• Aucun résultat trouvé

A Longitudinal P191 Single Ceramic Piezoelectric Transformer : Comparison between 3D Simulations and Experimental Results

N/A
N/A
Protected

Academic year: 2021

Partager "A Longitudinal P191 Single Ceramic Piezoelectric Transformer : Comparison between 3D Simulations and Experimental Results"

Copied!
4
0
0

Texte intégral

(1)

A Longitudinal P191 Single Ceramic Piezoelectric Transformer : Comparison between 3D

Simulations and Experimental Results

Faiza Boukazouha Research Center

in Industrial Technologies CRTI, ex - CSC Algiers - Algeria

f.boukazouha@csc.dz

Mohamed-Laid Tadjine

Laboratoire de Physique des Matériaux, Faculté des Sciences Physiques, Université des Sciences et de la Technologie Houari Boumediene, USTHB, B.P.32, El Alia,

Bab Ezzouar, 16111, Alger, Algérie

.

Abstract— In present study, three longitudinal piezoelectric transformers with size of 18.5×3.5×2 mm3, 26×3×1.5 mm3 and 24×3×2 mm3 are fabricated and tested with a soft PZT ( P191) piezoelectric ceramic. The input and the output part vibrate in longitudinal extensional mode. Three configurations of the input and the output voltage are proposed. The electrical measurements results are successfully compared to a three dimensional (3D) Finite Element Analysis using COMSOL Multiphysics software.

Keywords— piezoelectric transformer; 3D simulations;

longitudinal vibration; experimental characterisation.

I. INTRODUCTION

The demand of miniaturised electronic devices has increased with the rapid rise of portable equipments such as, notebook-type computer and digital camera. One interesting solution consists in combining inverse and direct piezoelectric effects to realise piezoelectric transformers (PT) in which the coupling between primary and secondary is achieved through mechanical vibration at the resonance frequency.

Piezoelectric transformers, compared with traditional electromagnetic transformers, provide many merits such as high voltage gain, high power density, high frequency, high efficiency, good isolation, no wires, small size, small weight, low loss, inexpensive and absence of electromagnetic noises.

In this article, a longitudinal piezoelectric transformer using ceramic of lead titanate zirconate (PZT) is P191, is made and tested. Both of the input and output part utilise the longitudinal extensional vibration mode along the length of the PT. After a brief description of its functioning, 3D finite elements simulation are realised and compared to experimental results.

II. EXPERIMENTS

The longitudinal PTs discussed in this work were based on a soft-lead titanate zirconate (PZT) ceramic (P191).

Millimeter scale was chosen to facilitate the PT manufacturing (using the facilities available at LMCPA, Laboratoire des Matériaux Céramiques et Procédés Associés, Maubeuge,

France). The values of the three dimensions were determined in order to favor the unimodal behaviour of the primary and secondary (L>w>t), and also to facilitate the experimental measurement of the PT characteristics (voltage gain, resonance frequencies). Moreover, the chosen dimensions are in the range of commercial PTs.

To manufacture the PZT ceramic, P191 powder (from Saint-Gobain Quartz) was pressed into cylinders (40 mm in diameter and 20 mm long) by cold isostatic pressing (CIP) at 300 MPa. The sample was subsequently sintered at 1150°C for 4 hours, and then diced into 18×3×1.5 mm3 pieces, followed by drying at 150°C for 15 minutes. As a final step, samples were also fired at 650°C for 5 minutes. The DC electric field applied for sample poling along the length direction were 1.5 kV/mm. The poling step was performed for 5 minutes in a preheated silicone oil bath held at 40°C. The main P191 ceramic properties and the transformer parameters were measured by the manufacturer (LMCPA) and are listed in table 1.

TABLE 1. P191 MATERIAL CONSTANTS USED FOR CALCULATION AND EXPERIMENT.

