UFFC | History | A Historical Review of Atomic Frequency Standards Used in Space Systems

Proceedings of the 1996 IEEE International Frequency Control Symposium (pp. 24 – 32)

© 1996 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.

Natarajan D. Bhaskar
The Aerospace Corporation, P.O. Box 92957
Los Angeles CA 90009

Joseph White
US Naval Research Lab
Washington DC 20375-5320

Leo A. Mallette
Hughes Space and Communications
Los Angeles CA 90009

Thomas A. McClelland
Frequency Electronics, Inc.
Mitchel Field NY 11553

Capt. James Hardy
USAF Space and Missile Systems Center
Los Angeles AFB, CA 90009



The remarkable navigational accuracy currently enjoyed by the users of satellite navigation systems is in major part due to the excellent performance of the on-board atomic frequency standards (AFSs). Since the laboratory demonstration of a Cesium AFS in the 50’s the performance and reliability of AFSs have significantly improved. Currently Cesium (Cs) and Rubidium (Rb) – AFSs are on-board many satellite systems–Navigational Satellites GPS (Global Positioning System-USA) and GLONASS (Global Navigation Satellite System-Russia) and the Communication Satellite MILSTAR (Military Strategic and Tactical Relay-USA). To test the predictions of General Theory of Relativity a hydrogen maser clock was flown in space. In all, we estimate that the total number of space-borne AFSs is no more than several hundred–by all accounts a relatively small sample size for any accurate reliability studies. The manufacturing technology of space qualified AFSs has vastly improved in the last three decades and the significant improvement in performance is primarily due to the maturity of the electronics industry. In this paper we present a historical review of the AFSs used in space systems. We will briefly review the uniquerequirements for space qualification for frequency standards and outline the performance characteristics of different AFSs which are presently on-board various satellite systems. We also present a brief discussion of the advanced AFSs for potential future space applications.



Since the dawn of civilization man has used a wide range of periodic phenomenon for time keeping–from the apparent periodic motion of heavenly bodies (sidereal clocks) to the periodic internal motion of electrons in atoms (atomic clocks). In this long development, the advent of the Quartz Crystal Oscillators (XOs) is a very important milestone [1]. The “Information Revolution” would simply not have occurred without the ubiquitous Crystal Oscillators. Historically, man’s need for accurate time was driven by the demands of navigation in vast open seas which are devoid of any landmarks. Celestial navigation required accurate portable clocks. With the development of radios, another class of navigation aids–radio beacons e.g., Loran, was born. Yet another technology–artificial satellites–made possible the more precise, line-of-sight radio navigation signals. The accuracy of radionavigation is critically dependent on the synchronization and timing accuracy of the transmitter and the receiver of the radio signals. The first satellite navigation was realized in the 1960’s, with the US Navy’s Navigation Satellite System known as TRANSIT. This was developed by the Johns Hopkins Applied Physics Laboratory. The TRANSIT satellites carried on-board Quartz Crystal oscillators (XOs) for stable and precise frequency generation. This system was based on the Doppler Shift of the received signal. TRANSIT is still operational though no new satellites will be added to the system.

In 1974, another US Navy Satellite system known as TIMATION extended the ‘state-of-the-art’ in satellite navigation by orbiting very precise AFSs (atomic clocks). These satellites were developed at the US Naval Research Laboratory (NRL), to provide accurate navigation information by transmitting ranging signals using a technique called side-tone ranging. The TIMATION concept required that a constellation of satellites be tightly synchronized to generate precise navigation reference signals. The first developmental satellites (TIMATION 1 and 2) used high performance XOs. The on-orbit performance of these XOs was exceptionally good (the typical accumulated time error after one day was on the order of 1 microsecond or about 300 meters in range error). Naturally, keeping the on-board clock synchronized to reduce the clock error to a few nanoseconds (range error to a few meters) would have required very frequent clock corrections to the satellite to be sent from the ground control stations. The superior frequency stability of the AFSs offered two distinct advantages–significantly improve the predictions of satellite orbit determination (ephemerides) by providing stable, low noise signals for ranging, and also eventually extend the time required between control segment updates to orbiting satellites. This made the satellite navigation a practical system to operate. TIMATION 3 (later National Technology Satellite–NTS-1) was the first satellite to carry-on-board Rb AFSs (Rubidium AFS). This pioneering work provided the stimulus for developing reliable AFSs for space applications.



