Development of an electric performance testing system for ceramic chips using a PZT actuator

A reno-pin contact test controlled by a solenoid motor is a method which is used to characterize electronic chips of electronic and telecommunication devices. Due to its lower efficiency and vibration and noise effects, an automatic efficient noise and vibration-free technique is needed to facilitate higher volumes of electronic chip checking in industry. A new electric performance testing system is developed in which the testing method is controlled by using a Piezoelectric Transducer (PZT) instead of a solenoid motor, which reduces vibration and noise. As the vertical displacement of the Reno-pin contact testing system is very short, using a flexible guide with a PZT actuator in the new Reno-pin contact system, the vertical movement of the Reno-pin is increased many times over that of the existing Reno-pin contact testing using a solenoid motor. The present electric performance testing system with a flexible hinge and a PZT instead of a solenoid motor is able to characterize electronic chips with low cost and high speed without noise and vibration.


Introduction
Nowadays, semiconductor devices are used in most electronic or communication equipment. Especially with miniaturization and highresolution implementation, the fine patterns of COF (Chip on FPC) and FPC (Flexible Printed Circuit) necessitate careful checking. In order to correspond to the fine pattern of the O/S test, it is necessary to develop a high-speed driving actuator. Currently, due to an increase in production and microminiaturization, in which the size of the chip used in semiconductor device production has decreased, it is necessary to produce an improved probe module which is a core part of any testing device with high precision and reliability. For the testing of the electrical properties of the chip, a Reno-pin contact test method is used. Currently, for ceramic chip burn-in testing, a Reno-pin high speed actuator system is used with a solenoid actuator (see Fig. 1).
Several problems occur when solenoid actuators are used for ceramic chip burn-in testing. Among them, electromagnetic noise affects the semiconductor directly and causes problem in drive speed control and fine adjustment of vibration displacement. 2 Wide applications of PZT actuators are observed in teleoperating 10 and energy harvesting at the micro level. [11][12][13][14] In the present testing set up, the main components are the Reno-pin guide, a lever mechanism, a PZT actuator, preload adjustment, a Renopin, and upper and lower fixtures. Details of the components are given in the following sections. Instead of a solenoid actuator, a piezoelectric PZT (Piezoelectric Transducer) actuator-based testing system is developed. A PZT actuator has the advantage of having high resolution, good response, lower power consumption, and less vibration. Hence, a lever amplification structure is designed using a flexible hinge to amplify the vertical driving displacement of the PZT actuator. The design of the prototype was verified by the comparison of experimental results with those of the simulation.  Figure 2 shows the flexure guide in which the left end of the flexure is rigidly fixed. Two grooves are made on the top and bottom faces of the flexure guide along the middle region, as shown in Fig. 2. b is the width of the flexure guide; h is the thickness; R is the radius of the grooves of the flexure guide; θ max represents the maximum angular rotation along the z axis. The following parameters are used for the perfect design of the flexure guide.

Mechanism Design and Operating System
(3) Where K = Correction factor, K t = Stress concentration factor, E = Young's Modulus. Vibration displacement of the PZT actuator (t) is 0.05 mm. Movement of the Reno-pin of the probe is increased 6 times with the help of lever mechanism as shown in Fig. 3. 3 All components of the lever mechanism in the present testing system are designed considering the theoretical design formula shown equations 1 to 4. Special attention is given during the design of critical sections where highest levels of stress concentrate during the working of the lever mechanism. We used a safety factor of around 1.5 during the design of the lever mechanism to prevent plastic deformation during operation. Figure 4 shows a 180 o rotated view of the meshed electric performance testing system with a lever mechanism, excluding the PZT shown in Fig. 9. The dimensions of different components and sections of the lever mechanism shown in Fig. 5 are given in mm. Specifications of the used PZT actuator in the present testing system are given in Table 1. The joint of the PZT with the lever mechanism are shown in Figure 6 and details of all components of the present    Joint of PZT with lever mechanism electric testing system with PZT preload adjustment as shown in Figure  6 are shown in Figure 7 for understanding all the components of the present testing system. When power is supplied to the PZT it expands and crosses a line as shown in Fig. 6. But when power is removed it contracts, which is 1/500 times of the expansion. Therefore, a certain amount of power is given for tightening the PZT actuator bolt. 4 The expansion and contraction of the PZT are accomplished by increasing and decreasing electric power which causes the Reno-pin to move upward and downward with the help of the lever mechanism, as shown in Fig. 4 and 5. Power supply to the PZT actuator is controlled according to the required moving frequency of the Reno-pin (see Fig. 15 (a)) which pushes the lever upward and downward by applying force.

