Teledyne LeCroy SDA III-CompleteLinQ - Manual (Page 19)

Operator's Manual
Serial Data Inputs
There are several scenarios for the configuration of the Seria Data Input(s) section.
If you are using a differential probe or if your signal is connected by one coaxial cable, use the 1 Input (or
Diff. Probe) button and select the input source.
If you have a differential signal transmitted on two two coaxial cables or two single-ended probes., use
the Input1-Input2 button and select the input channels used.
NOTE: There is no need to configure a math function to calculate the difference between to inputs. Doing
so adds additional computational steps, and unnecessarily uses RAM.
Lastly, any math function, memory trace, etc. can be used as an input, as well as any input channel.
When in Input1-Input2 mode, and when using traces that may have been the result of other processing
steps, be sure that both traces have the same record length and sample rate.
You can choose to upsample rate in the Serial Data Inputs dialog by a factor of two in order to provide a
higher sample density for analysis. In SDAII this was typically done to facilitate formation of eye diagrams
without gaps for bit rates integrally related to the sampling rate (for example, 20 GS/s is exactly eight
times 2.5 Gb/s), and especially for relatively short acquisitions. This, however, unnecessarily slowed
down the analysis process. When needing to upsample to remove gaps in the eye use the upsample con-
trol in the Eye Diagram dialog.
Crossing Levels
The Crossing level section of the Signal dialog determines the voltage level where the arrival time of each
edge of the signal is measured. The crossing level is set separately for the data and clock (if an external
clock is selected).. The Crossing Level on this dialog is for the data. Setting the crossing level to a value
that is not optimal can result in higher than expected deterministic jitter, since the error in the timing of
the edges will be different for rising and falling edges.
The Level Type can be either absolute or relative.
The Absolute crossing level can be set directly in volts (or watts for an optical signal) , or you can click the
Find Level button to automatically find the level. The level is found by locating the midpoint between the
highest and lowest signal levels in the current acquisition. When you select the Absolute crossing level,
the crossing time used by both the jitter and eye pattern measurements is determined as the time at
which the signal level crosses the specified threshold. When Relative level is selected, the level is auto-
matically set on each acquisition. The value set is the selected percentage of the signal amplitude (which
equals base - top).
The Slope selector determines which edges are measured when "Clock" is selected in the Signal Type
selector in the Signal dialog.. The choices for slope are Pos(itive), Neg(ative) or Both Select the choice
that corresponds to the edge type that is used for clocking of the data in your device, or that is of inter-
est in your analysis. For example, if you only latch data on positive edges, select Pos. If you clock on
both edges, you can select Pos or Neg rather than both in order to understand how the edge type affects
the jitter measurement.
The Hysteresis entry box sets the hysteresis level to use for edge detection, in units of vertical divisions.
The hysteresis is the vertical amount that the signal is required to travel beyond the crossing level to
allow detection of a crossing in the opposite direction. Incorporating hysteresis in the edge detection
algorithm prevents the software from finding false edges that would otherwise be detected due to noise
921143 Rev A 17