The configuration and dimensions of the studied transformers are shown in fig.1. The arrows indicate the direction of polarisation in the ceramic. Two electrodes are placed on the end surfaces. The center position is covered by

Values ρ (density) 7410 kg.m-3

D

s33(elastic compliances) 9.29×10-12 m2.N-1

0 T 33/ε

ε (relative dielectric constant) 3400 d31 (piezoelectric charge coefficient) -247×10-12 m.V-1 g33 (piezoelectric voltage coefficient) 20.2×10-12 V.m.N-1

k33 (coupling factor) 0.75

Qm (quality factor) 60

(2)

an electrode with length of 1mm. The absence of bonding favor the energy conversion.

Fig.1: Basic structure of longitudinal piezoelectric transformer When an alternative voltage V1 is applied between the input electrodes, the longitudinal mode (33) is created in the material through the electromechanical coupling factor k33 (inverse effect). With the transmitting vibration from the primary side, the longitudinal mode (33) induces through the same coupling factor k33 an electrical field in the secondary side (direct effect). A voltage V2 between the output electrodes will appear. By modifying the ground setting, three configurations are considered as it is shown on fig.2.

Fig.2 : Different configurations of longitudinal PT

III- 3D simulations

FEA (Finite Element Analysis) is a powerful tool to study the operation of piezoelectric transformer because the complete set of fundamental equations governing piezoelectricity can be solved in three directions. By considering no body force and free charge, the electrical behavior can be predicted : the electrical potential on nodes of the 3D meshed structure is calculated and compared to experimental results.

The commercial multiphysics software Comsol 3.5.a has been used to carry out a FEA. The primary and secondary sides are created in order to introduce local coordinate systems to define the material matrices. The module

"piezoelectric devices (pzd)" was used to solve the piezoelectric problem and the study type required for frequency sweep is "frequency domain". The unloaded transformer response was calculated for an applied sinusoidal voltage of 1V at the primary side.

Fig. 3 : Voltage gain versus frequency (configuration1)

Figure 3 shows that the voltage gain is approximately independent of the transformer dimensions for the first three modes when the resonance frequency is inversely proportional to the length.

50 100 150 200 250 300

0 5 10 15 20 25

Gain en tension (V2/V1)

masse 1 masse 2 masse 3

Fréquence [kHz]

Fig. 4: Influence of the ground setting on voltage gain

(3)

III. EXPERIMENTAL RESULTS,COMPARISON

Fig. 5 : Piezoelectric transformer test bench

Fig. 6 : Comparison between 3D model and experimental results for PT1

The PT was tested under sinusoidal excitation using the experimental set-up shown in fig.. The transformer was driven by a function generator (HP3314A). A digital oscilloscope (LeCRoy 64Xi-A) was used to measure both the input and output voltages,

TABLE 2. Comparison between 3D model and experimental results for PT1

PT1

Mode 1 Mode 2 Mode 3

fr

(kHz) Gain fr

(kHz) Gain fr

(kHz) Gain (a) 3D model 82.9 13.75 179.2 21 273.6 4.22 Measurements 81.4 11.2 163.8 16.2 269.7 1.61 (b) 3D model 82.1 16.44 179 21.5 271 4.5

Measurements 81.4 8 163.7 20.1 270 1.24 (c) 3D model 73.5 0.62 201.9 1.48 271.9 0.55 Measurements 74.2 0.62 188.7 4.1 267.5 0.63

PT2 Mode 1 Mode 2 Mode 3

(a) 3D model

Measurements 59.9 10.4 120.4 14.3 207 2.5 (b) 3D model

Measurements 60.1 8 120.4 20 207 1.35

(c) 3D model - - - - - -

Measurements - - 135.8 6.5 - -

PT3 Mode 1 Mode 2 Mode 3

(a) 3D model - - - - - -

Measurements 63.3 11.5 126.6 14.5 209 2.5

(b) 3D model - - - - - -

Measurements 63.2 9 126.8 21.2 209 1

(c) 3D model - - - - - -

Measurements - - 191.1 5.4 - -

CONCLUSION

This paper provided a strategy for modeling a single layer Longitudinal PT by a 3D numerical model using Comsol Multiphysics software. The interest and specificity of this method compared to experimental results are demonstrated under sinusoidal excitation conditions.

