NIST Time Scale Data Archive - 1999 Notes on NIST time scale, primary standards & services

Return to Archive index

2000

2001

2002

2003

2004

Current Year

Primary frequency standards developed and maintained by NIST are used to provide accuracy (rate) input to the BIPM. NIST-7, which has served as the U.S. primary standard since 1994, has been replaced by NIST-F1, a cesium fountain frequency standard. The uncertainty of the new standard is currently 1.3 parts in 10¹⁵.

The AT1 scale is run in real time using data from an ensemble of cesium standards and hydrogen masers. It is a free-running scale whose frequency is maintained as constant as possible by choosing the optimum weight for each clock that contributes to the computation.

UTC(NIST) is generated as an offset from our real-time scale AT1. It is steered in frequency towards UTC using data published by the BIPM in its Circular T. Changes in the steering frequency are usually made at 0000 UTC on the first day of any month, and the change in frequency in any month is limited to ±2 ns/day. The frequency of UTC(NIST) is kept as stable as possible at other times.

UTC is generated at the BIPM using a post-processed time-scale algorithm and is not available in real-time. The parameters that we use to generate UTC(NIST) in real-time are therefore based on an extrapolation of UTC from the most recent data available.

The table lists the parameters that are used to define UTC(NIST) with respect to our real-time scale AT1. To find the value of UTC(NIST) - AT1 at any time T (expressed as a Modified Julian Day, including a fraction if needed), the appropriate equation to use is the one for which the desired T is greater than or equal to the entry in the T₀ column and less than the entry in the last column. The values of x_ls, x, and y for that month are then used in the equation below to find the desired value. The parameters x and y represent the offset in time and in frequency, respectively, between UTC(NIST) and AT1; the parameter x_ls is the number of leap seconds applied to both UTC(NIST) and UTC as specified by the IERS. Leap seconds are not applied to AT1.

UTC(NIST) - AT1 = xls + x + y(T - To) Month xls(s) x (ns) y (ns/day) To (MJD) Valid until 0000 UTC on: (MJD) 1999-12 -32 -199896.5 -41.0 51533 51544 1999-12** -32 -199096.5 -40.0 51513 51533 1999-11 -32 -197896.5 -40.0 51483 51513 1999-10 -32 -196641 -40.5 51452 51483 1999-09 -32 -195426 -40.5 51422 51452 1999-08 -32 -194155 -41.0 51391 51422 1999-07 -32 -192884 -41.0 51360 51391 1999-06 -32 -191654 -41.0 51330 51360 1999-05 -32 -190383 -41.0 51299 51330 1999-04 -32 -189183 -40.5* 51269 51299 1999-03 -32 -187927.5 -40.5 51238 51269 1999-02 -32 -186779.5 -41.0 51210 51238 1999-01 -32 -185493 -41.5 51179 51210 * Provisional Value ** Rate Change in mid-month Bibliography Allan, D.W.; Hellwig, H.; and Glaze, D.J., "An accuracy algorithm for an atomic time scale," Metrologia, Vol.11, No.3, pp. 133-138 (September 1975). Allan, D.W. and Weiss, M.A., Accurate time and frequency transfer during common view of a GPS satellite, Proc. 34th Annual Symposium on Frequency Control, p. 334 (1980). Allan, D.W. and Barnes, J.A., Optimal time and frequency using GPS signals, Proc. 36th Annual Symposium on Frequency Control, p. 378 (1982). Drullinger, R.E.; Glaze, D.J.; Lowe, J.P.; and Shirley, J.H., The NIST optically pumped cesium frequency standard, IEEE Trans. Instrum. Meas., IM-40, 162-164 (1991).