Bond test apparatus and method

Abstract

A bond test apparatus 200 comprises: a flexure 272 having free and fixed ends; a test tool 210 coupled to the free end and configured to contact a bond during a test, so that application of a test for

Description

Field of the Invention The present invention relates to a bond test apparatus for testing the strength of bonds on electrical circuitry, such as a PCB or semiconductor device. In particular, the invention relates to an apparatus for measuring the force on a test tool during a bond test. Background to the Invention Semiconductor devices are very small, typically from 5mm x 5mm square to 50mm x 50mm square, and typically comprise numerous sites for the bonding of electrical conductors to a semiconductor substrate. Each bond consists of a solder or gold ball deposit, known as a “bump”, a copper pillar, or a wire adhered to the substrate. It is necessary to test the bond strength of the bonds, in order to be confident that a particular bonding method is adequate. Because of the very small size of the bonds, tools used to test the bond strength of these bonds must be able to measure very small forces and deflections accurately. There are several different types of bond tests that are used to test bond strength. For example, shear testing tests the shear strength of a bond by applying a shear force to the side of the bond and shearing the bond off the substrate. Pull testing tests the pull strength of the bond by pulling a ball deposit, or a wire embedded in a ball deposit, away from the substrate. In a push test, a force, or load, is applied in the vertical plane directly downward onto a bond. Machines that perform these tests typically comprise a bond test tool, be it a shear test tool, push test tool or a pull test tool, that can be positioned relative to the bond under test and then either the bond or the tool is moved in order to perform the test by measuring the force needed to break the bond. In many cases, the force applied between the test tool and the bond during a bond test is measured directly using strain gauges. For example, a pull tool may be mounted on one or more horizontal beams, or flexures, such that the application of a vertical pull force between the test tool and the bond causes the flexure to bend. By mounting a strain gauge on the flexure, the force on the test tool can be measured. However mounting strain gauges onto flexures and connecting them to form strain transducers is a difficult and expensive process, with the accuracy of the measured test force highly reliant on the precision with which the strain gauge has been mounted to the flexure. As strain gauges attached to flexures must inherently be deformed, or strained, on bending of the flexure, they may fail to recover completely following severe bending, such that strain gauges typically exhibit creep over time, making their outputs less accurate. Further, if a strain gauge is to measure the force on a flexure it must be attached to the flexure itself, for example by an adhesive bond. Such an adhesive bond may be deformed during deformation of the strain gauge, and may introduce a significant source of measurement creep. Sensors which provide a measurement of displacement rather than a direct measurement of force, and which preferably are not deformed by application of a force to the flexure, may advantageously exhibit very little creep. These sensors may therefore maintain better accuracy than the piezoelectric or piezo resistive strain gauges conventionally used to measure force in bond testing. EP3444588A1, for example, explored the use of displacement sensors such as magnetic field sensors and slotted photo-interrupters instead of strain gauges to measure bond test forces by measuring flexure deflection instead of strain. The present invention seeks to overcome these deficiencies in the prior art, and to provide an improved bond test apparatus and an improved method of measuring a force in a bond test apparatus. Summary of the Invention The present invention is defined in the appended independent claims to which reference should now be made. Preferred or advantageous features of the invention are defined in the dependent claims. Bond Test Apparatus In a first aspect there is provided a bond test apparatus comprising a flexure having a fixed end and a free end, and a test tool which is coupled to the free end of the flexure. The test tool is configured to contact a bond during a bond test, so that application of a test force between the test tool and the bond during a bond test displaces the free end of the flexure relative to the fixed end of the flexure. The bond test apparatus additionally comprises a linear encoder, and a processor which is configured to receive an output signal from the linear encoder. The linear encoder is configured to measure a displacement of the free end of the flexure relative to the fixed end of the flexure, and to deliver a digital output signal to the processor. The processor is programmed to convert the digital output signal from the linear encoder into a measurement of the magnitude of the test force applied between the test tool and the bond being tested. Traditionally the push / pull transducers (force sensors) used in bond testing applications use foil strain gauges configured in a wheatstone bridge to measure the surface strain on a flexure. Foil strain gauges are used on these transducers due to their strain rate, which makes them suitable for use in push- and pull- bond tests where the deflection of the flexure is high compared to the loading force applied to the flexure. As force is applied to the sample using a suitable tool attached to a flexure, then the flexure will deflect in a linear manner. The wheatstone bridge formed from strain gauges will produce an analogue output voltage proportional to the force applied to the flexure. As the analogue voltage produced by such transducers