Experimental results corroborate with theoretical predictions, few differences between theory and experiments were observed in resonance frequencies and voltage gain, they can be attributed to uncertainties in the material parameters and geometrical tolerances.

References

[1] C A Rosen. Ceramic Transformers and Filters.

Proceedings of the Electronic Comp. Symp. (1956). P 205- 211.

[2] H Joo, I Kim, J Song, S Jeong, Minsoo Kim. Piezoelectric Properties of Rosen Type Piezoelectric Transformer Using 0.01Pb(Ni1/3Nb2/3)O3-0.08Pb(Mn1/3Nb2/3)O3-

0.91Pb(Zr0.505Ti0.495)O3. Journal of the Korean Physical Society. Volume 56. (2010). P 374-377.

[3] M Guo, K H Lam, D M Lin, S Wang, K W Kwok, Helen L, W. Chan, X Z Zhao. A Rosen Type Piezoelectric Transformer Employing Lead Free K0.5Na0.5NbO3 Ceramics. J Mater Sci. (2008). P 709-714.

[4] H. Shin, H Ahn, Deuk-Young Han. Modeling And Analysis Of Multilayer Piezoelectric Transformer. Material Chemistry and Physics.

(4)

[5] O. Ohnishi, H Kishie, I Iwamoto, Y Sasaki, T Zaitsu, T Inoue. Piezoelectric Ceramic Transformer Operating in Thickness Extensional Vibration Mode for Power Supply.

Ultrasonic Symposium IEEE. (1992). P 483-488.

[6] M. Guo , D M Lin, K H Lam, S Wang, H Chan, X Z Zhao.

A Lead-Free Piezoelectric Transformer in Radial Vibration Modes. Review of Scientific Instruments 78, 035102. (2007).

[7] Y. H Jeong, S H Lee, J H Yoo, C Y Park. Voltage Gain Characteristics of Piezoelectric Transformer using PbTiO3 System Ceramic. Sensors and Actuators 77(1999). P 126-130.

[8] P. Laoratanakul, K Oschino. Designing a Radial Mode Laminated Piezoelectric Transformer for High Power Applications. IEEE International Ultrasonics, Ferroelectrics.

(2004) . P 229-232.

[9] A Mezheritsky. Quality Factor Concept in Piezoceramic Transformer Performance Description. IEEE transactions on ultrasonics, ferroelectrics, and frequency control. Vol. 53, no.

2. February 2006.

Références

Documents relatifs

In this essay, we explore how an interactive system, situated in public space, may foster explorative activities and enable the transformative power of play.. Through the analysis

Figure 15 shows the displacement measured at the surface of the transducer array when the central element “1” is excited by a sinusoidal signal having a frequency 481

Figure 18: Comparison of the linear and nonlinear models: relative error on the peak power and resonance frequency (left), frequency response for a direct excitation with amplitude U

With a combination of equilibrium and non-equilibrium molecular dynamics simula- tions we have studied the eect of strain on the thermal conductivity of bulk piezo- electric

Furthermore, the electromechanical conversion provided by the piezoelectric effect brings additional advantages : vibration frequency tracking if it varies over

Invariant (4) represents, in fact, the cross term of this energy penalty, while the two square terms, u~~ and fl~ belong to rubber and smectic parts of the free energy. In fact

It is shown that many characteristic features of the relaxation phenomena can be well explained qualitatively by this simple model if one assumes an appropriate

To avoid the influence of the lateral resonances over the operating thickness frequency, a short spacing between the bars in comparison with the lateral wavelenghts must