Fashionable oscilloscopes and digitizers are getting higher and higher. Greater bandwidth, higher vertical decision, and longer acquisition recollections. To not point out extra firmware instruments for application-specific measurements. With all these superior evaluation capabilities it’s typically onerous to recollect some very previous and easy guidelines that may enhance the accuracy and precision of your measurements. Listed here are just a few good concepts to assist.
Use the total dynamic vary of your instrument’s entrance finish
Digital devices feed their enter indicators to an analog-to-digital converter (ADC). The dynamic vary of the ADC is said to its variety of bits of decision. The instrument matches the enter sign to the ADC enter voltage vary utilizing attenuators or amplifiers. If the enter to the ADC is lower than its enter vary, it reduces the whole dynamic vary of the ADC. This may occur when customers arrange a number of traces on the display.
Some oscilloscopes and digitizers show software program solely supply a single show grid. When you attempt to show a couple of sign hint at full dynamic vary, the indicators overlap, making it onerous to view them. Most individuals confronted with this drawback cut back the vertical scaling of every channel. In case you have 4 traces, simply enhance the volts per division setting by an element of 4. Every hint now occupies solely 1 / 4 of the display and all 4 traces match on the display with no overlap. Downside solved? Probably not. You simply decreased the dynamic vary by two bits, you made your eight-bit oscilloscope a six-bit oscilloscope. You attenuated the sign, however the inside noise of the instrument is identical, the signal-to-noise ratio is now worse by two bits. Determine 1 reveals the impact of the lack of dynamic vary.
Determine 1 An instance of the lower in signal-to-noise ratio as a consequence of lowering the sign amplitude with a purpose to match a number of traces on a single grid.
The underside grid reveals the unique sign acquired a 50mV/division. The highest hint reveals the hint acquired at one quarter of the total display or 200mV/division. When you vertically increase the attenuated hint and show it on the authentic 50mV/division, the vertical noise has elevated considerably as you possibly can inform from the thickening of the displayed hint. Measurements made on the attenuated hint could have elevated uncertainty as a consequence of a poorer signal-to-noise ratio. This isn’t an issue for an oscilloscope or digitizer that has a number of grid shows, every grid shows the sign at full dynamic vary and a number of indicators may be in contrast, every in its personal grid. When you don’t have entry to a a number of grid oscilloscope, make it possible for any measurements are made on the total amplitude indicators, reserve the attenuated sign for visible comparisons solely.
Enhance dynamic vary and measurement accuracy by eliminating noise
Use sign processing within the type of averaging or filtering to cut back noise, enhance dynamic vary and measurement accuracy. Ensemble averaging, the place the nth samples of every acquisition are averaged collectively over a number of acquisitions, reduces Gaussian noise proportional to the sq. root of the variety of averaged indicators. This may carry a low stage sign out of background noise for higher measurements. It does require a number of acquisitions.
For a single acquisition, you possibly can cut back noise by bandwidth limiting the sign. The advance of dynamic vary is proportional to the sq. root of the bandwidth discount. Cut back the bandwidth by an element of 4 to attain a two-to-one enchancment in dynamic vary. This assumes that the sign has a low bandwidth and isn’t affected by the bandwidth discount. Determine 2 reveals the development that may be achieved utilizing both averaging or filtering.
Determine 2 Averaging a number of acquisitions or filtering a single acquisition can enhance the dynamic vary of the acquisition by eliminating noise.
The acquired sign is an exponentially damped sine wave. The highest hint reveals a uncooked acquisition. Notice that the sign disappears into the noise about three quarters of the way in which throughout the display. The middle hint reveals the common of a number of acquisitions. Within the backside hint, a Gaussian low move filter has been utilized to the acquired sign. Each averaging and filtering cut back the noise and enhance the dynamic vary of the measurement. The sign is clearly discernible after both kind of sign processing.
Bettering the accuracy of cursor measurements
Cursors are vertical and/or horizontal traces that may be moved over the oscilloscope or digitizer show to mark vital factors on a waveform. Cursor readouts present the time or amplitude of the waveform on the cursor location as proven in Determine 3. The waveform is a keyed RF service and the horizontal relative cursors are used to measure the width of the RF burst. It is a measurement that can’t be made with the instrument’s computerized measurement parameters. The cursor horizontal readout seems within the decrease right-hand nook beneath the timebase annotation field, it reads 8.06275 µs.
Determine 3 Horizontal relative cursors are used to measure the period of an RF pulse burst.
Is that basically the period of the burst? The reply is not any. This waveform has two million samples within the acquisition. The horizontal display decision is 1920 pixels. So, it’s apparent that not all of the samples are proven on the display. Instrument producers apply compaction algorithms to cut back the variety of displayed factors. They handle to indicate vital factors like peaks however there’s nonetheless quite a bit that you could’t see except you increase the show.
A extra correct technique to make this measurement is the used the zoom traces to horizontally increase the waveform at first and finish of the RF burst as proven in Determine 4.
Determine 4 Utilizing zoom traces to make extra correct placements of the cursors on the first and final pattern factors of the burst.
