13.07.2015 Views

NIST Technical Note 1337: Characterization of Clocks and Oscillators

NIST Technical Note 1337: Characterization of Clocks and Oscillators

NIST Technical Note 1337: Characterization of Clocks and Oscillators

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

E. APPENDIX - <strong>Note</strong>s <strong>and</strong> ErrataThe notes listed below are included to bring attention to notation <strong>and</strong> other problems. The pagenumber provides reference to the location <strong>of</strong> the problem or comment.1. page TN-4AM noise is especially important in the measurement <strong>of</strong> the (residual) phase noise addedby amplifiers <strong>and</strong> other signal h<strong>and</strong>ling components. Often the driving source for themeasurements is a frequency synthesizer that has phase <strong>and</strong>/or amplitude noise that iscomparable or larger than the added noise <strong>of</strong> the component under test. Most measurementsystems are configured so that the phase noise <strong>of</strong> the source cancels out to a largedegree (at high Fourier frequencies decorrelation effects limit the cancellation). In suchmeasurement systems, the AM to PM conversion factor <strong>and</strong> the AM noise <strong>of</strong> the sourcemay then set the noise floor. Discussion <strong>of</strong> the effect <strong>of</strong> AM noise <strong>and</strong> AM to PMconversion factors on the accuracy <strong>and</strong> precision <strong>of</strong> phase noise measurements is found in"Residual Phase Noise Measurements <strong>of</strong> vhf, uhf, <strong>and</strong> Microwave Components" by G. K.Montress, T. E. Parker, <strong>and</strong> M. J. Loboda Proc. 43rd Annual Symposium on FrequencyControl, pp. 349-359 (1989) <strong>and</strong> in "Accuracy Model for Phase Noise Measurements by F.L. Walls, C M. Felton, <strong>and</strong> A. J. D. Clements, 21st Annual Precision Time <strong>and</strong> TimeInterval Meeting (1989). The notation in these papers as well as that in other parts <strong>of</strong> theliterature differs from that given below. Our notation below is drawn from a modest level<strong>of</strong> consensus among individuals responsible for setting st<strong>and</strong>ards. We expect that it willgradually be adopted within st<strong>and</strong>ards committees. The following comments are directedspecificalJy at the specification <strong>of</strong> noise performance.The total power spectral density in a signal can be approximated by exp<strong>and</strong>ing eq 12-5 <strong>of</strong>paper B.2 (by Stein) <strong>and</strong> extending it to include the spectral density <strong>of</strong> relative amplitudefluctuations, Sa(f} The double-sideb<strong>and</strong> density written in single-sideb<strong>and</strong> form is given bySv(j)

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!