LeCroy SDA - Pj Breakdown; Alternate Jitter Breakdown Methods (Option ASDA-J Only); Effective Jitter

SDA Operator’s Manual
SDA-OM-E Rev H 333
DCD Duty Cycle Distortion is the mean difference between the width of
positive going pulses (low to high to low) and negative going pulses (high to
low to high) measured over all pulses in the acquired waveform. The widths
are measured at the same amplitude as specified for TIE (i.e., not necessarily
at 50% of the signal amplitude). This measurement is a component of DDJ
and included in the DDj value.
DDj The peak-to-peak jitter caused by systematic effects related to the
sequence of data transitions.
Pj Breakdown
The Pj Breakdown tab reveals a table of components of periodic jitter. This table lists the peak-to-
peak amplitude and rate (frequency) of each Pj component. The components are listed from
largest to smallest. The Pj readout below the grid on the display is the complex sum of the
components listed in this table.
Alternate Jitter Breakdown Methods (option ASDA-J only)
The ASDA-J option adds two additional jitter breakdown methods. These methods are termed
Effective and MJSQ and are selected in the Jitter Calc. Method control when the instrument
has the ASDA-J option present. The Effective and MJSQ methods provide alternate ways of
determining the random and deterministic jitter but do not include the breakdown of deterministic
jitter into periodic and data dependent parts. When either of these modes is selected, the jitter
breakdown button the jitter menu becomes grayed (unavailable) and only the basic jitter display is
shown. The DDj plot can still be viewed in this mode; however, the information from this plot is
not used in the computation of Dj.
Effective Jitter
The effective jitter mode is entered when Effective is selected in the Jitter Calc Method control.
Effective jitter is determined from the measured total jitter by evaluating the total jitter at several
bit error rate values and solving Tj = Tj
(sigma=1)
(BER)*Rje + Dje. The term Tj
(sigma=1)
is the total jitter
of a Gauss Ian (normal) distribution of jitter with a standard deviation of 1 second. The two
unknowns in this equation (Rje and Dje) are found by solving for several Tj values at BER levels
below 10
-10
. The jitter breakdown is the best-fit to the bathtub curve for very low BER values, but
does not take into account the jitter contribution at the very top of the bathtub curve. The figure
below shows the flow of the effective jitter measurement.