all accessible from one direction. This is a crucial choice, since the presence or lack of an articulating head has a large impact on the sensor and machine specification. In almost all circumstances, an articulating head will lead to a more flexible and more productive measuring system. Stylus changing or sensor changing? Active or passive scanning?
Articulation or fixed sensors? Articulating heads are a standard feature on the majority of computer-controlled CMMs Heads are the most cost-effective way to measure complex parts Fixed probes are best suited to applications where simple parts are to be measured Ideal for flat parts where a single stylus can access all features Articulating heads allow users to reorientate the probe without the need to re-qualify the stylus. This alone increases flexibility and saves time that fixed probe spend changing styli. Fixed probes are only the optimum choice when the range of parts to be measured is small and the parts themselves are simple.
Articulating heads - benefits Flexibility - a single, simple stylus can access features in many orientations Indexing and continuously variable solutions Flexibility Articulation brings the key benefit of simpler stylus arrangements that can be used to probe features in almost any orientation. In many cases, one stylus is all you need to inspect a complex component. By contrast, fixed probe systems typically require several styli to inspect a part, particularly where features can not be accessed by a straight stylus mounted vertically. This need for multiple styli makes fixed sensors slower to use than probes mounted on articulating heads - since indexing is faster than stylus changing. For the vast majority of measurement applications, an indexing solution is ideal. Indexing heads can position the probe in a finite number of equi-spaced positions around each of two rotary axes. The majority of machined features are oriented relative to the part coordinate system in multiples of 15 degrees. Renishaw indexing heads provide 7.5 degree increments in each axis to match this need. The need for subdivisions below 7.5 degrees is rare and tends to be combined with a requirement for extended reach on large contoured parts such as auto bodies. In these applications, the best solution is a servo head with a high torque so that it can carry long extensions. Renishaw articulating heads can be used with both scanning and touch-trigger probes, and can support extension bars for extra flexibility.
Articulating heads - benefits Repeatable indexing using kinematic principles: Method: 50 measurements of calibration sphere at {A45,B45}, then 50 with an index of the PH10M head to {A0,B0} between each reading TP200 trigger probe with 10mm stylus Results: Comment: Indexing head repeatability has a similar effect on measurement accuracy to stylus changing repeatability Result Span fixed Span index [Span] [Repeatability] X 0.00063 0.00119 0.00056 ± 0.00034 Y 0.00039 0.00161 0.00122 ± 0.00036 Z 0.00045 0.00081 0.00036 ± 0.00014 Test spec: Machine Specification: U1 = 0.4 + L / 1000 U3 = 0.9 + L / 1000 Controller: Renishaw UCC-1 Controller. Head: PH10MQ. Extension Bar: None. Probe: TP200. Stylus: 10x4mm Stainless Steel Software: Virtual DMIS V3.8.5 Test method: Machine underwent a 90 minute warm-up cycle. 50 measurements were made on a calibration sphere (13 points per measurement) with the head fixed at angle A45, B45. This establishes a baseline repeatability for the machine. Another set of measurements were made, with an index to A0, B0 and back again between each. The same qualification data was used throughout. Spans and standard deviations for the sphere centre position were established for both set of readings. Increases in span and standard deviation can be attributed to head index repeatability. Repeatability is defined as a 2 sigma value (i.e. 95% of readings will fall within this amount of the mean). Increases in repeatability were less than 0.4 microns in each direction (increase in 3D repeatability was approx 0.5 microns) - typically not a significant part of allowable measurement error. Most relative measurements on most parts are taken with the head in a fixed position, hence head repeatability is not a factor in the results.
Articulating heads - benefits Speed - indexing is faster than stylus changing (done during CMM moves) Dynamic response - simple, light styli make for a lower suspended mass Costs simple styli with low replacement costs small, low cost stylus change racks Speed Using a single probe and simple stylus combination to measure features in several orientations saves time. Re-orientating the sensor takes bout 2 to 3 seconds - and can be performed during positioning moves in proven inspection programs - thus having no impact on cycle times. By contrast, using a quill mounted scanning probe will either require a clumsy stylus cluster or will demand multiple stylus arrangements. To change styli, the quill must be moved to a docking position at the stylus rack, a change operation must occur, and the probe must then be positioned back to the part. If a complex stylus is used, time is wasted on longer clearance moves around the part. With Renishaw's kinematic stylus changing system, this time is minimised. Conventional stylus changing systems are slower - with these the measurement downtime can be as long as 30 seconds.