is low in magnitude, the output voltage is then amplified and measured to provide a force reading which can be recorded by software in the bond test apparatus. In the present invention, a linear encoder is configured to measure the deflection of the flexure, and to deliver a digital output signal to the processor. This provides a distinct advantage over the strain gauge transducers of the prior art, and also over the displacement sensors of EP3444588A1. In these prior art systems, the output signals produced by the sensors are analogue voltages which typically require amplification and conversion before they can be recognised and processed by the control electronics of the bond test apparatus. The Hall effect sensor of EP3444588A1 provided a distinct advantage over strain gauge transducers by providing a higher-magnitude output signal which removes the need for an amplification step, but the linear encoder used in the present invention has the additional advantage of providing a digital output signal that can be delivered directly to the processor without first requiring an analog-to-digital conversion step. By providing a digital output signal directly from the linear encoder, the risk of noise or electrical interference is greatly reduced compared to the sensor options typically used in bond testing. The arrangement of the present invention is thus both simpler and more economical than any of the force sensing options available for bond test apparatuses in the prior art. A linear encoder is a sensor which comprises two parts: an encoder readhead and a detectable member which is detectable by the encoder readhead. The readhead acts as the sensor, and the detectable member acts as a scale which is readable by the readhead. The linear encoder readhead detects linear movement of the detectable member and encodes the position of the detectable member relative to the readhead. The linear encoder converts the encoded position of the detectable member into a digital output signal which can be decoded into a position measurement by the processor. One of the detectable member or the encoder readhead may be coupled to (or fixed relative to) the free end of the flexure, while the other of the detectable member or the encoder readhead is fixed relative to the fixed end of the flexure. When a test force is applied between the bond and the test tool, and the flexure is deflected as a result, the resulting deflection will thus cause the detectable member to be displaced relative to the encoder readhead. In a preferred embodiment, either the encoder readhead or the detectable member is removably couplable to the free end of the flexure, while the other of the encoder readhead or the detectable member is removably couplable to an encoder mount which is fixed relative to the fixed end of the flexure. By removably mounting both portions of the linear encoder to the apparatus, it is advantageously straightforward to replace one or both of the readhead or the detectable member without disassembling the entire apparatus. This makes service and repair much more straightforward than it is with strain gauges which are adhered directly onto flexures. The linear encoder may be an absolute linear encoder or an incremental linear encoder. The linear encoder is preferably a non-contact encoder, in which the readhead does not contact the detectable member in operation, which ensures that friction between the readhead and detectable member does not alter the measurement accuracy. Magnetic Linear Encoder In a first preferred embodiment, the linear encoder is a magnetic linear encoder, and the detectable member is a magnetic detectable member. In this embodiment, the magnetic detectable member is a magnetic encoder code strip comprising a plurality of alternating magnetic poles. The magnetic linear encoder readhead is configured to sense changes in magnetic field as the magnetic encoder code strip moves linearly past the readhead. Suitable magnetic linear encoders are commercially available for use in other fields, for example the LM10 linear magnetic encoder sold by Renishaw (TM), or the RLB Miniature Incremental Magnetic Encoder sold by RLS (TM). Magnetic linear encoders based on the anisotropic magneto resistive (AMR) effect may also be used in preferred embodiments, for example the Bogen Magnetics IKP11. Optical Linear Encoder In a second preferred embodiment, the linear encoder is an optical linear encoder, and the detectable member is a distance scale marked at regular intervals. Suitable optical linear encoders are commercially available for use in other fields, for example ATOM, QUANTiC, TONiC and RESOLUTE optical encoder readheads available from Renishaw (TM), and corresponding optical scales such as RTLF-S, RTLC-S and RTLA-S available from Renishaw (TM). A linear optical encoder may comprise an interpolator chip, an optical encoder chip, and an optical scale. The interpolator chip acts as a digitiser which digitises sin / cosine signals which it receives from the optical encoder chip as it moves along the optical scale. By selecting an interpolator chip with a suitably high bit resolution, and an optical scale with a suitably high resolution, small flexure deflections may be measured to a high degree of accuracy. As the deflections experienced in bond shear testing are very small (typically in the range of 0.090-0.100 mm), it is challenging to measure this displacement accurately. The present inventors have found, however, that by using a linear optical encoder with an interpolator chip having a resolution of at least 18-bit, or 24-bit, or 26-bit, the measurement resolution of the encoder is high enough to measure shear test forces to a high degree of accuracy. Bond Test Transducers The linear encoder may be used to determine test forces for a number of different bond test types, in which different types of test tool are used, and different test forces are applied to the