Electricity: measuring and testing – Measuring – testing – or sensing electricity – per se – Cathode ray
Reexamination Certificate
2003-02-28
2004-06-22
Pert, Evan (Department: 2829)
Electricity: measuring and testing
Measuring, testing, or sensing electricity, per se
Cathode ray
C324S076220, C702S069000
Reexamination Certificate
active
06753677
ABSTRACT:
BACKGROUND OF THE INVENTION
Real time digital oscilloscopes operate by digitizing closely spaced consecutively sampled values of an applied waveform, storing the digital values in a memory, and then re-constructing the waveform as a displayable image (the “trace”) on a raster-scanned CRT by reading and processing the stored values. We shall call the stored digital values an “acquisition record” and note that its contents correspond to a definite time interval in the history of a signal's behavior. The length of that time interval is largely determined by the number of addressable memory locations devoted to a signal's acquisition (the memory's “depth”) and the rate at which the samples are acquired. That sample rate will be high enough to meet Nyquist sampling and bandwidth requirements, but is in all probability not high enough to directly produce a satisfactory reproduction of the waveformn, even though it does determine what that reproduction ought to look like. Accordingly, it is conventional to “fill-in” the “missing” samples before all or some portion of the acquisition record is used, by sending the portion of interest through an interpolation filter before being used by other mechanisms.
It often happens that the acquisition record contains many times more digitized signal data than is displayed at any one time, allowing the operator to select as the displayed trace a size (degree of zoom) and location (panning) of a subset of the acquisition record. To support panning and zooming, and because the sampled values generally have an arbitrary relationship to the horizontal pixel positions in the display, the digitized data stored in the acquisition record is algorithmically processed (“rendered”) to produce a graphics image in a frame buffer, from whence is created a raster-scanned display of quite modest scan line rates (as things go for CRTs). That is, the trace is not written immediately while the variations in its corresponding input signal are occurring. Only the taking of the samples is performed in real time, and once they are stored the raster-scanned display is a low speed representation of what the 'scope remembers having sampled. As mentioned, those sampled values in the acquisition record that correspond to the displayed trace are not necessarily the same in number as the number of horizontal pixel positions in the display, and digital signal processing techniques are commonly used to produce equivalent Y amplitude values that most probably would have been measured, had there been true alignment between the consecutive samples and the horizontal pixel positions. This rendering operation is essentially a process of interpolation.
Furthermore, the vast majority of oscillographic activities require that the displayed portion of the acquisition record be in some “defined relation” to a detected event originating outside the 'scope, whether that is a particular type of signal condition or transition occurring in the signal being measured, or a condition or transition in some other signal that is merely related to the signal being measured. The detection and operational response to this “defined relation” is called triggering; in the first case it is termed “internal” triggering, and “external” triggering in the second.
The term “real time” sampling is generally applied to the manner of digital oscilloscope operation described above. There is another mode of operation where a repetitive waveform cannot be densely sampled within a single instance thereof, but one (or perhaps more) samples can be obtained. For example, if the locations of those samples are allowed to “drift across” the waveform during successive instances thereof, enough samples can eventually be accumulated that a faithful reproduction of the waveform is obtained. (Provided, of course, that the waveform is repetitive, and that the trigger circuit re-triggers each time at the same relative location upon the successive instances of the waveform. The “drift” is then a precession relative to the trigger.) The old analog sampling oscilloscopes used this general technique (although they did not digitize and store values in a memory). We shall refer to the technique of digital oscilloscope operation that uses precession of samples relative to the trigger as one that practices “equivalent time” sampling. There is another equivalent time sampling technique that uses random sampling. It will be appreciated that the various manufacturers each have their own lexicography, and that there are many operational properties (and their associated descriptive terms) that arc reflections of differing internal architectures. We have chosen hereto use a particular division of the world of digital oscilloscopes into “real time” and “equivalent time” sampling, as defined above, because it is appropriate for the description that follows. It is believed that there is no significant inconsistency between our definitions and any of those promoted in the prior art. Accordingly, what follows herein should be understood according to our above definitions.
There are several useful modes of operation in connection with triggering, some of which deal with how to detect the trigger condition, and some of which deal with what to do once the trigger condition has been detected. In one (what-to-do) mode, the occurrence of the trigger indicates that thenceforth the acquisition record will be filled (or re-filled if it already has data in it) until one acquisition record's worth of subsequent data has been stored, and then the measurement ended. In another mode, some fractional portion of the acquisition is subsequently filled (the usual assumption being that the preceding portion has already got data), whereupon the measurement is ended. At the other extreme, the occurrence of the trigger can immediately end the storing of samples into the acquisition record.
The entire acquisition record can then be displayed, if desired. More typically, some selected subset of the acquisition record is displayed as the trace, beginning at some selected offset from that location in the acquisition record that corresponds to the occurrence of the trigger. That allows panning and zooming, which means that the location in the acquisition record corresponding to the trigger may or may not be a part of the displayed portion of the trace. If the trigger location is included in the visible portion of the acquisition record, then it is customary to indicate its location in the trace with some stylized indicia. (The algorithmic rendering mentioned above undertakes this annotation task, along with displaying a graticule, time per division indica and volts per division indicia, etc., by the inclusion of suitable visible pixels into the frame buffer.) The displayed and annotated trace may thus represent waveform activity that preceded the trigger, occurred after the trigger, or some combination of the two.
An aspect of real time digital 'scope operation that is of interest concerns how often a completed acquisition record is used to create a new displayed trace. In one case, a real time 'scope may capture, in just one instance of an event of interest, a complete description of that event and then display it indefinitely, regardless of what occurs next in the waveform. This is usually termed “single shot” operation. In another mode of operation, say, for a repetitive waveform such as a modulated sine wave, the acquisition record may be obtained, a portion rendered for display, and while that portion is being displayed for some brief period of time, a new acquisition record is triggered and taken, and then similarly rendered and displayed, etc. This amounts to a free-running, or “normal” mode of operation. The various manufacturers of digital 'scopes each have their own favorite terminology . . . ) This mode of operation allows the presence of variations in the waveforn to be observed.
Central to all that has been mentioned to this point is the notion of triggering. The ability to trigger on an event of interest allows the
Draving Steven D
Urban Ralph
Weller Dennis J
Agilent Technologie,s Inc.
Hollington Jermele
Miller Edward L.
Pert Evan
LandOfFree
Trigger jitter reduction for an internally triggered real... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Trigger jitter reduction for an internally triggered real..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Trigger jitter reduction for an internally triggered real... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3328448