Space-borne AFSs must maintain a high level of performance and stability throughout the mission duration in the harsh environment of space. The mechanical design must be such that the AFSs withstand the enormous shocks and vibrations of the launch. The extremes of temperature that may be encountered in space requires that the thermal and mechanical designs be such that the AFSs maintain excellent frequency stability over a broad range of temperatures. This requirement is expressed in terms of the Temperature Coefficient (TC) of the AFS. Typically, the magnitude of change in output frequency of a Cs AFS (Cesium AFS) should be no larger than 2×10-13/° C when the environmental temperature is varied between -10°C and +40°C. This performance should be achievable in the high vacuum in space (better than 10-5 Torr). Another requirement is the radiation sensitivity. Radiation is a major factor in the design of space AFSs, particularly in military applications. In outer space the AFSs will be continuously exposed to energetic particles and secondary radiation. Not only is the XO (Crystal Oscillator) affected by total dose and bursts but the servo electronics in the AFS is also subject to upset, latching, and DC offset effects due to radiation. Some key components in the AFSs are radiation hardened to significantly reduce the impact. Considerable radiation shielding is also added to further minimize the impact of the space radiation. In spite of this, during periods of intense solar activity, AFSs have clearly exhibited discernible sensitivity to radiation [2]. Another stringent requirement for space standards is the electromagnetic interference (EMI). Filter boxes, EMI gaskets, feedthrough filter capacitors and semi-rigid shielded coaxial cables are used to reduce the EMI effects. Another important space requirement is the magnetic sensitivity. Typically, the magnetic coefficient should be no greater than 10-12/Gauss. Several layers of m -metal shields are used to achieve this.

High reliability of AFSs is of paramount importance for space applications. Typically the time and frequency reference subsystems in satellites should exhibit a composite reliability of better than 0.9996 for a 7.5 to 10 year mission life. To achieve this reliability satellite systems fly multiply redundant AFSs. For example, with a unit reliability of 0.80, 4-fold redundancy can increase the subsystem reliability to 0.9996 or better. The reliability of individual standards are calculated using MIL-HNBK-217E. Such estimates must be viewed with great caution. Trustworthy estimates can only be obtained by operating a significant sample size of AFSs for extended periods of time. Consequently placing several AFSs in long-term life tests in a space-like environment is critical to establishing the true reliability and for ensuring mission success.



The launch of Sputnik in 1957 ushered the “Space Age.” The time-line of various space programs that have flown AFSs is shown in Fig. 1. The cumulative number of space-borne AFSs is shown in Fig. 2. To date about 300 AFSs have been flown in various satellite systems, a large fraction of which is in the two major Navigation Systems–GPS and GLONASS. Satellite descriptions are presented in Appendix-A.





In 1961 Space Technology Laboratories (now TRW) launched 3 Rubidium AFSs on Atlas rockets for demonstrating the feasibility of missile guidance using AFSs. These were sub-orbital flights. One of the three clocks could not be flight tested due to launch failure. The remaining two AFSs apparently performed well although the Atlas missiles had relatively short flight times of several minutes. No data are available on the performance of these AFS, but apparently there were concerns with high Rb lamp failures [3,4].





