Tektronix WFM 601i, Руководство пользователя

Страница: 92 / 135

Measurement Applications
3–14
Rev Aug 1994
WFM 601i Serial Component Monitor
Of the two methods described above, the EDH system is the most powerful.
First, it works with any video program material, since the source-end CRC is
calculated on a field-by-field basis. It does not matter if the source is a fixed
pattern test signal or moving pictures. Thus, EDH can monitor error rate while a
facility is in service. Second, the error information gathered by the WFM 601i is
the most comprehensive when using EDH, with errored seconds readouts for
both the active picture and full field. The big disadvantage of EDH is that it
requires calculation and insertion of the CRC at the source. At present, there is
very little equipment available that does this.
The AP CRC Value method does not require any special signal insertion at the
source, and therefore gets around this problem of the EDH system. But its
limitation is that it only works with fixed pattern signals, such as those from a
test signal generator. This precludes its use for in-service error monitoring. Also,
the error information is limited, with the only indicator being whether or not
there were zero errors. The AP CRC Value method is most useful for
manufacturing test environments and for situations where a zero error rate
condition is being verified.
Since the WFM 601i gives the AP CRC Values, this can be used to verify proper
operation of a video processor given a specified test signal. For example, a
manufacturer could specify the expected CRC values for various operating
modes with a given test signal. Any disagreement may indicate a problem. This
is analogous to “signature analysis” that is sometimes used to troubleshoot
digital logic.
Advanced Eye Pattern Modes
The WFM 601i eye pattern system was designed to facilitate comparison
between the serial rate signal and the video information it represents. Three
applications that highlight this capability are described below.
When video is serialized, a 270 MHz rate serial clock is derived from the 27
MHz rate parallel word clock. Often there will be slight phase modulation of the
serial clock between the transitions of the parallel clock. That is, if the eye
pattern formed between adjacent parallel clocks were viewed, ten eyes would be
displayed, but the spacing of the eye crossings might vary from one eye to the
next. This jitter is not random but correlated to the parallel word rate. Also, the
video patterns applied to the serializer may change at a 27 MHz rate or an integer
fraction of 27 MHz (for example, luminance samples may change at a 13.5 MHz
rate and color difference samples at 6.75 MHz). Thus any video pattern related
effects in the serial system typically show up at fixed locations with respect to
the parallel word. Because the 10-eyes are reconstructed over one tv-line, tv-line
correlated phase errors are superimposed over the 10-eyes. If the pattern repeats
often enough, and if the eye pattern display is triggered on parallel word
boundaries and tv-lines, these word/tv-line correlated effects can be observed.
Comparing the Two
Methods
Observing Word
Correlated Behavior