C om piled Septem ber 7,2016 W e describe an optical bench in w hich w e lock the relative frequencies or phases of a set of three lasers in order to use them in a cold atom s interferom etry experim ent. A s a new feature, the sam e tw o lasers serve alternately to cool atom s and to realize the atom ic interferom eter. T his requires a fast change of the optical frequencies over a few G H z. T he num ber of required independent laser sources is then only 3, w hich enables the construction of the w hole laser system on a single transportable optical bench. R ecent results obtained w ith this opticalsetup are also presented. c 2016 O pticalSociety ofA m erica O C IS codes: 140. 3320,120.3180,140.3550,120.3930. W i thi n the l ast decades,atom i nterferom eters have devel oped i nto a hi ghl y com peti ti ve tool for preci si on m easurem ents. 1 A tom i c fountai ns used as atom i c cl ocks are the best real i zati on of the ti m e uni t. 2 A tom i nterferom etry al so prom i sessensorsto be hi ghl y sensi ti ve to i nerti alforces. 3{6 T he use of sti m ul ated R am an transiti ons to m ani pul ate the atom i c wave packet has proven to be an e ci entway to obtai n hi gh accuracy devi ces. 5,6 In thi s l etter,we descri be a robust,com pact and versati l e l aser system for atom i nterferom eters usi ng alkal iatom s.Such experi m ents basi cal l y need two di erentopti calfrequenci es,w hose di erence rem ai nscl oseto the hyper ne transi ti on frequency.W hen they are tuned cl ose to the D 2 transi ti ons, they are used to cool and repum p the atom s i n a m agneto-opti cal trap (M O T ). W hen far detuned,and phase l ocked,they are used to i nduce sti m ul ated R am an transi ti onsforthe i nterferometer. 7 Si nce l asers are not used si m ul taneousl y for trappi ng and R am an transi ti ons, we have i m pl em ented a techni que to use the sam e two l asers for both functi ons on ourgravi m eter. 8 Ital l owed usto bui l d the w hol el aser setup on a 60 90 cm 2 opti calbench.O url asersetup i sshow n i n gure 1.A rstl aserL1 i s l ocked on an atom i c transi ti on,usi ng FM -spectroscopy 9 on a saturated absorpti on si gnal .T hi s l aser consti tutes an opti calfrequency reference and i s used i n our experi m ent to detect or push the atom s.A second l aser L2 i s al ternatel y used asrepum per orasm asterR am an l aser. Part ofthe outputs ofL1 and L2 are superi m posed on a fast photodetector (PD 12 ) (H am am atsu G 4176) and the frequency ofthe beatnote i sservo l ocked by usi ng a frequency to vol tage converter.A thi rd l aser L3 i s used al ternatel y as cool i ng or as sl ave R am an l aser.T he frequency di erence between L2 and L3 i s m easured w i th a second opti cal beat note on PD 23 . Fi nal l y, both L2 and L3 beam s are superi m posed and di rected through an acousto-opti calm odul atorei therto a m agneto-opti cal trap or to an atom i c i nterferom eter.B oth frequency l ocks of L2 and L3 use the sam e schem e w hi ch i sshow n i n gure 2.T he opti calbeatnote i ssued from the photode...

In atom interferometers based on two photon transitions, the delay induced by the difference of the laser beams paths makes the interferometer sensitive to the fluctuations of the frequency of the lasers. We first study, in the general case, how the laser frequency noise affects the performance of the interferometer measurement. Our calculations are compared with the measurements performed on our cold atom gravimeter based on stimulated Raman transitions. We finally extend this study to the case of cold atom gradiometers.