# oscilloscope

An oscilloscope ( lat. Oscillare "to rock", old Gr. Σκοπεῖν skopein "to look at") is an electronic measuring device which, in its preferred application, makes the time course of one or more electrical voltages visible on a screen . The oscilloscope displays a progression graph in a two-dimensional coordinate system, with the (horizontal) x-axis usually being the time axis and the (vertical) y-axis being the voltage axis . The resulting image is called an oscillogram .

There are analog and digital oscilloscopes, with the analog devices being almost completely replaced by the digital ones. In addition to the multimeter, the oscilloscope is one of the most important measuring devices in electronics and electrical engineering . The range of measurable voltages extends on the one hand from direct voltage to low-frequency voltage as it occurs in the electrical supply network , voltage in sound engineering to high-frequency signals in radios, televisions or computers; on the other hand, the range extends from a few millivolts to a few hundred volts ( peak value ) using standard accessories .

Sometimes the term oscilloscope is still used, which was correct in an early phase of development for a device that could write on paper. Sloppy terms are Scope , Oscar or Oszi .

Digital oscilloscope - with liquid crystal display
Analog oscilloscope from the 1970s - with tube display

## Measurement

In general, every process that can be mapped as a time curve of an electrical voltage can be represented with the oscilloscope by a continuous or discontinuous curve. It also has a rectangular display area. Periodic courses are preferably considered, the characteristic details of which of their “shape” are to be recorded. The x-deflection is used to display time.

The input voltages are usually connected directly via BNC sockets on the front or using a probe . In laboratory devices, the sockets are connected on one side to earth (housing, protective contact) via a protective conductor . Accordingly, each voltage to be measured must be earthed or potential-free on one side in the same way . Preferably there are 2 or 4 input channels for influencing the y-deflection of 2 or 4 input voltages.

Most oscilloscopes can use an input for the x deflection, which means that not only time-dependent functions can be represented (ty representation), but also xy representations (such as Lissajous figures or characteristics ). Occasionally there is a z-input that can be used to influence the intensity of the curve.

Many physical quantities can be represented by voltage signals via transducers . Then the oscilloscope and the details of how to peak-to-valley value , DC component or period , period , phase shift are measured.

Depending on the equipment, it is possible to form sums or differences between two channels or to display other than temporal relationships, for example in the form of

## Structure and setting options

Block diagram of an oscilloscope

The block diagram shown gives an overview of the structure of an oscilloscope.

The setting options are varied depending on the make: The options mentioned here are representative and by no means complete or are not available on every device.

### Vertical assembly

Typical structure of an entrance stage

Essential components for this are for each channel

• a high-quality amplifier from direct voltage to alternating voltage with high frequency (typically 100 ... 500 MHz)
• a selector switch for the voltage measuring range
• an adjuster for the height of the zero line (y position)
• an input selector switch with the possibilities of recording
• the alternating voltage component of the applied voltage (position AC)
• of the total voltage including DC component (position DC)
• the zero line (position GND).

Like any measuring device, an oscilloscope should influence the circuit to be examined as little as possible and falsify the applied signal as little as possible. This means that the input impedance should be as high as possible, while at the same time there should be as few reflections as possible on the measuring line. These demands cannot be combined with one another.

• A universal oscilloscope usually has an input resistance of 1 MΩ and an input capacitance of 20 to 50 pF. With a probe , the resistance can be increased and the capacitance reduced, but mostly without using the lowest measuring ranges for small voltages.
• With special oscilloscopes for anechoic high-frequency applications, the input resistance can be 50 Ω.

A special feature of the oscilloscope: the voltage zero point is neither fixed at the edge of the image nor fixed on the center line, but always where it is individually placed for optimal use of the screen.

### Trigger assembly

For triggering:
Thin line: a sawtooth voltage continuously present at the input.
Thick line: Part of the input voltage that is visible on the screen at a given scale (when triggered on a positive rise).

An applied signal is continuously measured and drawn over and over again from the left edge of the display area to the right. In order to obtain a steady picture with the periodic signals, it is necessary to stop the screen scan until the signal to be displayed has reached a defined initial state. Only then is a new display triggered. The runs are thus identical and always refresh the picture.

Usually these are set

• the trigger level (LEVEL, continuously adjustable voltage)
• the trigger rise with which the signal crosses the trigger level
(SLOPE, + or -)
• the trigger mode (MODE, normal or automatic).

If the set trigger condition is not met by the selected trigger signal, the time base remains in waiting position during normal operation; in automatic mode, a display is created, albeit free-running. For example, DC voltage cannot trigger; Free-wheeling is also helpful for looking for the signal path until the vertical assembly is correctly set.

The trigger source from whose voltage curve is to be triggered comes into question

• each of the channels (CH1, CH2, ...)
• an external trigger input (EXT)
• the supply network (50 Hz; LINE), since network-synchronous events are often to be recorded.

Depending on the equipment of the oscilloscope, there are also special trigger circuits that z. B. recognize TV signals or the I 2 C bus cycle and use them to trigger.

### Horizontal assembly

A time base , which must also meet high requirements, ensures that the image runs horizontally . It has adjustment options for

• the time measuring range
• the horizontal position of the beginning of the picture.

In the case of an analog oscilloscope, it generates a voltage that rises strictly linearly over time from the trigger time (" sawtooth voltage "), which is used for the horizontal deflection.