bonds being tested. In one preferred embodiment the test tool is a pull test tool, and the bond test apparatus is configured to perform a pull test on the bond during the bond test so that the test force is a pull force. The pull force is applied in a vertical direction, typically by applying a load force to pull the test tool vertically upwards away from the bond. For example, the test tool may comprise a Cold Bump Pull (CBP) test tool comprising tweezer jaws configured to grip a bond. In another preferred embodiment the test tool is a push test tool, and the bond test apparatus is configured to perform a push test on the bond during the bond test, such that the test force is a push force. The push force is applied in a vertical direction, typically by applying a load force to push the test tool vertically downwards into the bond, or alternatively by applying a load force to push the bond vertically upwards into the push test tool. In both pull and push tests, the flexure is configured to deflect in a vertical plane in response to a test force applied in a vertical direction between the test tool and the bond. In these embodiments, the linear encoder is thus arranged to measure vertical deflection of the free end of the flexure relative to its fixed end. In another preferred embodiment, the test tool is a shear test tool, and the bond test apparatus is configured to perform a shear test on the bond during the bond test, such that the test force is a shear force. The shear force is applied in a lateral, or horizontal, direction by placing the shear test tool alongside the bond to be tested and either applying a load force to the test tool to push the test tool sideways into the bond, or alternatively by applying a load force to the stage on which the bond is mounted to push the bond sideways into the test tool. In order to measure shear force, the flexure for a shear transducer is configured to deflect in a lateral plane in response to a shear test force applied in a lateral direction between the test tool and the bond. As the magnitude of the flexure deflections experienced in shear test are much lower than those experienced in push or pull tests, linear encoders require a high measurement resolution in order to measure the deflections of the shear transducer. As the test forces are measured more simply and accurately by the linear encoder, the bond test apparatus preferably does not contain a strain gauge configured to measure a strain applied to the flexure during the bond test. This advantageously simplifies manufacture and improves reliability of bond test apparatuses, as there is no need to manually adhere strain gauges onto flexures. Preferably the digital output signal from the linear encoder is not amplified before being delivered to the processor. Output Signal Processing The bond test apparatus comprises a processor, or a controller, configured to receive the digital output signal of measured displacement from the linear encoder and, based on the digital output signal and a known stiffness of the flexure, to determine the force on the flexure. The processor may advantageously be programmed to control the apparatus based on the calculated force on the flexure, such that the sensor provides feedback to the processor, based on which the processor controls the apparatus. Prior to use, the stiffness of the flexure may be found by a calibration procedure in which the flexure is loaded with a known force, and the displacement of the first end of the flexure relative to the second end of the flexure is measured. Prior to carrying out any bond tests, the bond test apparatus is calibrated against a known load, according to calibration processes known in the art. By applying a load to the flexure using standard traceable weights, the magnitude of the flexure displacement versus these weights can be used to calibrate the apparatus. When the apparatus is working correctly then the digital output signal from the linear encoder will directly correlate flexure displacement versus load. The processor is preferably programmed to convert the digital output signal from the linear encoder into a measurement of the test force by comparing the displacement measured by the linear encoder with stored calibration data of flexure stiffness and flexure load force. Cartridge In a second aspect, there is provided a cartridge for a bond test apparatus, the cartridge being removably mountable to a bond test apparatus. The cartridge may comprise: a flexure having a fixed end and a free end; a test tool, the test tool being coupled to the free end of the flexure and configured to contact a bond during a bond test, so that application of a test force between the test tool and the bond during a bond test displaces the free end of the flexure relative to the fixed end of the flexure; a linear encoder which is configured to measure a displacement of the free end of the flexure relative to the fixed end of the flexure and to deliver a digital output signal to a processor which is programmed to convert the digital output signal into a measurement of the magnitude of the test force. The features of the bond test apparatus discussed above in relation to the first aspect may be provided in the form of a cartridge. Alternatively, the processor may be housed in the bond test apparatus, while the other force sensing components (the flexure, test tool and linear encoder) are housed in the cartridge and configured to deliver the digital output signal to the bond test apparatus through an electrical data connection. A cartridge system may be used to enable different test tools to be exchanged and used one at a time on the same bond test apparatus. For example multiple cartridges may be provided, with each type of bond test being performed by its own dedicated cartridge. Thus the invention may be realised in either form factor: within the cartridge of a multicartridge system, or integrated