Experiments of the fiber-connected interferometer for MIIRA project K. Sato, J. Nishikawa, M. Yoshizawa, T. Fukushima, Y. Toni, K Matsuda, K. Kubo, H. Iwashita, S. Suzuld, D. Saint-Jacques MIRA Project, National Astronomical Observatoiy, Mitaka, Tokyo, 1 8 -8588, JAPAN

ABSTRACT Fiber optics is useful for connecting the element telescopes of a long-baseline interfemmetet ExperimUs with a fiberconnected inteferometeraie canied out as a xirt ofMIRA-L2 interferorneterunder constnic1ion MIRA-L2 is an inteifemmeter consisting of two 300mm siderostats whith are placed on a 30m south-nozth baseline. For this IseIine, it is necessaly to use fibers ofabout lOOmlong Fiber optics ll be essential for the interfemmeteic annection oflarge telescopes, such as Subam, c3errun arI Keck on the surnirñt ofMauna Kea In this case, fiber length ofabout 1 km is needed We achieved white light fringes using lOm and lOOm fibers. With the lOm fiber, sibffity of 0.7 s achieved for both He-Ne laser and white light

sours. Visibility of 0.7 for white light s ubtained by compensating the dispersion effect of fiber with ass plates. Experiments ll be exteiided to longerfibeis such as l000m.

Keywords: o interfezometer, fiber oics, highresolution imaging

1. INTRODUCTION Experiments offiber-connected interferometerare canied out as a jut ofMIRA-L2 intezfemmeterundei consliuctiolL MIRA-L2 is an iriterferometer consimng oftwo 300mm sidemstats which are placed on a 30m south-rrth baseline. 11 outhne of MIRAanI the basehne is now extided to 30m. For this 1.2 is descnlxd by Sato et aL'. The MIRA-L2 system has been giaded baseline, it is necessazy to use thefiber ofabout lOOm long. Fiber oplics ll be essenimi for the itIteifeJometertO connect the Iaige telescopes, such as Subam, Gemini, and Keck on the innnit ofMaunaKea (e.g Nishikai et aL2, Maiiotü et aLa). Forthis case,

fiber length of about l000m is needed. By using fibeis, we can reduce the rnimber of miirors to guide the adigii along conqlicatedand Iongpath& Single-mode fibers act like good filters ofthe beam pmfile. This fealuie makes the cahbzation of the thnge visibility easier One ofthe difficulties to use fibers as a wave-guide ofthe inleifemmeter is the coupling OfStaiiight to the fiber, since the athght is dancing aixi speckled because ofatmospheric tuthuleiice (e.g. Coudé dii Foresto aix! Ridgway', Shakian

and Rodch&). Hoiver, once the light is pit into the fibei diffiion, oiica1 surface quality and mihgmnait are no longer prublems. Ciitular core fiber is basically neutral th respect to polazization. However, the poIazation ate of output beam is generally an effipse of iarJom shape azl orientation, because ofunnted birefringence due to bend, tst, tenqeiaUne variation and so on ( Coudé du Foiesto and Ridgway4, Shaldan and Roddier ). Acllc the fringes disappear vhen the fiber is shakeii slowly by haid The fiber leigth has to be made eipal. Ifnot, the dispersion offiber reduces the VISibility. It is shown that the made veiy small at neir infrared wivelengths effect of fiber diersion is smaller than that of regular glass aix! the effect can (Dyerand Chiistenseii7, arl Coudti duForesto etaL). Reynaud et aL diseussed about the control offiber leiigth& lathispaper, we descnbe the experiments with fibers to develop a fiber-connected interferometet In section 2, meaairemenls of fiber length are

KS.(corresporidence): e-mail: koichioptikm&nao.acJp; telephone: +81422-34-3812; fax +81-34-3812

1102

In lnterferometry in Optical Astronomy, Pierre J. Lena, Andreas Quirrenbach, Editors, Proceedings of SPIE Vol. 4006 (2000) • 0277-786X/00/$1 5.00

Downloaded From: http://proceedings.spiedigitallibrary.org/ on 09/16/2015 Terms of Use: http://spiedigitallibrary.org/ss/TermsOfUse.aspx

descnbeci In section 3, wedescnbe inteifemmeiric e'qeiiments thfibers oflOm and lOOmlength. For the experiments in tIs paperS we use the staiidard fibers (low inrnnsic-bireftingence) of lOni, lOOm aix! l000m long. The lOm fiber is a single-mode fitr

male by OZ O LtcL (mode field diametere 4 it in, open1ing ivelength: 633nm). The lOOm arid l000m ones are tl production ofSpecTran Optics Cornxmy (mode field diameter 4.3 ji m, operatingwavelength: 63Qnrn).