In the late 60’s General Radio developed a Rubidium AFS for NASA. It was the first AFS developed and qualified for space application. However, this AFS was never flown [5]. Navigation Technology Satellite (NTS-1) launched in 1974, a precursor to the GPS, was the first satellite to fly Rb AFSs. These were commercial Efratom standards (Datum, Inc. Efratom Time and Frequency Products) and were modified by NRL for space operation. NTS-1 carried 2 Rb AFSs. This pioneering work laid the foundations for the development of space qualified AFS. NTS-2, which was launched in 1977 was the first satellite to carry a pair of Cesium AFS on-board. This AFS was specifically designed for space application and was built by Frequency and Time Systems (FTS) in collaboration with NRL. This design was strongly influenced by the on-orbit performance of the AFSs in NTS-1. The NTS-1 & 2 clearly established the excellent improvement in frequency stability of AFSs over the XOs and also demonstrated the feasibility of producing AFSs for space applications. Each AFS in NTS-2 performed well in space for about eighteen months.



The navigational requirements of GPS could only be met with AFSs on satellites. The first 3 GPS Block I satellites (also known as Navigation Development Satellites) were each launched with 3 Rb AFSs on board. These Rubidium AFSs consisted of physics packages from Efratom, Voltage Controlled Crystal Oscillators (VCXO) from Frequency Electronics, Inc. (FEI) and the integrating electronics was designed and fabricated by Rockwell Autonetics. The mean life of most of these AFSs was low (typically less than a year). The numerous early failures of these Rb AFSs raised a lot of concern regarding the reliability of these AFSs. The on-orbit data and the ground test data were systematically analyzed to find the root causes of the on-orbit failures. This led to the identification of the principal wear-out mechanisms that were limiting the operational life of these AFSs [6,7,8]. In the Rubidium AFSs the principal wear-out mechanism was identified to be the consumption of metallic Rb by the glass walls of the Rb-resonance lamp that is used to optically pump the Rb-vapor in the resonance cell [9]. This led to the establishment of optimum fill that is needed for the required life, but not far in excess which is known to create instability in the discharge. Generic causes of early failures were identified to be–depletion of Rubidium in the lamp, transformer malfunction and magnetic field tuning hits. It should be noted that one of the Rb AFSs of this vintage lasted for over 12 years [10,11]. This apparent low reliability of Rb AFSs together with the better overall performance of Cesium AFSs resulted in the remaining GPS Block I satellites to each carry 4 AFSs (3 Rubidium AFSs and 1 Cesium AFS). The Cesium AFSs were built by FTS. The first GPS satellite to fly a Cesium AFS was NAVSTAR-4 which was an embarrassing failure. This AFS performed well during ground testing and turned-on normally on orbit. However, when the satellite came into view 12 hours later, the clock was found non-operational. This failure was attributed to the power supply in the AFS. Depletion of Cs in the source was one of the major causes of early failures in some of the Cs AFSs. In general root causes of these failures were identified. This resulted in numerous design improvements. One of the significant design modifications was the temperature controller that was added to the thermal base plate of the Rubidium AFSs (first flown on NAVSTAR 7), to maintain the AFS at essentially constant temperature in spite of the fluctuations in the satellite temperature.

Each GPS Block II/IIA satellites carry on-board 2 Rubidium AFSs and 2 Cesium AFSs. These standards demonstrate excellent on-orbit performance [12,13]. However, more recently, concerns regarding their reliability have been raised. Apparently a significant number of on-board AFSs exhibited performance deficiencies–significant enough to be detected in the User Range Error (URE). Consequently a number of on-board AFSs had to be switched off, and the redundant AFSs were turned on. Since GPS is a fully operational constellation, detailed investigations of the anomalies are not carried out. Van Melle has calculated the average operating life of the GPS AFS to be about a year for Rubidium AFSs and about 2 years for Cesium AFSs [10,11]. This observation raised a lot of concern regarding the reliability of space-borne AFSs. Van Melle [10,11] believes that many of the “switch-outs” are not ‘hard’ failures. Some of these AFSs will undoubtedly be used again. Furthermore, it is unclear if some of the observed performance anomalies were not due to external factors occurring in the satellites. This observation apparent low reliability should not cast a pall over the reliability of AFSs as a whole. Some of it may be process related, unique to this particular build. Caution must be exercised in drawing any firm conclusion from the limited data. A more fundamental question is whether our models used for estimating the reliability need significant improvements. This issue deserves full attention.



