Need for robust RF MEMS in future space applications O. Vendier, J-C.Cayrou, R. Barbaste, C. Drevon, J.L. Cazaux

Transcription

1 Need for robust RF MEMS in future space applications O. Vendier, J-C.Cayrou, R. Barbaste, C. Drevon, J.L. Cazaux Alcatel Space, 26 Avenue Champollion, BP1187, Toulouse, France

2 General trends for satellites Outline MEMS to get in as MMIC did. Always the same story? Study case 1 : RF MEMS within microwave equipment, MIPA project Study case 2 : RF MEMS for reconfigurable antenna, ARRESAT project Issues to be solved Conclusions

3 Telecom Satellite : Current trends From direct TV to interactive multimedia services- more and more antennas Improved performances

4 Telecom market trends!strong constraints on cost : Size and weight C 30% 32% 30% Ku 55% 48% 42% Ka 5% 10% 18% Autres 10% 10% 10% Power consumption Current situation Analog transparent repeater C, X et Ku band Trends for future generations Improved performances for space segment (Ka, Q/V band, higher bandwidth, ) Increased complexity (Multi-spot coverage, reconfigurability, OBP,...)

5 Standard analog transparent repeater Receiver : LNA and Convertisseur de Fréquence IMUX : Input Multiplexor band pass filters CAMP : Chanel amplifier and linearizer Transmitter : TWTA or solid state (SSPA) OMUX : output Multiplexor pass band filters -

6 Multimedia payload LNA assembly Down conversion OBP Up conversion 2:1 per beam or m for n Ka /IF m-for-n IF/BB m-for-n BB processing : BB/IF m-for-n IF/Ka m-for-n Redundancy Ring LNA assembly 2:1 per beam or m for n Ka /IF m-for-n IF/BB m-for-n - Gain control - Channelization - Routing BB/IF m-for-n IF/Ka m-for-n LCAMP + TWTA + HPI LNA assembly 2:1 per beam or m for n Ka /IF m-for-n IF/BB m-for-n - Broadcasting BB/IF m-for-n IF/Ka m-for-n Redundancy Ring LCAMP + TWTA + HPI FGU FGU FGU FGU

7 Is Space a conservative industry? How new technologies are inserted into space products? How long will it take to see MEMS flying on-board a commercial satellite? OK, ok. Let s look at history. How was it with MMICs?

8 1987/88 : First MMICs designed for ALCATEL SPACE 1989 : First MMICs flown by COMSAT on-board ITALSAT F-1 12 GHz Amplifier

9 1987, Ann Arbor, Michigan Fabricated at University Of Michigan on GaAlAs/GaAs HEMT process First MMIC ever?

10 MMIC Globalstar Decision for FM FM delivery Preliminary phase : foundries open in US and Europe. Space industry start to get informed Lauch of AMOS, Arabsat 2, Telecom 2D Knowledge phase : Foundries selection. Constitute a team of designers, first MMIC designs Pre-Indus phase : Intensive work on packaging. Start evaluation and qualification works Industrialization phase : Developement of first FM equipment with MM MEMS We are here! FM MEMS? Can we make it faster?

11 You could start thinking of fabricating satellite subsystems like we make CMOS chips for labtops 1-kg-class satellite could be mass-produced you could put up a constellation of hundreds of these little guys in low Earth-orbit But There is a long way to go before all satellite functions can be duplicated by MEMS-based components Some MEMS are flying in experimental satellite : (from IEEE Spectrum, July 2001)

12 General trends for satellites Outline MEMS to get in as MMIC did. Always the same story? Study case 1 : RF MEMS within microwave equipment, MIPA project Study case 2 : RF MEMS for reconfigurable antenna, ARRESAT project Issues to be solved Conclusions

13 Space prototype radiating element design RF MEMS reliability µswitch mechanical optimization RF MEMS packaging RF MEMS reliability Prime contractor End-user : V band active feed RF MEMS technology provider End-user : ACC automotive radar sensor

14 Space application objective Use of microtechnologies in V band satellite equipments µswitch (SPDT) for redundancy Focal array integrated antenna High G/T ratio antenna front-end for satellite link in Q/V band Low NF of the LNA chain Low insertion losses before the first LNA Use of ultra low loss RF MEMS switch Advanced packaging concepts (0 level and 1st level) Active feed concept Integration of radiating elements and LNA chain Compliance with space requirements (Life time, reliability )

15 Antenna Gateways-users mission links in Ka & Q-band Ka-band Users beams (64 spots) Forward link In Ka band (20 GHz ) Return link In Ka band (30 GHz ) In Q band (40 GHz ) Return link Forward link In V band (50 GHz ) Q-band Gateways beams (9 big cities) USERS GATEWAYS

16 Gregorian antenna Antenna coverage iso-contours 1 m Elevation angle (degree) Feed horns to be replaced by MIPA prototype Azimuth angle (degree)

17 Active receiving feed key elements Anticipated source efficiency : 66% Patch array Stacked patch array Insertion loss target : 1.3 db (Max 2dB) between A&B SPST first results at 50GHz : Insertion loss=0.4 db Isolation 20 db Slot transition LNA 1r Filter 2r LNA 2n Filter 1 SPDT1 SPDT2 LNA 1n Filter 2n LNA 2n EMS and LNA interconnect substrate A B C

18 RF MEMS packaging approach CPW Sealing cap h BCB RF-MEMS substrate x2mm 2 Return Loss (db) x3mm Frequency (GHz) 0-level test assemblies on a Bosch RF-MEMS wafer