PH10M indexing head - design characteristics Flexible part access This video shows scanning features in three different orientations, using a single, simple stylus. The probe is fitted to a PH10M indexing head, which allows rapid reorientation of the probe to allow access to the three features. In this instance, the head reorientations are programmed to occur during positioning moves, meaning that they have no impact on the measurement cycle time. By contrast, a fixed sensor would need either a large, complex stylus arrangement or a time-consuming stylus change to measure these features. Rapid indexing during CMM positioning moves give flexible access with no impact on cycle times
Articulating heads - benefits Automation - programmable probe changing with no manual intervention required touch-trigger, scanning and optical probing on the same machine Stylus changing - even greater flexibility and automation optimise stylus choice for each measurement task Sensor changing Having access to a range of different types of probe can be invaluable if you have many different types of parts to inspect. For some parts, a touch-trigger probe will be ideal, whilst others will need a scanning sensor (to inspect complex forms, for instance), and yet others may need an optical sensor (where the material to be inspected is soft). In some cases you may need an extension bar to offset the probe from the quill to allow access to a deep feature. The ACR1 probe changing rack allows you to store up to 8 different sensors and automatically change between them. This increases your flexibility and allows you to pick the best extension / probe / stylus combination for each key feature on your components. Stylus changing Stylus selection is important. In general, you should aim to use the shortest and stiffest stylus available, with the minimum of joints. Long and complex styli inevitably bend under probing forces. Whilst most of this can be corrected with simple calibration, there is another consequence of selecting large styli that should be avoided where possible: the impact on dynamic response when scanning. Larger and hence heavier styli reduce the dynamic response of a scanning probe, thus acting to limit inspection speeds. It may be better to incur a quick stylus change delay than to compromise scanning performance with a sub-optimal stylus. A stylus module changing rack means that you can change styli automatically in the middle of an inspection cycle. Stylus changing racks are generally passive, simple devices that allow styli to be swapped quickly, thus minimising the impact on inspection cycle time.
Articulating heads - disadvantages Space - a head reduces available Z travel by a small amount - can be an issue on very small CMMs PH10MQ in-quill version of PH10 indexing head reduces Z travel requirements In order to articulate a probe, a motorised indexing or servo head needs to pivot about a point somewhere below the end of the CMM quill. This means that some of the available Z travel of the CMM is used up by the body of the articulating head. On very small machines, this can be an issue. In these circumstances, a fixed, in-quill probe can be the best solution. However, on all but the very smallest machines, the benefits of articulation tend to out-weigh concerns about the size of the head. For the minimum impact on Z travel, an in-quill version of the PH10 - the PH10MQ - can be fitted and provides a solution with minimal intrusion into the machine's operating volume.
Fixed sensors - benefits Compact - reduced Z dimension makes minimal intrusion into the measuring volume - ideal for small CMMs Simplicity - fixed passive sensors are less complex for lower system costs Note: an active sensor is more complex and more expensive than a passive sensor and an articulating head combined Fixed sensor Articulating head Z space Fixed sensors can be more compact than systems that include a motorised head. The images shown above show a compact, passive scanning sensor mounted directly to a CMM quill. However, active sensors and some designs of quill mounted probes are very large, and use excess Z space of the machine. A further point to consider is the size and complexity of stylus clusters on fixed probes, which can act to reduce available working area even when the probe itself is compact. Simplicity A scanning system using a fixed, passive sensor is simpler than one that includes an articulating head - two axes of motion are eliminated. Clearly, users of such systems do not have to pay for a head - reducing purchase costs. Active sensors have 3 axes of motion in them, so are actually more complex than an articulating head and a passive sensor. They also tend to be more expensive. Stylus length Some fixed sensors can carry very long styli, suitable for access into deep parts. The smaller scanning probes mounted to articulating heads tend to be able to carry shorter styli. However, this capability must be compared with the reach that is achievable with a probe fitted to an extension bar on an articulating head. Stylus length - fixed sensors can be larger than those fitted on articulating heads, making it possible to carry longer styli
Fixed sensors - disadvantages Feature access - large and complex stylus arrangements are needed to access some features Large 'crank', 'T' and 'star' styli are often used with conventional scanning sensors. They are needed to probe features that could not otherwise be reached with a simple, straight stylus. Long, complex styli present programming difficulties since extra care must be taken to avoid collisions with the part, fixtures, machine structure and stylus changing system.