The Russian Global Navigation Satellite System (GLONASS) consists of a constellation of 24 satellites in three orbital planes. This system is operated by the Russian Military Space Forces. The pre-operational phase (Block I: 1982 to 1985) were launched with two “BERYL” Rubidium AFSs per satellite. The Block lIa, IIb and IIc satellites (1985 to present) carried three on-board “GEM” Cesium AFSs. These AFSs are reported to demonstrate excellent on-orbit performance [14,15]. More recently, a new generation of Cesium AFSs (“MALAKHITE”) with improved performance and reliability (5 years) will be deployed on the GLONASS-M satellites. The Rb and Cs AFSs were built by the Russian Institute of Radionavigation and Time (RIRT, formerly: Leningrad Scientific Research Radiotechnical Institute (LSRRI). A passive hydrogen maser is also expected to be flown on GLONASS in the near future.



The Russian PARUS (or Tsikada-M) is a navigational spacecraft in the CICADA series for locating ships on the ocean. The satellite system consists of a constellation of six satellites with an inclination of 83 degrees and a period of 105 minutes. PARUS is the military equivalent of the commercial Tsikada satellite navigation system. The satellites in this system have been launched from 1974 (Cosmos 700) to 1996 (Cosmos 2327). It’s believed that the PARUS satellites are equipped with Rubidium AFSs [16]. We have not been able to obtain much information regarding these AFSs.



A navigation experiment (NAVEX) as a part of Spacelab Mission Dl was launched on shuttle flight STS-61A It consisted of experiments in navigation, time transfer, one way ranging, and relativistic effects and was implemented by the German Spacelab Mission (DFVLR). This package was launched with a FTS 4000 Cs AFS (S/N 168) and an Efratom FRK-H Rubidium AFS (S/N 955). The flight extended from 30 October to 06 November 1985. Both AFS were off at launch and were commanded on 12 hours after launch. These were the first AFSs to be returned from orbit, thus providing one of the most accurate measurement results on the relativistic effect of moving clocks [17,18,19].



Milstar is a US military communication satellite system, which, after full deployment will consist of a constellation of 4 geosynchronous satellites, interconnected by cross-links. Presently, two satellites have been deployed. Milstar, popularly called the “switchboard in the sky”, is a ‘Time-of-Day’ digital communication system which transmits and receives encrypted and encoded messages. It carries on-board several Rubidium AFSs. The constellation is required to operate autonomously and also survive the environmental threats of natural and man-made nuclear radiation. These requirements which are only met by AFSs and not by Quartz Crystal Oscillators (XOs). These Rubidium AFSs are manufactured by FEI. MILSTAR Rubidium AFSs have been tested rigorously, far and above the normal acceptance tests required for space-hardware. Their long-term ground performance has been good and shows a promise of high reliability. The limited available on-orbit performance data points to good space performance as well [20,21,22].

Table I shows the characteristics of the space qualified Rubidium AFSs. The typical Rb AFS stability numbers are calculated after removing the linear drift in Table II, characteristics of the space qualified Cesium AFSs are presented. Their on-orbit performance compare very favorably with the ground test data. Comparison of the two Tables clearly show that the Rubidium AFSs are substantially lighter and consume less power than the Cesium AFSs. However, the Rubidium AFS exhibit aging which has to be taken into account for accurate time error estimation. Rubidium AFSs consistently show negative aging and is found to be highly modelable. Although the complete origin of this aging is not fully understood, it’s deterministic nature makes it highly acceptable for very demanding applications. Early GPS IIR satellites are expected to fly only Rubidium AFSs. These standards have demonstrated excellent performance in the ground tests [23].

Table I. Characteristics of Space Qualified Rubidium Atomic Frequency Standards.



Space System















First Launch






Power (watts)






Mass (kg)














Table II: Characteristics of space qualified Cesium AFSs.