into a bond test apparatus which does not connect to removable cartridges. Method In a third aspect, there is provided a method of measuring a test force in a bond test apparatus, the method comprising the steps of: providing a flexure, the flexure having a fixed end moveable relative to a free end and a test tool coupled to the free end of the flexure; providing a linear encoder arranged to measure the displacement of the free end of the flexure relative to the fixed end; contacting a bond with the test tool; and applying a load force to the first end of the flexure, so that the test tool applies a test force to the bond. The method may comprise the steps of measuring, using the linear encoder, the displacement of the free end of the flexure relative to the fixed end of the flexure; delivering a digital output signal from the linear encoder to a processor; and converting the digital output signal into a measurement of the magnitude of the test force. In a fourth aspect, there is provided a method of measuring a test force using the bond test apparatus of the first aspect, or the cartridge of the second aspect. The method may comprise the steps of: contacting a bond with the test tool; applying a load force to the first end of the flexure, so that the test tool applies a test force to the bond; measuring, using the linear encoder, the displacement of the free end of the flexure relative to the fixed end of the flexure; delivering a digital output signal from the linear encoder to the processor; and converting the digital output signal into a measurement of the magnitude of the test force. The features of the bond test apparatus of the first aspect, and the cartridge of the second aspect, are applicable to the methods of the third and fourth aspects, and vice versa. The method may be a method of performing a bond test, or part of a method for performing a bond test. In the methods of the third and fourth aspects, the digital output signal from the linear encoder may be delivered to the processor without amplification. The digital output signal from the linear encoder may be converted into a measurement of the test force by comparing the displacement measured by the linear encoder with stored calibration data of flexure stiffness and flexure load force. Features described with reference to one aspect of the invention may equally be applied to other aspects of the invention. In particular, it should be clear that features described in relation to the first aspect of the invention may be applied to the second aspect of the invention, and features described in relation to the third and fourth aspects of the invention may be applied to one another. Brief Description of the Drawings Embodiments of the present invention will now be described in detail, by way of example only, with reference to the accompanying drawings, in which: Figure 1A is a simplified front view of a bond test apparatus that may be used with the present invention; Figure 1B is a cross-sectional view showing the test tool supported by the tool mounting bracket of the test tool cartridge of the bond test apparatus; Figure 1C is a perspective view showing the test tool clamped to the tool mounting bracket; Figure 1D is a perspective view of a test tool mount and drive portions of the bond test apparatus, including an anti-backlash cylinder; Figure 1E is a perspective view of the test tool cartridge of the bond test apparatus; Figure 1F is a perspective view of the bond test apparatus; Figure 2 is a perspective view of an exemplary pull test transducer usable in aspects of the present invention; Figure 3 is a perspective view of a magnetic linear encoder usable in aspects of the present invention; Figure 4A is a perspective view of a traditional prior art shear loadcell without its cover; Figure 4B is a perspective view of the shear loadcell of Figure 4A with its cover in place; Figure 4C is a side view of the shear loadcell of Figure 4A without its cover; Figure 5A is a perspective view of a shear loadcell according to an embodiment of the present disclosure; Figure 5B is a side view of the shear loadcell of Figure 5A; Figure 5C is a side view of the shear loadcell of Figures 5A and 5B, with the linear encoder readhead removed to show the encoder scale; Figure 6A is a graph of encoder count response measured for a variety of different applied loads applied to a shear cell with an optical linear encoder according to an embodiment of the present disclosure; and Figure 6B is a graph showing the linearity of encoder counts vs applied load for the shear cell measured in Figure 6A. Detailed Description Bond Test Apparatus Figure 1A is an illustration of a bond testing apparatus in accordance with the present invention. The apparatus comprises a test tool 10 mounted to a test tool cartridge 11, which is itself mounted to the main body of the bond test apparatus 12. Beneath the test tool, the bond test apparatus includes a motorized stage table 13, on which samples or substrates 100 to be tested can be mounted. The test tool 10, mounted to cartridge 11, can be a shear tool, push tool or a pull tool and can be switched in order to perform different tests. Shear tools are used, for example, for applying a force horizontally across the board to shear a bond off the substrate, and push tools are used, for example, to apply a vertical compression force on a component on the substrate. The force applied by these tools is measured. A pull tool may, for example, have a hook at the bottom of the tool that is used to hook an electrical lead, which is attached between a component and a sample circuit board, with a vertical force being applied to the tool to pull the lead off the board and measure the force required to pull the lead off the board. An example of a suitable shear tool is described in US 6,078,387, the contents of which are incorporated herein by reference. An example of a suitable pull tool is