Fixed sensors - disadvantages Programming complexity - complex stylus clusters mean more attention must be paid to collision avoidance DANGER! Possible collisions with: component fixturing stylus change rack other styli in rack machine structure Programming Complex styli can mean more complex inspection cycles. As well as having to consider multiple stylus tips, programmers must concern themselves with avoiding collisions between complex styli and the component, stylus change rack, the fixture and the machine structure. By comparison, probes with simple styli mounted on an articulating head require less space in which to move, present less risks of collision, and hence require simpler programming.
Fixed sensors - disadvantages Machine size large stylus clusters consume measuring volume larger stylus change racks you may need a larger machine to measure your parts Large 'crank', 'T' and 'star' styli are often used with conventional scanning sensors. They are needed to probe features that could not otherwise be reached with a simple, straight stylus. However, to move such styli around the part during measurement without colliding with either the part or the machine structure is difficult. Much of the operating volume of the measuring machine becomes unusable for part measurement. In many cases, this will require users to specify a larger machine than they really need. A further consequence of fixed probes is that the typical stylus arrangement is larger than on a scanning system fitted with an articulating head. This means that stylus changers are larger, further reducing the volume available for measurement. Star styli consume greater working volume
Fixed sensors - disadvantages Speed - stylus changing takes longer than indexing up to 10 times slower than indexing indexing can be done during positioning moves Dynamic response - heavy styli increase suspended mass and limit scanning speed Accuracy - complex styli compromise metrology performance Speed Fixed scanning sensors typically require the stylus to be changed in order to inspect all of the features on a complex part. Many such stylus changes can be eliminated with an articulating head, which provides flexible access to the component. A stylus change involves moving the CMM quill to the rack where the styli are stored, followed by a manoeuvre to swap styli. The probe must then be moved back to a position to measure. On some CMMs this can be a lengthy procedure, taking up to 30 seconds. With Renishaw stylus changing systems this can be much quicker - typically around 10 seconds - but the fact remains that cycle times are increased. By contrast, an articulating head can index in just two or three seconds, and this action can be combined with movement of the machine around the part on proven programs, so that there is no impact on the cycle time. Accuracy Large styli are heavy, which impacts on the dynamic performance of a scanning probe, limiting scanning speeds. The longer the stylus, the more bending there will be under the contact forces between the stylus and the component. Whilst such effects can be addressed by calibration, it is best to maximise stylus stiffness to avoid unnecessary deflections. Styli with joints suffer from poor repeatability and compromised metrology performance compared to shorter, simpler styli.
Which inspection solution will suit your application? Probing applications Touch-trigger or scanning? Dynamic effects on scanning performance Articulation or fixed sensors? Stylus changing or sensor changing? Active or passive scanning?
Why change styli? Optimise your measurement repeatability for each feature by selecting a stylus with: Minimum length Longer styli degrade repeatability Maximum stiffness Minimum joints Maximum ball size Maximise the effective working length (EWL) Test results: TP200 repeatability with stylus length Stylus length 10mm 50mm Uni-directional repeatability 0.30 µm 0.40 µm 2D form deviation ±0.40 µm ±0.80 µm 3D form deviation ±0.65 µm ±1.00 µm Guidelines for stylus optimisation: Keep styli short and stiff The more the stylus bends or deflects, the lower the accuracy. Probing with the minimum stylus length for your application is recommended and where possible the use of one piece styli is suggested. Probing with excessive styli/extension combinations should therefore be avoided. Keep the stylus ball as large as possible This will ensure maximum ball/stem clearance whilst providing a greater yet rigid Effective Working Length (EWL). Using larger ruby balls also reduces the effect of surface finish of the component being inspected. Effective working length (EWL) is the penetration that can be achieved by any ruby ball stylus before its stem fouls against the feature.Generally the larger the ball diameter the greater the EWL. The test data shows the repeatability of a TP200 touch-trigger probe with two different length styli. It is clear that extra stylus length increases variation in measurement results. This principle applies to scanning probes as well. Long styli that feature joints will show much greater repeatability ranges than shorter styli. The lesson from this is only to use long or complex styli when absolutely necessary to measure difficult features. It is generally better to use a shorter stylus for the majority of measurements and switch to a longer stylus only for those features that demand it.