IIa/IIb/IIc & M












First Launch








Power (watts)








Mass (kg)


















Among the various AFSs H-maser is uniquely suited for testing the fundamental physical laws. However, designing a space qualified hydrogen maser is a challenging task. Dr. Vessot and colleagues of Harvard Smithsonian Astrophysical Laboratory have successfully met this challenge in their pioneering experiments conducted in 1976, known as the Gravity Probe-A (GP-A) experiment. In this experiment the gravitational red shift resulting from the difference in the Newtonian gravitation potential between the space clock and the ground clock is measured. A hydrogen maser was launched in sub-orbital trajectory to an altitude of 10,000 miles by use of a Scout-D rocket system. The flight duration was about 2 hours during which time the space H-maser clock’s frequency variation was measured relative to the frequency of a ground maser clock. The experimental goal was to measure the red shift to 50 ppm which could only be achieved by comparing the frequency of the space maser clock relative to that of the ground maser clock to an accuracy of 2×10-14 over time intervals of 100 seconds. Many large systematic frequency shifts had to be either eliminated or accounted for to reliably measure the gravitational red shift. To obtain the desired stability particular emphasis was placed on the thermal and mechanical design of the maser. Thermal sensitivity of 2 x 10-14/° C was reported for the space maser. This experiment tested Einstein’s prediction to a precision of better than 50 parts per million. Several studies are in progress to design tests of relativistic gravitation using maser clocks in a space probe in a polar orbit approaching within 4 solar radii of the sun’s center. This flight is intended to reveal the behavior of the second order in the red shift [24,25].

A joint NASA/SAO (Smithsonian Astrophysical Observatory) technology experiment to demonstrate the performance of hydrogen masers in space is now in progress. In the planned SAO/NASA Maser experiment, the new space maser will be flown on a European Space Agency (ESA) EURECA Spacecraft, and frequency measurements will be made by time transfer using existing laser ranging stations. This maser is based on the GP-A design. With a laser timing precision of 20 ps short-term frequency measurements with a precision of approximately 4×10-15 is anticipated. This satellite will be deployed from the space shuttle and boosted to a higher orbit. It is expected to remain in operation from 6 to 8 months and then retrieved by the shuttle [24].

NRL in the 80’s, led the development of a space qualified H-maser for GPS. The goal of this effort was to produce a frequency standard with very good long-term stability for autonomous operation of GPS. This program included development of dielectrically loaded maser cavities for reduced size and weight and digital servo systems to reduce the effects of cavity frequency errors on the maser. Masers for space applications were also developed at the Hughes Research Laboratories (HRL) in the mid 80’s.- However, none of these masers were ever flown [26,27]. Table III shows the characteristics of the various space qualified masers.

Table III. Characteristics of Space Qualified Hydrogen Maser AFSs.



Not Flown




Hughes Research Labs



Neutchatel Obs.

First Launch

1987 (built)



Power (watts)



Mass (kg)










To a large measure the remarkable improvement in the performance of the AFSs over the last two decades is in large part due to the excellent progress in the electronics-both performance and packaging. Many manufacturers have introduced sub-miniature Rubidium AFSs into the market whose absolute performance and performance per dollar is very impressive [28,29]. This miniaturization will strongly impact the development of Space Rubidium AFSs. Another area where there is considerable promise is the development of “smart clocks.” Conceptually this would require additional sensors and servos to sense and correct environmental effects and other variations–microwave power level, magnetic field, and electron multiplier gain, are just a few. This could nearly eliminate the effects of temperature and magnetic field changes resulting in excellent long-term stability. For the immediate future the traditional Rubidium AFSs and Cesium AFSs of the present design physics packages is expected to be used, albeit with significantly improved performance and reliability. However, there are a number of research and development efforts in the area of new types of standards (i) optically pumped Cs AFSs: Optical pumping, state selection and detection will be done using solid state diode lasers. This would eliminate the use of bulky “A” and “B” magnets in the physics package. This will result in lighter beam tubes, improved efficiency and better performance [30]. However, laser diode frequency must be “locked.” Temperature of the diode laser has to be controlled to a very high degree. (ii) Diode laser pumped Rubidium AFSs: Replacing the Rb discharge lamp with a diode laser tuned to the correct transition frequency offers substantial simplification in the design of the physics package in the Rubidium AFSs. The diode laser can optically pump with nearly 100% efficiency thereby substantially improving the SNR (Signal-to-noise ratio) of the microwave clock signal. This would eliminate the need for the Rb 85 filter cell and also the parasitic effects of the radio frequency used in the discharge lamp. Theoretically, a factor of 100 improvement over the existing design should be realizable. Ongoing experiments show that the light shift effects of the diode laser are very significant. Diode laser frequency has to be “locked”. Very encouraging results have started to appear in this area [31,32,33].