2. FIBER LENGTh MEASUREMENT It is necessary to measure fiber length of 1OO1OOOm long by some optical method in order to eq1ize the length for the interferometiic experiments the length ofshort fiber cn be measuzedby a tape measure. An optical fiber analyzer (e.g OTDR optical time domain zfiectometer) can measure the fiber length, but tl accumcy ofO.5-lm is inequate. A laser iange meter (MM3OR, Sokkia Co. LtcL)s used forthis purpose. This laserrange meter is a hand-lld type and is used for general purpose in the air The measurement of distrn is made by analyzing the phase difference between the rmn light from the target aii the reference light inside the meter iith U) lighi(650nm) modulated at two frequencies, 470 kHz aixi 60 MHz Ths meter has X D), where D is the distance to be msuiei This error becomes aboul±Smm to lOOm. an accuracy of The optical layout ofthe equipment is shown in Figure 1.

Li

Fiber (L)

L2 Figure 1. Layout ofthe equipment to measure fiber length The laser light fromthe meter is fed into the fiberby a lens. The output light from the fiber is p1 diiectly into the icciver1ens of

the metet The fiber length L is calculated by the equation, L= (2 X LM—LO> where L is the fiber length measured th the laser ninge metei LO the length in the air (LO=Ll+L2; Li and L2 aie the lengths shoi in Figure 1), LM the reading of the laser iange meter, IL the fiber length measured th a tape measure aii n the refmctive index offiber (n1.4564 is assumed). The results ofthe measuiements are thoi in Table 1. In this table, it is found that the length measuid th a tape mxre is longer than the one with the laser zange meter by about 1.4 percent On the other hand, the diffeiences (no.l-no2.) of1engh of two lOm fibers are —9mm and -5mm. measured with the laser mnge meter arI a tape measure, respectively. The two lOam fibers are equalized by measuring the length with a 100-rn tape measure, simultaneously The difference (no. i-no.2) of two lOOm fibers is fouixl to be +7mm by measuring th the laser range metet 11 length of l000m fiber is measured by the laser

range meter This length is calculated by adding two times the length of 3 18.928m, which coiresporis to the 470kHz modulation, to the reading LM of the laser range meter The length LL of i000rn fiber listed inTable 1 is the 1ue given by the maker It is found that the length of I000m fiber can be measured with the laser range meter Within this error of a few mm, we can search the position of the wine light fringes with the interlërometer described in the next sect on.

1103

Downloaded From: http://proceedings.spiedigitallibrary.org/ on 09/16/2015 Terms of Use: http://spiedigitallibrary.org/ss/TermsOfUse.aspx

Table 1. Fiber lengths measured th the laser range meter MM3OR (unit in rn) LM: the reading of MM3OR, LO: the length in the air (LOL1+L2; Li andL2 are

the lengths sho in Figure 1), L: the fiber length iIcu1ated by the equation, L=(2 X LM—LO)/n, n: the refractive index of fibei, IL: the fiber length measured with a tape measure.

If)

L

LL

(L-LL)

7.625

0.286

10.275

10.133

+0.142

LM

fiber lOm-no.1 lQrn-no.2

7.632

0.286

10.284

10.138

+0.146

iOOm-no.1

73.944

0.185

101.485

100 000

+1485

lOOm-no.2

73.989

0.185

101.478

i000m-no.i

90.781

0.185

1000.473

100.000 +1.478 987.000 +13.473

3. INTERFEROMETRIC EXPERIMENTS The layout ofthe inteifemmetric expetiment is shown inFigure 2. Two light sources, He-Ne laser aml white light lamp, are used

for the expemneit The light goes thraugh a pinhole axiii is cothniated by a lens. The ve front is dwided with two diaphragms inthe cothmatedbeaxu Eachbeam is fed into a singlernode fiberby a lens. Fibers of lOmand lOOm long are

path # 1

lens

FC

fiber

FC

lens

dispersion compensator

Figure2. Overall layout of the interferomatric experiment

1 104

Downloaded From: http://proceedings.spiedigitallibrary.org/ on 09/16/2015 Terms of Use: http://spiedigitallibrary.org/ss/TermsOfUse.aspx

Figure 3. Fnnges of laser and white light

visibility of white light 0.8 0.7

A.

0.6

>, 0.5

lOm fiber -"-lOOm fiber

6 0.4 Cr)

> 0.3 0.2 0.1

0 —4

—2

0

2

4

compensation (mm) Figure 4. Fnngevisibilites of white light

used for the experiment. The output beams from the fibers are collimated again by a lens. The delay line consists of an electnc moving stage, on which the exit of fiber in jnth #1 and its collimator lens are set up. 'fl rx.1 and no.2 fibers are set in puths #1

and #2, respectively The beams are combined again by a lens. Tbe interference fiinges are detected with a CCD TV cameia The compensation of chromatic dispersion is made with a glass compensator wiith is inserted just after the exit of the fiber 11