Zoom traces Z1 and Z2 horizontally increase the start and ending of the burst. The pattern counts within the zoom traces are smaller than the display decision so the compaction algorithm isn’t used. The cursors monitor on the acquired sign and the zoom traces. The cursor on zoom hint Z1 (yellow hint) marks the start of the RF burst which begins at a zero crossing. The cursor on zoom hint Z2 (pink hint) marks the top of the burst. The cursor horizontal readout reveals the burst size as 8.33295 ms, a extra correct end result.
Constructed-in measurement parameters
Oscilloscope and digitizer help software program provides built-in measurement parameters. Most oscilloscopes embody about twenty or extra widespread measurement parameters like amplitude, frequency, rise time, and fall time to say just a few. Utility-specific software program packages can enhance the variety of out there parameters to over 100. The usual parameter measurements are normally based mostly on IEEE normal 181 which employs statistical methods to make the measurements on pulse waveforms as proven in Determine 5.
Determine 5 IEEE Normal 181 bases pulse measurement parameters on a statistical dedication of the highest and base values of a measured pulse.
The amplitude values of the heart beat high and base are decided by forming a histogram of the acquired samples of the waveform, that is proven as an inset on the appropriate of the display. A sq. wave or pulse waveform could have a histogram with two distinct peaks. The typical or imply values of the higher histogram peak known as the “high”. The imply of the decrease valued peak known as the “base” of the waveform. Utilizing the statistical imply of many pulse measurements suppresses the consequences of waveform aberrations like noise, overshoot, and ringing. The heartbeat amplitude is the distinction between the highest and base. The utmost worth of the waveform minus the highest is the constructive overshoot. Likewise, the distinction between the waveform minimal worth and the bottom is the adverse overshoot. The heartbeat width is the time distinction between the main and trailing edges crossing the mid-amplitude or median between the highest and base. The height-to-peak worth of the waveform is the distinction between the utmost and minimal amplitudes. The transition time measurements, like rise and fall time, measure the time to transition from 90% to 10% of the heart beat amplitude. If the waveform isn’t a pulse, the measurement engine sees that as a result of the waveform histogram has kind of than the 2 peaks that outline a pulse. In that case, the amplitude measurement reverts to a peak-to-peak measurement and signifies the truth that the waveform isn’t a pulse utilizing a measurement standing icon beneath the parameter readout.
In virtually all circumstances, the measurements made utilizing measurement parameters are much more correct than these made utilizing cursors. They’re additionally made routinely saving quite a lot of time.
Measurement Statistics
How do instrument measurements fluctuate from measurement to measurement? Measurement statistics reply that query. Many devices embody statistics reporting together with the fundamental measurement parameters as proven in Determine 6.
Determine 6 Measurement statistics report how measurement values fluctuate over a number of measurements displaying the final worth, imply, minimal, most, normal deviation, and whole inhabitants.
Some oscilloscopes embody all occasion measurements. Time-related measurements, like frequency and width, report one worth for every cycle of the measured waveform. In case you have 100 cycles of a sign on the display, the measurement engine provides 100 measurements for every acquisition. Amplitude-related measurements solely add a single worth per acquisition. You possibly can purchase a great many measurement values over many acquisitions. Measurement statistics present very helpful views of this knowledge. The desk beneath the waveform show, proven expanded within the blue field, lists the final worth measured within the acquisition, the imply values of all of the acquired values, the minimal and most values of the set, the usual deviation of the set, and the whole inhabitants of all of the measurements. It additionally features a standing indicator and an iconic histogram of all of the measurement values.
The amplitude measurement reviews that 11,873 values are included within the statistics. The imply or common worth is 237.5457 mV. The imply worth is reported with the next decision than the final worth as a result of the imply is an averaged worth. As we noticed within the waveform, averaging the averaging course of improves the vertical decision of the measurement, the identical factor occurs should you common a number of measurements, therefore the extra vital figures within the imply worth.
The biggest worth is 241.5 mV reported as the utmost, the smallest worth, minimal, is 234.8mV. These values might help detect transient occasions that happen in the course of the acquisitions. Different instruments can plot measured values versus time to see when transient occasions happen and match them in time with potential sources.
The usual deviation describes the distribution of the measured values in regards to the imply, on this case is it 826 µV. The imply and normal deviation are helpful in understanding the distribution of the measured values as is the enduring histogram. The enduring histogram may be expanded to view the full-sized histogram for a extra detailed evaluation with its personal set of histogram measurements. All of those measurement instruments assist perceive the dynamics of the actual measurements. A information of the distribution of measurements allows you to set up take a look at limits for indicators.
Conclusions
These instruments and methods might help enhance the measurement accuracy and reliability of your devices. Different methods may be gleaned from producer’s webinars and utility notes. The extra you study your instrument, the extra correct and dependable your measurement outcomes will probably be.
Arthur Pini is a technical help specialist and electrical engineer with over 50 years of expertise in electronics take a look at and measurement.
Associated articles:
