Waveguide optical deflector for an optical disk trackingactuator using a surface acoustic wave device

Akira Arimoto, Kouji Muraoka, Takeshi Shimano, Kazuto Senda, and Sachiko Ishikawa

Optical disks are widely used in data processing. However, one of the problems is slow access time

compared with that of magnetic disks, because in the former dual mechanical actuators such as a galvanomir-ror or voice coils are used. To achieve higher speed access, it is necessary to replace mechanical components

with an electronically controlled system. A surface acoustic device is introduced into the fine access actuatorand its feasibility is tested. Less than 1-ms fine access can be expected.

1. Introduction

Optical disks are widely used in data processingsystems as well as in compact disk and videodisk sys-tems. Rapid progress is now being achieved in termsof data capacity and rewritability. Data capacity isexpected to increase with shorter wavelength laser di-odes' and rewritable optical disks with magnetoopticand phase change recording materials now available.2

However, current major unsolved problems are slowdata transfer and tracking access times compared withthose of magnetic disk systems. This slowness is be-cause optical disks employ a dual actuator system com-prised of coarse and fine actuators, each one utilizingmechanical components such as galvanomirrors.These components have slow response times and takemore than 20 ms to stabilize, because of the mechanicalvibration. It is difficult to design a coarse actuatorwithout mechanical components, but these compo-nents can be removed from the fine actuator. Forfaster access, waveguided and electronically controlledcomponents are promising. In this study, ultrahighspeed tracking access is attained using a surface acous-tic wave (SAW) deflector in the optical disk pickup.

II. Necessity for Reduced Access Time

Figure 1 compares the access times of optical andmagnetic disks. In conventional 13.3-cm (5.25-in.)

The authors are with Hitachi, Ltd., Central Research Laboratory,1-280 Higashi-Koigakubo, Kokubinji, Tokyo 185, Japan.

Received 14 July 1989.0003-6935/90/020247-04$02.00/0.© 1990 Optical Society of America.

optical disk systems, the total access time is 60 ms. Ofthis 60 ms, coarse access occupies 40 ms and fine accesstakes 20 ms. On the other hand, the access time of themagnetic disk is only 20 ms, that is, one-third that ofthe optical disk. This difference is due to the trackpitch difference between the two disk systems and theuse of a dual mechanical actuator in the optical disksystem. If an electronically controlled SAW deflectorreplaces the mechanical fine actuator, the fine accesstime can be expected to be <1 ms, which is negligiblecompared with the coarse access time. This reductionin the fine access time is investigated in detail.

Ill. Optical Pickup Usirnj a SAW Deflector

The optical pickup configuration is shown in Fig. 2.3Ti-diffused LiNbO3 is used as the waveguide. ,The

laser beam emitted from the laser diode is sent to thewaveguide through a rutile prism and a coupling lens.SAWs are excited in the waveguide by an Al interdigi-tal transducer (IDT) located on the substrate. Thecenter frequency iS 250 MHz. The laser beam de-flected by the SAW is collimated through a cylindricallens placed behind the SAW deflector and is sent to theoptical disk through the beam splitter and objectivelens. The RF signal and the spot control signals areobtained by the photodetectors. A tracking servo-mechanism is utilized to reduce tracking error signals.A voltage controlled oscillator (VCO) circuit is used tochange the tracking error voltage to correspond to theSAW frequency change. The grating placed betweenthe cylindrical lens and the beam splitter is used to getthe tracking signals.4 Plus minus first-order beamsare used for this.

IV. Fundamental Properties of the SAW Deflector

Before performing the tracking experiments, severalbasic properties of the SAW deflector were investigat-ed.

10 January 1990 / Vol. 29, No. 2 / APPLIED OPTICS 247

20 40 60 (ms)

ca) BUL PU

(b)SA PU

100

80

60

40

20

.