19 RF MEMS reliability Three topics are reviewed and discussed: Technology and design aspects of MEMS switches functioning, resistive and capacitive, shunt and series, types of actuation, materials and processes Failure mechanisms 4 main failure modes expected: stiction, creep, dielectric charging, electromigration Aging tests overview tests + relevance for MIPA Creep, fatigue Stiction Deformation Degradation & Charging of the dielectric

20 RF MEMS Reliability Stress, stress gradient and creep Study the stress relaxation vs temperature Study the shape of beams and compare with simulation Nanoindentation Profilometry Temperature [C] Stress [MPa] Creep study Time [min] Simulation Nanoindentation

21 General trends for satellites Outline MEMS to get in as MMIC did. Always the same story? Study case 1 : RF MEMS within microwave equipment, MIPA project Study case 2 : RF MEMS for reconfigurable antenna, ARRESAT project Issues to be solved Conclusions

22 The reflectarray : An emergent technology of reconfigurable antennas at moderate cost Reflectarray antenna combines Reflector advantages : Array advantages : Pattern control possible through a modification of the element phase response Benefits from the breakthrough of the MEMS technology low losses and low consumption (as compared to diodes) Switch Integration of phase shifters inside the array elements High integration of the antenna Control circuit Reflectorplane High scale microelectronic process (MEMS on Wafer) Objective : Low manufacturing costs Reflect-array principle Via hole Radiatingelement Multilayercircuit

23 Objectives of the ARRESAT Project Demonstrate the feasibility of broadband and low losses controllable phase-shifting elements Beam forming performances assessment of the reflectarray Application to a Multimedia Ka band mission in circular polarization Design and breadboarding of controllable radiating phase shift elements using MEMS switches Design and Measurement of a RF Passive Reflectarray Context Collaborative effort between Thalès and Alcatel Space

24 Definition of the phase shift element Main requirements Low losses, Circular polarisation, Broadband Compatible with MEMS switch design Commutation of Dipole elements Proposed solution The phase is controlled by the orientation of the resonating dipole Isolation between elements achieved with via holes thin metallic grid Commandes BF Trous métallisés: passage des commandes λ/4 Plan de masse Multicouche BF Cellule mono-pola

25 Implementation Dipoles and MEMS on glass substrate Report of the hexagonal component on a PTFE substrate 13,04 mm 6,520 mm Use of a Pyrex protection 11,3 mm DC command routed through via holes Glass Substrate 0,5 mm resistive lines on Metclad and Glass 1,9mm 500 µm 300 µm Dipole elements MEMS (Con/Coff=100) signal entrée membrane 150 µm 100 µm liaisons résistives vers trous cellules hexa. Résistance Substrat Metclad 8.5 mm Ground Plane 110 µm signal sortie

26

27 Details of the 12 switches/cell

28 Unpackaged cell Glass lid Currently under measurement Packaged cell

29 Reflectarray Antenna

30 Ka Band Concept Validation More than 300 phase shift elements Radiation of a on axis directive beam The Phase shift distribution law over the aperture is achieved by rotating the phase shift elements

31 Radiation Pattern Very satisfactory Performances Radiating Efficiency > 60 % over more than 1,5 GHz Co/cross polarization > 35 db over 2 GHz (10%) better than the cross polarization of the feed Non coherent radiation of the cross polarization component

32 General trends for satellites Outline MEMS to get in as MMIC did. Always the same story? Study case 1 : RF MEMS within microwave equipment, MIPA project Study case 2 : RF MEMS for reconfigurable antenna, ARRESAT project Issues to be solved Conclusions

33 Issues to be solved : 1, Reliability ESD sensitive devices MEMS needs controlled environment to be reliable humidity sensitive inert gas preferred (damping effect needed) temperature sensitive sensitive to dielectric charging (capacitive switch) Mechanical protection

34 Issues to be solved : 1, Reliability A vast domain Very dependent on the application : need billions of switching like in a Phase-Shifter? need a single switching after 15 years in a redundancy ring? need no alteration in a cavity performance? What are the degradation mechanisms? Are there any accelerated tests possible? A mine of good subjects for future researches

35 Issues to be solved : 2, Packaging Space is an harsh environment Temperature cycling (-40/+65 C operat. Temperature) Radiation (Esp. cumulated dose) Hermetic sealing=leak rate specified according to MIL 883 STD (hybrid are in vacuum after few years in space ) Life time >15 years Which Packaging? Chip level Module level

36 Issues to be solved : 2, Packaging Problem : There is a trend towards non-hermetic for active (MMIC, ) functions MEMS needs some level of hermeticity Glob-Top for MMIC LNAs Transitions are best avoided! Keep insertion losses at minimum!!!

37 Issues to be solved : 2, Packaging A solution : A System-In-Package with a micro-machined package but : A lot of questions arises for an industrial production: who s doing what, necessary investment for automatic process, which partnership between die manufacturer, module assembler, Not easy to solve for small production as for space applications... Courtesy of

38 Conclusion MEMS are very promising technology (ies) for future Space applications It could be the biggest technical revolution since the beginning of the commercial explosion of satellite business One more thing is needed and will definitively come soon : a new type of designer : innovative ; mastering both electrical and mechanical knowledge then, maybe, one day in not-so-far future we will see :

39

40 Need for robust RF MEMS for future space applications Acknowledgements t To all our partners from the different consortia on all our projects