Stylus changing Many probe systems now feature a repeatable stylus module changer access to features that demand long or complex styli different tips (sphere, disc, cylinder) needed for special features Automated stylus changing allows a whole part to be measured with a single CMM programme reduced operator intervention increased throughput Stylus changing Unless you are measuring a simple component, you will need to change your stylus configuration to suit different measurement tasks. This can be done manually using a threaded connection, but probe systems are now available with a repeatable, automated means to switch styli. This greatly increases your flexibility by allowing you to access features that demand long or complex styli, as well as using different tips (sphere, disc, cylinder) needed for different surface configurations. With automated stylus changing, all of this can be achieved automatically, reducing operator intervention and increasing measurement throughput.
SP600 stylus changing Stylus changing The video shows how quick and simple an automatic stylus change can be. In just 10 seconds, the stylus is switched and the probe is ready to measure again. The repeatable kinematics mean that the stylus can be changed many times and will always be in the same position relative to the probe, so there is no need to re-qualify after each change. Rapid stylus changing with the passive SCR600 stylus change rack
Why change sensors? Not all parts can be measured with one sensor: Scanning probe ideal for features with functional fits where form is important digitising contoured surfaces Not all parts can be measured with one sensor: Touch-trigger probe ideal for discrete point inspection, for size and position control compact for easy access to deep features Sensor changing A scanning sensor is the most flexible sensor you can fit to your machine, since it can perform discrete point measurement as well as scanning. However, it is not the optimum sensor for discrete point measurement - touch-trigger probes can be faster - nor for more specialised tasks like measurement of soft materials (where a non-contact sensor will be best). Optical probes ideal for pliable surfaces inspection of printed circuit boards
Probe sensor changing The requirement... The solution… Automatic, no requalification, easy programming automatic switching automatic sensor recognition automatic electrical connections automatic alignment of sensor The requirement... If the range of features and parts that you must measure demands a range of sensors, then a sensor changing system is essential Sensor changing One sensor may not be suitable for all your measurement needs. If you can accept the limitations of a single type of sensor, then manual or automated stylus changing will be sufficient. However, if the range of features and parts that you must measure demands a range of sensors, then a sensor changing system is essential. Sensor changing requires a repeatable, automated joint between the probe head or CMM quill and the probes that you want to use. You will also need a means to store those sensors that are not in use on the machine. Renishaw articulating heads can be fitted with a unique and highly repeatable Autojoint, that allows automatic switching between different probe configurations. The newly attached sensor will be in the same position as when it was calibrated, meaning that you can start to measure without re-qualification.
New ACR3 probe changer for use with PH10M Probe changing Video commentary New ACR3 sensor changer No motors or separate control Change is controlled by motion of the CMM The ACR3 is a simple, passive device that uses the machine motion to affect the unlocking and locking actions needed to change sensors. The lateral movement of the rack causes the Autojoint to lock or unlock, disconnecting the old sensor and connecting the new one. In this instance, a scanning sensor is swapped for a touch-trigger probe, highlighting how different types of sensor can be integrated into a single probing routine. Quick and repeatable sensor changing for maximum flexibility
Which inspection solution will suit your application? Probing applications Touch-trigger or scanning? Dynamic effects on scanning performance Articulation or fixed sensors? There are two main types of scanning probe technology - 'active' and passive. What are the differences and is the choice of technology important for your scanning performance? Renishaw scanning systems are based around 'passive' sensors, whilst other metrology suppliers provide 'active' sensors. Stylus changing or sensor changing? Active or passive scanning?