described in US 6,310,971, the contents of which are incorporated herein by reference. As shown in Figs. 1B and 1E, the test tool 10 is typically attached to the cartridge 11 by a tool mount bracket 70 having cantilever arms 72, 74 fixed at one end to the cartridge 11 by screws 73, with the free ends of the arms 72, 74 supporting a clamp 76. As shown in Figure 1C, the tool 10 is clamped in clamp 76 by means ofclamp screw 78. However, any suitable means for attaching a test tool to the cartridge mount plate may be used in a system in accordance with the present invention. Figure 1D shows the retaining channel 71 on the cartridge mount plate 21 into which the test tool cartridge 11 is slid and then secured using one or more screws 22. This design permits different test tool cartridges, having different test tools, to be used with the bond test apparatus, as appropriate for the type of test that the user is conducting. The cartridge mount plate 21 includes a data port 23 that couples with an electrical connector on the test tool cartridge 11 for transferring data from the transducers of cartridge 11 to a PC, such as data representing the force required to shear a solder ball off a substrate or pull a lead off a substrate. An interchangeable test tool cartridge design for a bond test apparatus is well known in the prior art. See for example the Dage 4000 multipurpose bond tester available from Dage Holdings Limited, 25 Faraday Road, Rabans Lane Industrial Area, Aylesbury, Buckinghamshire, United Kingdom. The cartridge 11 is moveable in a z direction normal to the surface of the substrate 100 on the stage table 13. This allows the test tool 10 to be positioned vertically relative to the substrate 100 under test so that it can contact a particular bond during a test. Relative x-y movement between the test tool 10 and the table 13 in a direction parallel to the plane of the substrate 100 is typically achieved by moving the table 13. Movement of the table 13 in x and y is achieved using suitable servo motors or stepper motors, coupled to the table 13 via a lead screw and nut, ball screw and nut, or suitable belt-drive mechanism (not shown), as is also well known in the prior art, such as the Dage 4000 Multipurpose Bond Tester referenced above Also shown in Figures 1A and 1F are control devices, comprising two joystick controls 14, 15 to allow for controlling movement of the table 13, and a keyboard 16. A display 17, a light 18 for illuminating the substrate 100 under test, and a microscope, aiding accurate positioning of the test tool 10, are also shown. These features are also all well known in the prior art, such as the Dage 4000 Multipurpose Bond Tester referenced above. Figure 1D shows that the mounting plate 21 and its connection to the main body 25. As has been described, the test tool (not shown in Figure 1D) must be moveable towards and away from a substrate under test. This is achieved by moving the cartridge mount plate 21, to which the test tool 10 is attached, relative to the main body 25 of the device in a direction towards and away from the substrate, herein referred to as the z-axis direction or axial direction. The cartridge mount plate 21 is rigidly coupled to a moving block 24, using screws 75. The moving block 24 is coupled to the main body 25 via a ball screw (or lead screw) and nut and nut block (not shown) that are driven by a servo motor or stepper motor 26. In order to remove the problem of backlash, an anti-backlash mechanism as described in US 9,170,189 may be included. This mechanism is shown in Figure 1D, and preferably comprises a pneumatic piston 27 and cylinder 28. Figures 2 &3 Figures 2 and 3 illustrate a pull test transducer 200 usable in a bond test apparatus or cartridge according to an embodiment of the present invention. In the pull test transducer 200 shown in Figure 2, a pull test tool 210 is attached to a backing plate 211 by a tool mount bracket 270 having a pair of parallel flexure arms 272. The backing plate 211 may be a portion of a bond test apparatus, or alternatively the backing plate 211 may be a portion of a cartridge which is attachable to a bond test apparatus. The flexure arms 272 are fixed at one end to the backing plate 211 by screws 273. The pull test tool 210 is mounted on the free ends 275 of the flexures 272, so that the application of a pull force in a downwards direction to the test tool 210 will cause downwards deflection of the free ends of the flexures 272. As shown in Figure 2, a magnetic linear encoder 230 is mounted on the pull test transducer 200. The features of the magnetic linear encoder 230 are illustrated in Figure 3. The magnetic linear encoder 230 consists of two parts: a linear encoder readhead 240, and a magnetic encoder code strip 250. The magnetic encoder code strip 250 comprises a plurality of alternating north (N) and south (S) magnetic poles, and the magnetic linear encoder readhead 240 is a printed circuit board (PCB) with sensing components that are configured to sense changes in magnetic field as the magnetic encoder code strip 250 moves linearly past the readhead. In the preferred embodiment illustrated in Figure 2, the magnetic encoder code strip 250 is coupled to the free end of the flexures 272, while the linear encoder readhead 240 is mounted to the backing plate 211 by screws through screw holes 245. The linear encoder readhead 240 is thus fixed in position relative to the fixed ends of the flexures 272, while the magnetic encoder code strip 250 moves upwards or downwards when the flexures are deflected. In order for the processor to convert a measured flexure deflection into a measurement of test force, the relationship of flexure-free-end-position versus load-on-flexure must be calibrated or known first. Linear encoders output a known number of counts for every interval of the scale that moves past the readhead. For example if the readhead and scale produce 10,000 counts / mm, then this defines the encoder part of the rate in terms of displacement. In order to calibrate the system, known loads are then applied to the flexure according to standard calibration practices in the art. If the transducer is found to have flexure displacement of, for example, 3mm for 100g applied force, then the loading rate is 300 counts / g. This loading rate can then be stored in the processor and used to convert sensed flexure deflections back into measurements of force. The relationship between deflection and load is based on the physical material properties of the flexures and is linear as long as the deflections are within the materials elastic limits. This relationship is governed by Hooke’s law. When a pull test is performed, a pull force is applied to the test tool 210, causing deflection of the free ends of the flexures 272 relative to their fixed ends. As a result, the magnetic encoder code strip 250 is displaced linearly downwards past the stationary encoder readhead 240. The flexure beam displaces linearly in relation to the applied force, so a measurement of flexure deflection can be reliably converted into a measurement of the force applied to the flexure. The readhead 240 senses the changes in magnetic field as the alternating magnetic poles of the code strip 250 are displaced past the readhead, so that the code strip 250 functions as a distance scale which is readable by the readhead. The linear encoder readhead 240 detects linear movement of the code strip 250 and encodes the position of the code strip 250 relative to the readhead 240. The linear encoder readhead 240 converts the encoded position of the code strip 250 into a digital output signal, and transmits that signal to a processor (not shown). The processor then decodes the digital output signal from the linear encoder into a position measurement which indicates the magnitude of flexure deflection during the pull test. As the stiffness of the flexures 272 is known from calibration procedures that are standard in the art of bond testing, the processor converts the linear deflection measurement into a measurement of the pull force applied to the test tool 210. This apparatus provides significant advantages compared to prior art bond test transducers in which strain gauges are used to measure force directly by sensing the strain on a bent flexure. The linear encoder used in the present invention advantageously provides a direct digital positional encoded output without the need for post-amplification or conversion from analogue to digital. The readhead and code strip are also easy to install, remove and replace using screws, compared to the painstaking and unreliable process of adhesive bonding and wire bonding which is required to connect a strain gauge to a transducer flexure. Although a magnetic linear encoder 230 is illustrated in Figures 2 and 3, this may be replaced by an optical linear encoder consisting of an optical encoder readhead and a readable distance scale marked at regular intervals. Likewise, although the invention is illustrated in Figure 2 in the context of a pull transducer, linear encoders may be similarly used to sense flexure displacement, and thus test force, in transducers designed for push testing or shear testing. Prior Art Shear Loadcell Figures 4A, 4B and 4C illustrate a traditional prior art shear loadcell 400 used for shear testing in a bond test apparatus. The loadcell 400 has a tool tip 410 which is connected to a shank 420 by a pair of flexures 430. Multiple strain sensing elements 440 are connected to the two flexures 430, and connecting wires 450 from the strain sensing elements extend through a central passage in the shank and out of a cable exit 460. A cover 470 is attached to protect the sensing elements in use. During a shear test, the loadcell shank 420 is held stationary in the bond test apparatus, and a bond is biased laterally into the tool tip 410, typically by moving a stage on which the bond is held. The bond and the tool tip 410 exert a shear force on one another, and the flexures 430 are deflected sideways under the shear force. The strain on the flexures is sensed by the strain sensing elements 440 and communicated to the bond test apparatus as an analogue electrical signal. As discussed above, this arrangement has a number of drawbacks which arise from the requirement to accurately bond the strain-sensing elements to the flexures. Shear Loadcell using Linear Encoder Figures 5A, 5B and 5C illustrate a shear loadcell 500 according to an embodiment of the present disclosure, while Figures 6A and 6B illustrate experimental results obtained using such a device. Like the shear loadcell of Figures 4A-4C, the shear loadcell 500 has a tool tip 510 which is connected to a shank 520 by a pair of flexures 530. Instead of strain-sensing elements, however, the shear loadcell 500 has a linear encoder (illustrated in Figures 5A-5C as a linear optical encoder, though a linear magnetic encoder could also be used) made up of an optical encoder readhead 540 and an optical encoder scale 550. The optical encoder scale 550 is mounted on an arm 555 which extends from the deflectable end of the loadcell (the same end as the tool tip 510), while the readhead 540 is mounted to the shank 520. During a shear test, the loadcell shank 520 is held stationary in the bond test apparatus, and a bond is biased laterally into the tool tip 510. The bond and the tool tip 510 exert a shear force on one another, and the flexures 530 are deflected sideways under the shear force. The deflection of the flexures causes the arm 555 and the optical scale 550 to move sideways relative to the readhead 540 on the shank, so that the encoder readhead senses the displacement and sends an encoded digital signal to the processor in the bond test apparatus. The processor then decodes the digital signal and converts the displacement measurement into a measurement of test