11

c

L

(c) MAGNETIC C A I S

DISC

Fig. 1. Comparison of access time betweendisk systems.

optical and magnetic

50 200 250 300

SAW Fre(iquncY (Mliz)

Fig. 3. Deflection bandwidth.

A. Deflection Bandwidth

The diffraction efficiency vs SAW deflector fre-quency properties are shown in Fig. 3. When an IDT(A) with 250-MHz center frequency is used, over 40%efficiency is obtained between 210 and 290 MHz. Thebandwidth is 80 MHz. This value corresponds to a 40disk track bandwidth when using an objective lenswith 4.5-mm focal length. The deflector can covermore than 100 disk tracks when two IDTs, one with acenter frequency of 175 MHz (B) and the other of 250MHz, are utilized simultaneously. This is representedby the dashed line. These numbers are quite suffi-cient for fine access in optical disk systems.

B. Deflection Speed

The deflection speed between two positions ischecked by switching back and forth between frequen-cies. This speed is detected with a streak camera.The measured switching time is -10 us, as shown inFig. 4; this time ensures that the fine access time is<100 gs-much faster than that obtained when using amechanical deflector.

C. Wavefront Aberration

A 0.349-X wavefront aberration was obtained, as canbe seen in Fig. 5, when the collimated output beam

-T R

t-I O( St

Fig. 4. Deflection speed properties.

from the SAW deflector was sent to a Mach-Zehnderinterferometer. Even though this wavefront is notaberration free, a focused laser spot of -2 Acm can beexpected. The aberration is mainly caused by theerror in displacement of the cylindrical lens placedbehind the deflector. In this experiment, the datatrack pitch is 3 Mm.

Laser Diode

Collimator Lens

Ruti L i N b 03

Objective lens

Detector

I DT

Cylindrical Lens

IDT: Interdigi tal TransducerSAW;Surface Acoustic Waves Beam Splitter

Lens

Mi rror

Fig. 2. Optical pickup using the SAW deflector.

248 APPLIED OPTICS / Vol. 29, No. 2 / 10 January 1990

0

- Detector(a)SAW PU

Def lec. Lens

S i Disc

(b)BULK Pu A Detector

Fig. 5. Wavefront aberration.

V. Configuration Difference Between Conventional BulkPickup and SAW Waveguide Pickup

The optical configurations of conventional andSAW pickups are shown in Fig. 6.

The main difference between these two pickups istheir deflector position. In the conventional bulkpickup, the deflector for tracking is placed between thebeam splitter and the objective lens. In this case, thelaser spot on the detector moves only slightly, evenwhen the laser beam is deflected by the mechanicaldeflector. This is because the laser beam passesthrough the same path on the way to and from theoptical disks, thus going through the deflector twice.In the SAW waveguide pickup, the deflector is placedbetween the collimating lens and the beam splitter,because it is difficult to insert the laser beam into thewaveguide on the return from the disk as in conven-tional pickups. In the SAW pickup, the laser beam onthe detector moves due to deflection of the laser beam.We must take this problem into consideration whendesigning an optical pickup.

Figure 7 shows a five-segmented detector used in theexperiment. A and C detectors are used for TR, andB 1-B3 detectors are used for RF and AF signals. Thelaser beam moves transversely, but in this configura-tion the AF, TR, and RF signals are not interfered withby the laser spot motion on the detectors.

VI. Tracking Experiment Using a SAW Deflector

To perform the tracking, the three-spot methodwhich makes use of a diffraction grating is employed.When the tracking servo is switched off [Fig. 8(a)] andon [Fig. 8(b)], the tracking error and RF signals areshown. When the servomechanism is off, the laserspot is fixed and the tracking error signal and the RFsignal vary, because the laser spot crosses the datatracks due to decentering. When the tracking servo-mechanism is switched on, the continuous RF andsuppressed tracking error signals are observed. TheTR error signals are reduced to one-seventh of thepeak value, ensuring that tracking using the SAW de-vice is performed successfully. The fluctuation of theRF signal amplitude caused by variations in the dif-fraction efficiency that are due to the frequency

E | Lens

L i;<T I isc

BS Def lectorFig. 6. Comparison of the optical pickup configuration: (a) SAW

pickup and (b) conventional pickup.