1105

Downloaded From: http://proceedings.spiedigitallibrary.org/ on 09/16/2015 Terms of Use: http://spiedigitallibrary.org/ss/TermsOfUse.aspx

white light thnges are searched by channg the delay arotmd the position predicted from the differences offiber length oven in

Table I. The white light fringes were found at the po1ion where the length ofno. 1 is shorter by 4mm for the lOm fiber. For the lOOm fibei; the fringe s found at the position there the length ofno.2 fiber is shoiterby 6miu These results are constU the differences offiber length sho in Table 1. Example offringes of lOm fiber is shown in Figuie 3 for Ias and white

th

light. These figms ai thtained by pmcessing digitally the TV images th a personal compilet The visibility is dthved from

the digiiized intensity disth•bution offringe images. Fns of compensation for chromalic dispsion are thoi in Fige 4. The visibilities are plotted against the thickness ofthe ass compensatoi, sere the thickness of gss is expressed as a positive vahie in mm when the gjass in poth #1 is thicker In this se, the ViSibility oflaser fringe is about 0.7. I1 visililities of white light fringes are about 0.4 for the lOm fiber and about 0.2 for the lOOm fibei iespectively. For the lOm fibex a compensation of

+1mm is iequate. The fringes thoi inFiguie 4 are obtained by maldng the corrqensinion of dispersion of +1mm. The compensation for 1OOmfibers notadequate, althoughthe ViSibility is increased a little th a compensation of -2mm.

4. CONCLUSION AND FUTURE WORK We are now canying out the development of the fiber optics interferometer for tI M[RA-I.2 aIx! MIRA-SG projects. We described some results ofbasic expenments for this pirpose. By using a laser range metei we n msui arwi equalize the

length of two long fibeis thin a few mm. The white light flinges axe fouii by changing the delay around the position predicted by this measwement. Moie picise equalization may be male by the msurements of the white light flinges themselves. White lightfzinges with fibers of lOm ari lOOm long were thtained, but tI&S11,i1itY offringes is zath low To

inciase the sibilitç we ll continue the experiments to find out the causes and to improve the innimaita T[1 effect of polarization due to the fiber will be inVeStigated. We will make experimits th poIaiizifion-preseMng fibeis (tIe aixIauI single-mode fibers wei used hue). We made expenments to compensate for tl thormiic dispsion with agIas plate. The expenment th a spectrometer is planned in the next stage to iIIVeStiUe the chmmatic diersion effects more precisely Inteiferometric experiments with l000m fiber 1l also be m1e in the next sage. 1 coupling ofstathght to the single mode ilbei; whose core diameter is veiy small, is a difficult problem, since tl star image is dancing and specklecL The caq11l te male forthe MIRA-L2 system in the nearfutute.

REFERENCES 1. K Sato, J. Nishilcawa, M Yoshizas T. Fukushinia, Y. Machida, Y. Honma, R Kuiban, S. Swuld, Y. Toru, K Kubo, K Matsuda, amlR Iwashita, "Developments ofthe optical andinfiared inteiferometer, MIRA-L2", Pmc. ofSPIE Vo13350, 212-218, 1998.

2. 3.

J. Nishilcawa K Sato, T. Fukushima lvL Yoshiza Y. Machida, Y. Honma, "MiRA-il, M1RA-ffl andMlRA-SG pmject: The future plan oflong-baseline optical/IRinterferometer in Japan", Proc. ofSPIE Vol.3350, 184-191, 1998. J.-lvL Maiioth, V. Coudé th Foresto, G. Penin PeiqianZhao andP. Lena, "Inteifemmetzic connection oflarge gmund-based

4.

telescopes", Astron. Aslrophys. Suppi Ser. 1 16, 381-393, 19%. V. Coudé du Foresto and S. Bidgway, "FLUOR A stellar interferometerusing single-mode infiaiedfibers", ESOPrOC. 39,

731-740, 1992.

5.

S. ShaklanandF. Roddier, "Coupling starlight into single-mode fiber optics", Applied Oçtics, 27(1 1), 2334-2338, 1988

6.

S. ShaklanandF. Roddi "Single-mode fiber optics in a long-basehi interferometer", Ap1ied Oç4ics, 26(1 1), 2159-2163, 1987.

7. 8.

S. D. DyerandD. A. Chiistensen, 'Diersion effects infiber optic inteiferomett', Oi Eng36(9), 2440-2447, 1997. V. Coudé du Foresto, 0. Pemn, andM Bocc "Mininiiation of fiber dispersion effects in double Fourier stellar

9.

inteaferometers", Astron. Astrophys. 293,278-286, 1995. F. Reynaud, J. J. Alleman, and P. Connes, "Inteaferomethc control of fiber lengths fora coherent telescope army', Applied Optics, 3 1(19), 3736-3734, 1992

1106

Downloaded From: http://proceedings.spiedigitallibrary.org/ on 09/16/2015 Terms of Use: http://spiedigitallibrary.org/ss/TermsOfUse.aspx