force using stored calibration data for the shear loadcell 500. The linear encoders used in the present apparatus produce a digital signal output. Absolute encoders give a position output directly. For this to happen the scale or code strip is encoded with the position information and the encoder reads and outputs this. One advantage of absolute encoders is they can be switched on and straight away they will broadcast their location without having to see a reference position. Incremental encoders by contrast only broadcast numbers of pulses (or a sin &cosine waveforms) relative to any motion. Frequency and number of pulses relate to speed and distance of travel. Scales or code strips comprise of alternating dark / light equally-spaced lines. In some embodiments, encoder scales consist of black (non reflective) and light (reflective) lines of equal width, which together form a line pair which repeats at known intervals. Optical encoder scales are typically formed by a continuous pattern of line pairs, optionally with a secondary pattern with greater spacing which is used as a reference, or for error correcting. A less common alternative is the nonius encoder, which uses a number of fixed pitches to create an absolute encoder. The conversion from position to force (conventionally given in units of grams) can be as basic as 300 counts = 1 gram force. In this case each count of sensed flexure deflection = 1 / 300 = 0.00333333g of applied force on the flexure. In a preferred embodiment which is particularly suitable for shear testing applications, the encoder readhead 540 may be made up of an optical encoder chip linked to a 26-bit interpolator chip, and the optical encoder scale 550 may have a scale of 0.080 mm per line pair.The interpolator chip acts as a digitiser which digitises sin / cosine signals which it receives from the optical encoder chip as the optical encoder chip moves over the optical scale. Using these components, the linear encoder can measure displacements to very high resolutions, typically down to 1.19 nm / count. Measuring flexure displacement in shear transducers is a significant challenge because the deflection during a shear test is very small - typically 0.090-0.100 mm. This makes accurate measurement of the displacement extremely difficult. With this linear encoder arrangement, however, the present inventors are able to achieve approximately 84 million counts for a flexure deflection of 0.100 mm. In the exemplary optical linear encoder with a line pair pitch of 0.080mm and a 26bit encoder readhead, then for 0.080 mm travel the readhead would measure 2A26 counts or 67,108,864. In order to measure applied force, this information on deflection must be calibrated against applied loads of known magnitudes. If the flexure deflects 0.1mm for 250 grams of force, for example, then the linear encoder will output: 0.100 / 0.080 = 1.25 where 0.080 is the scale pitch; 1.25*67,108,864 = 83,886,060 counts so 250g of applied force produces 83,886,060 counts of encoder travel. The relationship between applied load and flexure deflection (counts from the encoder) can then be calculated as: 83,886,060 / 250 = 2.98 x10-6 counts per gram of applied force; or 0.1mm / 83,886,060=1.19nm flexure deflection per encoder count. Therefore a flexure deflection of 1.19nm = 2.98 x 10'6 grams of applied force on the flexure. Actual shear transducer measurements were obtained by the inventors for a generic shear transducer with a designed 0.1 mm displacement, an optical linear encoder with a line pair pitch of 0.080mm and a 26bit interpolator. This experiment determined that: Actual full-scale displacement (0.072mm) = Encoder pitch (0.000,000,001,19mm) x 60,628,160 counts. So, for a 250g pull transducer the smallest unit of measure will be: Transducer capacity (250g) + Counts (60,628,160) = 0.000,004,12g. Testing the same encoder arrangement with a generic push / pull transducer of 3 mm displacement, the inventors found that: Displacement full scale (3mm) - Encoder pitch (0.000,000,001,19mm) = 2,521,008,403 counts. So, for a 100g pull transducer the smallest unit of measure will be: Transducer capacity (100g) + Counts (2,521,008,403) = 0.000,000,039,66g This sensing resolution is high enough to give accurate force measurements for all types of bond test, so this linear encoder solution will work for force transducers designed for push, pull and shear testing. For a conventional push / pull test flexure deflection of 3 mm, for example, this linear encoder arrangement achieves approximately 2.5 billion counts. This resolution is higher than needed for any conventional bond testing application, so that noise may be filtered out, or resolution downgraded in firmware, while still providing highly precise measurements. Similar high-resolution results may be achieved using alternative 5 encoder line pair spacings and encoders other than 26bit encoders. Figures 6A and 6B illustrate test data obtained with a shear test loadcell and the optical linear encoder described above, in which the encoder readhead contains an optical encoder chip linked to a 26-bit interpolator chip, and the optical encoder scale has a scalespacing of 0.080 mm per line pair. Using this setup, the encoder count response was 10 measured for a variety of different applied loads. At each applied loads, the deflection of the flexure caused a corresponding displacement between the encoder readhead and the encoder scale, and the readhead measured the displacement as a number of encoder counts. As shown in Figure 6A, three different loads of 2 g, 100 g and 250 g were applied to the loadcell, each resulting in a steady count being measured by the readhead. 15 Figure 6B charts the encoder counts measured in Figure 6A, showing the high degree of linearity of encoder counts vs applied load. This linearity advantageously makes this linear encoder arrangement both highly accurate and straightforward to calibrate.