°- A TR=A-C

Detector Bo3 RF=B, B2 +B 3

Fig. 7. Five-segmented detectors in this experiment.-- A '

__ ~ ~~~~~~~~~ I

(a) -r Fi (I F' F'

_ ~~~~~AF

'1' PI

-R

(b) -o ( D r'4

Fig. 8. Tracking experiment: (a) TR off and (b) TR on.

change in the SAW deflector can be easily correctedwith an automatic power control.

VIl. Track Jump Experiment

Figure 9 shows the waveforms of the track jumpexperiment. Before jumping, stable reduced and con-

10 January 1990 / Vol. 29, No. 2 / APPLIED OPTICS 249

RmS: .061 P-V: 0. 349

- TR Jump

- TR

- RF

Fig. 9. Tracking jump experiment.

Ist

ode

IDT:Interdigital TransducerSAW:Surface Acoustic Waves

Fig. 10. Compact SAW deflector.

tinuous RF signals are observed. When the TR jumpsignal is applied and the SAW frequency is forced tochange from one frequency to another, the TR errorand RF signals fluctuate for a short time but soonstabilize. The horizontal axis unit is 100 As, so a trackjump can be done within 100 As. This means that thefine access time can also be <100 ,us.

Vil. Compact SAW Deflector

In the TR experiment, the optical system is compli-cated and impractical because optical transmittance is<10%. Figure 10 shows the next generation SAWdeflector. A laser diode is attached and a geodesic lensis fabricated for beam collimation. This geodesic lensis chromatic aberration free, and it is quite importantbecause the diode laser oscillates in multiple modesdue to the mode-hopping and the feedback of the laserbeam. More than 30% of the laser beam is insertedinto the waveguide. The deflection was confirmed buttracking was not completed, mainly due to a fabrica-tion error in the geodesic lens.

IX. Conclusion

To reduce access times and replace magnetic disksystems, elimination of mechanical actuators from an

optical pickup is necessary. An electronically con-trolled SAW deflector with a fine access time of <100,s was introduced. Further integration of the opticalpickup, such as fully integrated pickup proposed bythe Osaka University group,5 will reduce the coarseaccess time.

References

1. M. Ishikawa, Y. Ohba, Y. Watanabe, H. Sugawara, and M. Yama-moto, "Room-Temperature cw Operation of Transverse-Mode-Stabilized 680-nm InGaAlP Visible Light Laser Diodes Grown byMOCVD," in Technical Digest, Conference on Lasers and Elec-tro-Optics (Optical Society of America, Washington, DC, 1986),paper TUQ1.

2. M. Ojima, T. Nakao, H. Sukeda, N. Ohta, H. Yasuoka, T. Nishida,and M. Terao, "High Speed Overwritable Optical Disk," Proc.Soc. Photo-Opt. Instrum. Eng. 899, 154-159 (1988).

3. A. Arimoto, K. Muraoka, T. Shimano, K. Senda, and S. Ishikawa,"Wave Guide Optical Deflector for Optical Disk Actuator UsingSurface Acoustic Wave Device," in Technical Digest, Conferenceon Integrated and Guided-Wave Optics (Optical Society ofAmerica, Washington, DC, 1989), paper MBB5.

4. G. Bouwhuis, "Apparatus for Reading a Flat Record Carrier,"U.S. Patent 3,876,842 (1975).

5. S. Ura, T. Suhara, H. Nishihara, and J. Koyama, "An Integrated-Optic Disk Pickup Device," IEEE/OSA J. Lightwave Technol.LT-4, 913-918 (1986).

250 APPLIED OPTICS / Vol. 29, No. 2 / 10 January 1990