Reliable high performance AFSs are the backbone of many Space Programs. AFSs are no longer a mere laboratory curiosity. Rubidium and Cesium AFSs are the two that have been extensively used–over 300 of them have been flown thus far. In the last 20 years performance and reliability of these standards have vastly improved. Commercial-of the Shelf AFSs (COTS) show considerable promise for space applications. Considerable work remains to be done to demonstrate their viability. Laser diode pumped clocks also show promise-however, challenging task of space qualification and demonstrating reliability still remains to be done. For space applications requiring excellent very long-term stability mercury ion standards also show promise [34]. This type of standard has not yet been flown.



The authors would like to thank Dr. T. English (Datum, Inc., Efratom), Dr. V. Reinhardt (Hughes Space and Communications), Mr. W. Riley (EG&G), Mr. M. van Melle (Rockwell), Dr. R. Vessot (Smithsonian), Mr. H. Winthroub (Aerospace), Dr. P. Daly (University of Leeds-U.K.), Dr. H. Nau (Germany), Dr. R. Schmid (Germany) and Dr. H Hellwig (Air Force Office of Scientific Research) for their valuable suggestions in the preparation of this manuscript.



1. E. Bottom, Introduction to Quartz Crvstal Unit Design, Van Nostrand-Reinhold, New York, 1982.

2. P.Tally (The Aerospcace Corporation), private communication, 1989.

3. H. l. Winthrob (The Aerospcace Corporation), private communication, 1996.

4. J. M. Andres, D. J. Farmer, and G. T. Inouye, “Design Studies for a Rubidium Gas Cell Frequency Standard,” IRE Transactions on Military Electronics, October 1959, pp.178-183.

5. W. Riley (EG&G), private communication, 1996

6. R.P. Frueholz (The Aerospace Corporation), private communications, 1996.

7. R. P. Frueholz, M. Fun-Fogle, H. U Eckert, C. H. Volk, and P. F. Jones, “Lamp Reliability Studies for Improved Satellite Rubidium Frequency Standard,” Proceedings of the 13th Annual Precise Time and Time Interval (PTTI) Applications and Planning Meeting. Dec.1-3, 1981, Washington, DC, NASA Conference Publication 2220, pp. 767-788.

8. C. H. Volk, R. P. Frueholz, T. C. English, T. J. Lynch, and W. J. Riley, “Lifetime and Reliability of Rubidium Discharge Lamps for Use in Atomic Frequency Standards,” Proceedings of the 38th Symposium on Freqeuncy Control, pp. 387-400, 1984.

9. Vanier and C. Audoin, The Ouantum Physics of Atomic Frequency Standards, Adam Hilger, Philadelphia, 1989.

10. M. Van Melle (Rockwell Space and Operations Center), private communications, 1996.

11. M. Van Melle, “Cesium and Rubidium Frequency Standards Status and Performance on the GPS Program,” 27th Annual Precise Time and Time Interval (PTTI) Applications and Planning Meeting, NASA Conference Publication 3334 Conference, Nov. 29-Dec. I, 1995, pp. 167- 180.

12. H. Hellwig and M. Levine, “Cesium Clocks Deployed in the Global Positioning System: Design and Performance Data,” Proceedings of the 38th Annual Symposium on Frequency Control. pp 458-463, 1984.