Claims

1. A bond test apparatus comprising:a flexure having a fixed end and a free end;a test tool, the test tool being coupled to the free end of the flexure and configured to contact a bond during a bond test, so that application of a test force between the test tool and the bond during a bond test displaces the free end of the flexure relative to the fixed end of the flexure;a linear encoder; anda processor which is configured to receive an output signal from the linear encoder; wherein the linear encoder is configured to measure a displacement of the free end of the flexure relative to the fixed end of the flexure;wherein the linear encoder is configured to deliver a digital output signal to the processor, and the processor is programmed to convert the digital output signal into a measurement of the magnitude of the test force.

2. A bond test apparatus according to claim 1, in which the linear encoder comprises an encoder readhead and a detectable member which is detectable by the encoder readhead, in which one of the detectable member or the encoder readhead is coupled to the free end of the flexure, and the other of the detectable member or the encoder readhead is fixed relative to the fixed end of the flexure.

3. A bond test apparatus according to claim 2 in which the encoder readhead is removably couplable to the free end of the flexure, or in which the detectable member is removably couplable to the free end of the flexure.

4. A bond test apparatus according to claim 2 or 3 in which the encoder readhead is removably couplable to an encoder mount which is fixed relative to the fixed end of the flexure, or in which the detectable member is removably couplable to an encoder mount which is fixed relative to the fixed end of the flexure.

5. A bond test apparatus according to claim 2, 3 or 4, in which the linear encoder is a magnetic linear encoder; and in which the detectable member is a magnetic detectable member.

6. A bond test apparatus according to claim 5, in which the magnetic detectable member is a magnetic encoder code strip comprising a plurality of alternating magnetic poles.

7. A bond test apparatus according to claim 2, 3 or 4, in which the linear encoder is an optical linear encoder; and in which the detectable member is a distance scale marked at regular intervals.

8. A bond test apparatus according to any preceding claim, in which the test tool is a pull test tool, and in which the bond test apparatus is configured to perform a pull test on the bond during the bond test, such that the test force is a pull force.

9. A bond test apparatus according to any of claims 1 to 7, in which the test tool is a push test tool, and in which the bond test apparatus is configured to perform a push test on the bond during the bond test, such that the test force is a push force.

10. A bond test apparatus according to claim 8 or 9, in which the flexure is configured to deflect in a vertical plane in response to a test force applied in a vertical direction between the test tool and the bond.

11. A bond test apparatus according to any of claims 1 to 7 in which the test tool is a shear test tool, and in which the bond test apparatus is configured to perform a shear test on the bond during the bond test, such that the test force is a shear force.

12. A bond test apparatus according to claim 11, in which the flexure is configured to deflect in a lateral plane in response to a shear test force applied in a lateral direction between the test tool and the bond.

13. A bond test apparatus according to any preceding claim, in which the bond test apparatus does not contain a strain gauge configured to measure a strain applied to the flexure during the bond test.

14. A bond test apparatus according to any preceding claim, in which the digital output signal from the linear encoder is not amplified before being delivered to the processor.

15. A bond test apparatus according to any preceding claim, in which the processor is programmed to convert the digital output signal from the linear encoder into a measurement of the test force by comparing the displacement measured by the linear encoder with stored calibration data of flexure stiffness and flexure load force.

16. A cartridge for a bond test apparatus, the cartridge being mountable to a bond test apparatus and the cartridge comprising: a flexure having a fixed end and a free end;a test tool, the test tool being coupled to the free end of the flexure and configured to contact a bond during a bond test, so that application of a test force between the test tool and the bond during a bond test displaces the free end of the flexure relative to the fixed end of the flexure;a linear encoder which is configured to measure a displacement of the free end of the flexure relative to the fixed end of the flexure and to deliver a digital output signal to a processor which is programmed to convert the digital output signal into a measurement of the magnitude of the test force.

17. A cartridge according to claim 16, in which the linear encoder comprises an encoder readhead and a detectable member which is detectable by the encoder readhead, in which one of the detectable member or the encoder readhead is coupled to the free end of the flexure, and the other of the detectable member or the encoder readhead is fixed relative to the fixed end of the flexure.

18. A method of measuring a test force in a bond test apparatus, the method comprising the steps of:providing a flexure, the flexure having a fixed end moveable relative to a free end and a test tool coupled to the free end of the flexure;providing a linear encoder arranged to measure the displacement of the free end of the flexure relative to the fixed end;contacting a bond with the test tool;applying a load force to the first end of the flexure, so that the test tool applies a test force to the bond;measuring, using the linear encoder, the displacement of the free end of the flexure relative to the fixed end of the flexure;delivering a digital output signal from the linear encoder to a processor; andconverting the digital output signal into a measurement of the magnitude of the test force.

19. A method of measuring a test force using the bond test apparatus of any of claims 1 to 15, the method comprising the steps of: contacting a bond with the test tool;applying a load force to the first end of the flexure, so that the test tool applies a test force to the bond;measuring, using the linear encoder, the displacement of the free end of the flexure relative to the fixed end of the flexure;delivering a digital output signal from the linear encoder to the processor; and converting the digital output signal into a measurement of the magnitude of the test force.

20. A method according to claim 18 or 19, in which the digital output signal from the linear encoder is delivered to the processor without amplification.

21. A method according to claim 18, 19 or 20, in which the digital output signal from the linear encoder is converted into a measurement of the test force by comparing the displacement measured by the linear encoder with stored calibration data of flexure stiffness and flexure load force.

Images