Passive sensors Simple, compact mechanism no motor drives no locking mechanism no tare system no electromagnets no electronic damping springs generate contact force force varies with deflection Passive sensor technology Passive scanning probes use a simple mechanism to sense stylus deflection in 3 axes. During measurement the springs in the suspended probe mechanism generate a force to match the deflection. This contact force varies with deflection, but it is highly repeatable. Typical scanning deflection Force Deflection
Active sensors Complex, larger mechanism motors generate contact force force modulated by motors deflection varies as necessary longer axis travels Displacement sensor Axis drive motor Active scanning technology Active scanning sensors use a motorised mechanism that is able to control the stylus deflection and modulate the contact force with the component. Motors generate the contact force, rather than springs. This arrangement attempts to avoid the non-linear force / deflection characteristics of the sensor by limiting their effect through mechanical design. This approach uses a complex electro-mechanical design to try to avoid accuracy problems, with relatively simple compensation. The result is a large, expensive and more fragile sensor. Force Controlled force range Deflection
Scanning a ‘defined’ surface Most scanning is performed on ‘known’ or ‘defined’ features feature size, position and form vary only within manufacturing and fixturing tolerances Renishaw passive scanning: CMM moves around feature adaptive scanning keeps deflection variation to a minimum small form errors accommodated by sensor mechanism small force variation due to deflection range Active scanning: CMM moves around a pre- defined path form errors accommodated in the sensor force variation is controlled by probe motors The majority of scanning on CMMs involves measuring defined features - precision geometric forms whose size, position and shape is known to lie within a small tolerance band. These features are scanned to check that they fall within the required tolerances. Active scanning systems use their motors to control the contact force between the stylus and the component during the scan. The force is controlled within a narrow range, but there are small variations. The scan involves servoing the stylus carrier over the small range of deflections necessary to accommodate the errors in the form of the feature. Usually these are small - measured in tens of microns. Passive scanning systems use the machine’s drives to accommodate surface variations. When performing a scan using a pre-defined scan path, the probe must accommodate form errors. These errors will cause the probe deflection to vary, resulting in slight variations in contact force. If Renishaw adaptive scanning algorithms are used, the scan path is continuously adjusted based on the probe deflection, resulting in a still smaller force variation. The result is a variation in contact force that differs little from that seen with active sensors. Furthermore, sensor calibration ensures that any variation in contact force does not affect measurement results. Active force control does not significantly reduce force variation in most scanning applications
Scanning sensor design factors Passive sensors contact force is controlled by CMM motor drive compact sensor that can be mounted on an articulating head short, light, simple styli low spring rates Active sensors contact force is controlled by probe motor drive large, fixed sensor long, heavy styli motors required to suspend the stylus to avoid high contact forces Passive scanning Contact forces between the stylus and the component cause the stylus to bend. These deflections are not visible to the eye, but must be corrected to avoid measurement error. Passive sensors are smaller and lighter than active sensors since they do not contain a motorised mechanism. When mounted to an articulating head, simpler and lighter styli can be used, meaning a lower spring rate is needed. Active scanning Contact forces and stylus bending must be accounted for. However, active sensors are fixed to the CMM quill, meaning that they need larger and heavier styli to access all the features on a part. This necessitates either higher spring rates and hence contact forces, or larger probe deflections. An active mechanism is needed to control these heavy styli without having excess contact forces. Compact passive sensor Complex active sensor
Scanning probe calibration Passive sensors probe characteristics, including stylus bending, are calibrated simple calibration cycle sophisticated non-linear compensation Active sensors smaller variation in contact force, but styli are less stiff calibration of probe mechanism characteristics and stylus bending effects at fixed force still required Passive scanning Passive sensors use a compensation algorithm derived from a calibration routine. Renishaw compensation algorithms include sophisticated 3rd order polynomials that convert the probe's non-linear output into true deflections. Active scanning By controlling the contact force, stylus bending is also controlled. This means that any errors due to these contact forces can be quantifed and compensated for. Although the contact force is controlled, the probe deflection is not directly related to it, so the way in which the position sensing system behaves with deflection of the mechanism must still be calibrated. Compact passive sensor Complex active sensor
Scanning probe calibration Constant force does not equal constant stylus deflection although active sensors provide constant contact force, stylus bending varies, depending on the contact vector stylus stiffness is very different in Z direction (compression) to in the XY plane (bending) if you are scanning in 3 dimensions (i.e. not just in the XY plane), this is important e.g. valve seats e.g. gears F F High deflection when bending Constant force does not mean constant deflections Some active sensors are promoted on the basis that they provide a constant contact force, which implies better measurement performance. Whilst it is indeed possible to maintain force in a narrow range, this does not necessarily result in better performance. 2D scanning The key is stylus bending. When the contact force is perpendicular to the axis of the stylus (probing in the XY plane with the probe aligned with the Z axis), the stylus is under a bending load. It deflects according to the formula = F.l3 / 3EI. Stiffness in this plane is relatively low. Bending of the stylus during scanning in the XY plane is indeed constant if the force is modulated successfully. This approach allows for accurate measurement with a simple assumption of constant stylus bending. 3D scanning However, as soon as the contact vector introduces a component in the Z direction, things change significantly. Stiffness in the probe’s Z direction is far higher than in the XY plane, since the stylus is in compression rather than bending. This results in far lower deflections under scanning loads in this direction. When scanning a sphere, for instance, the deflection reduces with the angle as shown in the graph. The constant force = constant stylus bending assumption breaks down when tougher measurement tasks are faced. Therefore, if your scanning application involves anything other than XY measurements, constant force (and an assumption of constant deflection) will not be helpful. An calibration approach which takes account of variation in the deflection characteristics of the stylus is the best way to measure complex parts like gears, valve seats or mould cavities with accuracy. Deflection Low deflection in compression 90 180
Scanning probe calibration constant force does not result in better accuracy how the probe is calibrated is what counts Passive sensors passive probes have contact forces that are predictable at each {x,y,z} position scanning probe axis deflections are driven by the contact vector sensor mechanism and stylus bending calibrated together Active sensors contact force is controlled, and therefore not related to {x,y,z} position no relationship between contact vector and probe deflections separate calibration of sensor mechanism and stylus bending Passive sensor calibration The mechanism of a passive probe is moved under contact forces between the stylus and the component. As it moves, springs resist the contact force until equilibrium is reached. As such, any deflection of the probe is the result of a predictable contact force for any one stylus configuration. With a passive mechanism, it is possible to calibrate the characteristics of the probe mechanism (e.g. small errors in position feedback and any non-linear effects) at the same time as compensating for stylus bending. Both sources of potential measurement error can be detected and removed in one process. Active sensor calibration By contrast, an active sensor controls the force between the stylus and the component, moving the sensor mechanism as appropriate to achieve this. The contact force is constant, whatever the position of the stylus within the probe’s frame of reference. This means that there is no fixed relationship between the contact vector and the probe deflection, resulting in a need for separate calibrations for the probe mechanism and the stylus bending characteristic. Unlike a passive scanning probe, when measuring in 3D with an active sensor, the stylus bending cannot be accurately inferred from the probe position.
Active or passive scanning - conclusion Both active and passive systems achieve the basics - accurate scanning within their calibrated operating range Their performance and costs differ Look at the specification of the system before making your choice Renishaw scanning systems are based around 'passive' sensors, whilst other metrology suppliers provide 'active' sensors. Both active and passive systems achieve the basics - accurate scanning within their calibrated operating range. Look at the specification of the system before making your choice and consider the following: The capability to scan accurately at high speeds Cycle time for discrete point measurement Overall cycle times for typical parts, including stylus change time Purchase and service costs ?
Questions to ask your metrology system supplier Do my measurement applications require a scanning solution? How many need to be scanned? How many need discrete point measurement? If I need to scan, what is the performance of the system? Scanning accuracy at high speeds Total measurement cycle time, including stylus changes If I also need to measure discrete points, how fast can I do this?
Questions to ask your metrology system supplier Will I benefit from the flexibility of an articulating head Access to the component Sensor and stylus changing What are the lifetime costs? Purchase price What are the likely failure modes and what protection is provided? Repair / replacement costs and speed of service
Questions? apply innovation