13. T. B. McCaskill, and J. A. Buisson, “Performance of Global Positioning System (GPS) On-Orbit Navstar Clocks,” Proceedings of the 1995 Frequency Control Svmposium. May 31-June 2. 1995, San Francisco.

14. Y.G. Gouzhva, A.G. Gevorkyan, and V.V. Korniyenko, “Atomic Frequency Standards for Satellite Radionavigation Systems,” Proceedings of the 45th Symposium on Frequency Control, pp. 591-593, 1991.

15. Y.G. Gouzhva, A.G. Gevorkyan, A.B. Bassevich, P.P.Bogdanov, and A.Y. Tyulyakov, “Comparative Analvsis of Parameters of GLONASS Space borne30 Frequency Standards when used Onboard and on Service Life Tests,” Proceedings of the IEEE 47th Annual Symposium on Frequency Control, pp. 65-70, 1993.

16. P. Daly (University of Leeds), private communications, 1996.

17. H. Hellwig (Air Force Office of Scientific Research), private communications, 1996.

18. R. Schmid (Germany), private communications, 1996.

19. German Aerospace and Research Establishment (DLR), “Navigational Experiment NAVEX on Spacelab Mission D1,” Final Report. October 1989.

20. T. A. McClelland, I. Pascaru, and M. Meirs, “Development of a Rubidium Frequency Standard for the Milstar Satellite System,” Proceedings of the 41st Annual Symposium on Frequency Control, pp. 66-74.

21. N.. D. Bhaskar, L. Mallette, T. A. McClelland, and J. Hardy, “Rubidium Atomic Frequency Standards for the MILSTAR Satellite Payload,” AIAA Space Programs and Technology Conference Proceedings (September 1995): Paper 3542.

22. M. Bloch, J. Ho, T. McClelland, M. Meirs and N. D. Bhaskar, “Performance Data on the Milstar Rubidium and Quartz Frequency Standards–Comparison of Ground Tests in a Simulated Space Environment to Results Obtained on Orbit,” Proceedings of the 1996 IEEE Frequencv Control Symposium.

23. W.Riley, Rubidium Atomic Frequency Standard for GPS Block IIR, ION-GPS-92, September 1992.

24. R. Vessot (Harvard-Smithsonian Center for Astrophysics), private communications, 1996.

25. E. M. Mattison and R. Vessot,” High Precision Time Transfer to Test a Hydrogen Maser on Mir,” Proceedings of the 27th Annual Precise Time and Time Interval (PTTI) Applications and Planning Meeting NASA Conference Publication 3334.

26. H.T.M. Wang, “An Oscillating Compact Hydrogen Maser, “Proceedings of the 34th Annual Symposium on Frequency Control. 1980, p. 369.

27. E.S. Sabisky and H.A Weakliem, “An Operating Development Model Spacecraft Hydrogen Maser.

28. T.A. McClelland. I Pascaru, I. Shtaerman, C. Stone, C. Szekely, J. Zacharski, and N. D. Bhaskar, “Subrniniature Rubidium Frequency Standard: Performance Improvements,” Proceedings of the 1996 IEEE Frequency Control Symposium.

29. R. Frueholz, “The Effects of Ambient Temperature Fluctuations on the Long-Term Frequency Stability on the Miniature Rubidium Atomic Frequency Standard,” Proceedings of the 1996 IEEE Frequency Control Symposium.

30. K. Hisadome and M. Kihara, “Prototype of an Optically Pumped Frequency Standard,” Proc. 45th Annu. Freq. Cont. Symp., 1991, pp 513-520.

31. N. D. Bhaskar, “Potential for Improving the Rubidium Frequency Standard with a Novel Optica Pumping Scheme Using Diode Lasers,” IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control, Vol. 42, pp 15-22, 1995.

32. G. Mileti, J. Q. Deng, F. L. Walls, J. P. Lowe, and R. E. Drullinger, Recent Progress in Laser Pumped Rubidium Gas Cell Frequency Standards,” Proceedings of the 1996 Frequency Control Symposium.

33. J.C. Camparo, “Reducing the Light-Shift in the Diode Laser Pumped Atomic Clock,” Proceedings of the 1996 IEEE Frequency Control Symposium.

34. J.D. Prestage, R. L. Tjoelker, G. J. Dick and L. Maleki, “Space Flyable Hg+ Frequency Standard,” Proceedings of the 48th Annual Svmposium on Frecuency Control, 1994, p. 747.


Description of Satellite Systems


The 24 satellites in the NAVSTAR GPS constellation (completed in March 1994) are in six orbital planes and are operated by the US Department of Defense. The circular 20,200 km orbits have an inclination of 55 degrees (Block I used 65 degrees). The GPS constellation was developed in Blocks.

GPS Block I. The ten Block I satellites were numbered 1 through 11 (7 was a launch failure) in the developmental/system testing phase and were built by Rockwell International in Seal Beach California. These satellites were designed for a 4 year life and were launched from Vandenberg AFB (beginning in February 1978).

GPS Block Il/IIA. The 28 Block Il/IIA satellites were numbered 13 through 40. They weighed 840/930 kg, were built by Rockwell International in Seal Beach California and-are designed for an 7.5 year life. These satellites were launched beginning in February 1989. The Rubidium AFSs were the same as Block I, and the Cesium standards were built by three companies; FEI built two, Kernco built three, and the remainder (a build of 56) were built by FTS.

GPS Block IIR. The 24 Block IIR satellites are being built by Lockheed Martin. These satellites are designed for a 10 year life and will be launched beginning in 1997. First few of these satellites will be carry 3 Rubidium AFSs which are built by EG&G. Subsequent satellites will carry 1 Cs and 2 Rb AFSs.

GPS Block IIF. The Block IIF satellite contract was recently (April 1996) awarded to Rockwell International. Each of these satellites will carry 3 Cs and 1 Rb AFS.



The 24 satellites in the Global Navigation Satellite System (GLONASS) constellation are in three orbital planes and are operated by the Russian Military Space Forces. The circular 19,100 km orbits have an inclination of 64.8 degrees. GLONASS populated two of the three orbital planes in 1990) and populated all three orbital planes in January 1996. The satellites weigh about 1440 kg; a large part of which is the pressure vessel. The pressure vessel is used to keep the electronics at terrestrial temperatures and pressure; therefore, minimizing the effects of the space environment. The system we developed in Blocks.


1 GPS 12 was the qualification vehicle and was not flown.

Block I. The 11 satellites in this pre-operational phase (1982 to 1985) had a design life of 1 year. The first launch of GLONASS I was on Cosmos 1414 on 12 October 1982. These satellites used the “BERYL” Rubidium AFSs.

Block IIa. This Block of satellites was launched starting in December 1985 and had a design life of one year. The GEM frequency standards were introduced with the six satellites in this Block.

Block IIb. The 12 satellites in this Block had a design life of two years and were launched in 1987 and 1988. There were two launch failures, and for this reason, only six of the 12 satellites reached the desired orbit.

Block IIc. This Block of satellites was launched starting in September 1988 and had a design life of three years. The constellation was completed on 18 January 1996.

GLONASS-M. The next generation of GLONASS satellites weigh 1480 kg, have design lives of 5 years and have the improved “MALAKHITE” Cs AFSs with a stability of 1×10-13/day

These AFSs were built by the Russian Institute of Radionavigation and Time (RIRT, formerly: Leningrad Scientific Research Radiotechnical Institute (LSRRI)).



The Milstar system is a multi-service communications system providing the US Army, Navy and the Air Force with command and control of strategic and tactical forces through all levels of conflict. The system consists of three segments: Terminal, Mission Control and Space. The space segment will consist of low data rate, medium data rate and crosslink communications payloads on six geosynchronous satellites. The satellites will be able to operate autonomously and are required to survive the environmental threats of nuclear radiation.