Finite Element Simulation and Discussion
The actuator mechanism is designed using the design software CATIA V5 as shown in Fig. 7. Then it is imported into ANSYS Workbench 11.0 for structural analysis. PZT actuator, fixture, Reno-pin guide, lever and preload adjustment are shown in Fig. 7. The main objective of 3D finite element analysis is to check the amplification factor of the vertical displacement of the Reno-pin applying the vertical displacement to the top contact joint of the PZT Actuator.
When the Reno-pin contact testing system is modeled in an FEM environment using ANSYS finite element software, the PZT actuator is removed and the vibration vertical displacement produced by the PZT actuator is applied at the top contact zone of the PZT actuator with the flexible lever mechanism. as shown in Fig. 8.
Finite element meshing of the whole structure shown in Fig. 8 is given in Fig. 9. At different parts of the structure, the mesh density is different. At critical zones, where higher levels of stress can concentrate, higher mesh density is made to achieve higher accuracy of the solution, as shown in Fig. 9.
The structure, as shown in Fig. 9, is made of two different materials, namely Al6061 and S45C. The effective mechanical properties of those materials are given in Table 2. A 3D hexagonal solid element is used to mesh the structure. The number of elements used in the present FE simulation is 87116. After meshing the structure, as shown in Fig. 9, the mechanical properties, as shown in Table 2, are used for the solution of the problem using 3D FE modeling.
The bottom surface of the fixture is fixed, which means all degrees of freedom of its bottom surface are stopped. Also, a vertical displacement of 0.05 mm is applied at the contact point of the PZT with the lever mechanism. As for the remaining parts of the structure, the reliable boundary conditions are set to default by the ANSYS FEM software. Figure 10 shows the vertical deformation of the lever mechanism. From the figure it can be clearly observed that at the contact point of   Fig. 10 Directional deformation of the lever mechanism the PZT actuator with the lever mechanism, the input vertical displacement is 0.05 mm and the output vertical displacement at the contact point of the Reno-pin with the lever mechanism is 0.3535 which is more than 6 times of the input vertical displacement. The vibration amplification setup as shown in Fig. 11 was designed to carry out harmonic analyses to determine the frequency and amplitude of different vibration modes. The frequencies of six different modes were determined, which are shown in Table 3. Among them, the frequency of the first mode is equal to the natural frequency. At the upper fixture for mode 1, the starting frequency is 213.12, which is equal to the natural frequency. At the first and second modes the lever moves down and upward, respectively. Same phenomenon is also observed in third and fourth moves and in these modes ten thousand lever vibrations are observed. For the fifth and sixth modes, the vibrations of the lever and the upper fixture are somewhat different from the others. Harmonic analysis is carried out from the results of the modal analysis on the contact point of the PZT. To observe the variation of the amplitudes at different modes, frequency response analysis is carried out, which is shown in Fig. 12. Figure 12(a) shows amplitude as a function of frequency. In Fig. 12(a), six peaks are observed where amplitude changes abruptly. Among them, in the second mode, (frequency = 257.11 HZ) and in the fourth mode (frequency = 568.73 Hz), the amplitudes are equal and higher than those of other peaks [7].
In order to employ the present developed testing structure a prototype PZT Actuator drive controller is required. The controller shown in Fig. 15(a) was connected to the PZT actuator of the present testing structure and to a computer to control the testing structure and this controller is shown in Fig. 15(a) and is controlled using LabVIEW Fig. 11 Different modes of the testing system during simulation by 3D finite element modeling Software and the movement of the Reno-pin is stored in the computer which is shown in the computer screen in Fig.15 (b). Sin wave program for the Piezo Amplifier controller and a constant frequency and voltage of the analog is given to control. At this point, the frequency and amplitude of the voltage were controlled. 5,6 The testing capacity of the previous testing system using a solenoid motor is 3000 times per minute. The main drawbacks of the previous testing system are vibration and noise and the inability to test full 24 hours. Due to the high-speed production of semiconductor chips a need has developed to accelerate the testing of chips. The required testing numbers to characterize the electronic chips is increased to more than 3600 per minute in the present developed testing system using PZT (see Figs. 13 and 14) instead of a solenoid motor, in which the maximum distance between the chips and Reno-pin is kept at 0.35 mm. Other advantages of this testing system are freedom from noise and vibration and the ability to operate 24 hours without stopping. To identify the driving speed of the probe of the Reno-pin, high speed vision cameras were used and the filming capacity of these high speed cameras was 500 fps and these cameras identified 60 meetings of the probe in one observation.
Reno-pin movement of the present testing system during operation obtained by 3D finite element modeling is shown in Fig. 10 and its maximum is 0.3535. On the other hand, the movement of the Reno-pin was measured experimentally using a displacement sensor (Keyence LK-G35) as a function of time and this is shown in Fig. 16 and its maximum displacement is 0.3499. Variation of the vertical displacement of the Reno-pin of the present testing obtained by 3D finite element modeling to that of the experimental study is about 1%. So this variation is within a reasonable limit. Theoretical study of the movement of the Reno-pin of the present testing system during operation by 3D finite element modeling as shown in Fig. 10 and experimental analysis as shown in Fig. 16 agree well qualitatively and quantitatively.

Conclusions
To check the electrical properties of the key electronic components of IT products, a noise free high electric performance system was developed using an electromagnetic PZT actuator moving a Reno-pin probe instead of a solenoid Actuator. 8 A reliable prototype of the testing mechanism has been developed by the analysis of 3600 vibration tests at 60 Hz. As can be seen from the analysis of 200 Hz, a PZT testing actuator device without constraints can be designed to avoiding resonance and higher performance can be achieved. A comparative study of the vertical movement of the Reno-pin in the present testing system during operation using 3D finite element modeling and the experiment show the reliability and appropriateness of the present developed testing system.