In the case of a digital oscilloscope, the process is scanned and the data from the measuring points are stored in a data memory that is repeatedly overwritten in a ring. Here the time base ensures the time interval in which the measurement data are obtained and written to the memory. These are then used to build the image from a specified distance to the trigger time. The data store accepts data for a longer period of time than the period shown on the screen. This means that the history of the trigger event (“pre trigger”) can be displayed.

With digital oscilloscopes there is also the convenient option of displaying one-off events ("single"). As of the trigger event, the data memory is only written with a defined number of measuring points, but is no longer constantly overwritten. This allows a transient signal to be recorded and displayed for as long as desired.

Convenient oscilloscopes have two time bases. In addition to the main time base, there is a second time base that can be used to generate enlarged sections with faster processing. This starts after an adjustable delay time after triggering the main time base; or it can be triggered after the set delay time due to a second trigger event. In this way, an event can be resolved much more finely than is possible with the main time base if the event occurs a longer distance after the trigger event. The second time base can be omitted if the data are recorded from the outset and written to the memory much more densely than they can be used for image construction. For a better resolution of the event, a section of the data is shown spread out.

### Measuring ranges

In order to be able to read measured values, the screen contains a grid. It is preferably equipped with 10 divisions (“div” for short) horizontally and 8 divs vertically. A measuring range is here - preferably different from the measurement - not by a zero and a full scale value , but by a scale ( "scale") or in deflection.

The indication is used for a quantitative description of the time on the screen

${\ displaystyle {\ frac {\ Delta t} {\ Delta x}} = {\ frac {1} {\ text {Horizontal speed}}} = {\ text {Horizontal scale}}}$

Typical adjustable scales are 10 ns / div… 1 s / div with three settings per power of ten in the factors 1, 2 and 5.

But 20 ps / div or 10 ks / div are also offered.

The setting options therefore typically extend over the large range of around eight powers of ten, and in some cases a few more.

The specification serves as a quantitative description of the voltage on the screen

${\ displaystyle {\ frac {\ Delta U} {\ Delta y}} = {\ frac {1} {\ text {Sensitivity}}} = {\ text {Vertical scale}}}$

Typical adjustable scales are 2 mV / div… 5 V / div in the same steps as for time.

## species

### Digital oscilloscope

Oscillogram recorded with a digital oscilloscope

#### overview

Digital storage oscilloscope
Compact DSO
Connections of a DSO, here for RS-232 , printer and GPIB

Today digital oscilloscopes (DSO, English: Digital Storage Oscilloscopes ) are predominantly used. After an analog technical amplification, they convert the voltage values into digital signals at discrete times and store the data in a data memory for each channel . These are then used to build the image, but can also be stored on an external memory after the measurement or transferred to a PC .

There are different levels of equipment as well as mixed forms between analog and digital oscilloscopes. In addition to the measurement options mentioned above, digital oscilloscopes contain other functions, for example:

• Pre-triggering for viewing the history of an event that triggers the trigger, for example when looking for the cause of a voltage spike
• Averaging over many image runs to suppress interference with periodic signals
• Calculation of rise time, pulse width, amplitude, frequency, etc.
• Calculation of frequency spectra / FFT , histograms and statistics
• automatic adjustment to an unknown signal.

The input voltage is digitized with an analog-digital converter (ADC) with a resolution of 8 to more than 12 bits. Flash converters are mostly used . In the case of high speed requirements, the converters available for 2 to 4 channels are operated in parallel, which then work time-shifted (interleaved) for 1 channel. An 8-bit ADC can resolve in 256 steps; Over a measuring range of 10.24 div there is a relative resolution of 25 steps / div, which is sufficient for viewing in the vertical direction.

In addition to the resolution in the y-direction (voltage), the time resolution is also an important parameter: It is determined on the one hand by the bandwidth of the analog input amplifier and on the other hand by the sampling rate at which the signal is sampled. Since digital oscilloscopes represent an application of discrete-time signal processing , the sampling rate and the Nyquist-Shannon sampling theorem play a central role. The sampling rate is usually specified in “megasamples per second” (MS / s or Msps) or “gigasamples per second” (GS / s or Gsps), i.e. the number of samples per second. At the beginning of 2009, even in the lower price segment (800 to 2000 €) of the DSO, the sampling rates were in the range of 1 GS / s with bandwidths (−3 dB) between 60 and 200 MHz.

Example: If a point density of 50 S / period is considered desirable on the screen for a curve that is not simple, this is possible at a sampling rate of 1 GS / s up to a signal frequency of 20 MHz. The undersampling described below then starts around the 25th harmonic .

Another parameter is the memory depth, which in the oscilloscope means the number of measuring points that can be saved. It is given as a total number or per channel. If it is only a matter of viewing the image, a horizontal point density of 50 S / div is sufficient, with an image width of 10 div this means a storage depth of 500 points, for pre-trigger with the trigger event on the right edge of the image another 500 points. However, if it is desired to isolate the cause of timing anomalies in a complex digital data stream, you may find it useful. U. Millions of points required as memory depth.

DSOs are often implemented on the basis of FPGAs , since the small number of pieces and the flood of data to be processed and stored cannot always be achieved with a DSP . Above a sampling rate of approx. 1 GS / s, DSOs often use several AD converters per channel in parallel (interleaved mode), which sample the signal out of phase. At very high frequencies, the low clock jitter is the strongest quality criterion.

The development towards ever smaller devices has made it possible not only to create very compact DSOs for use in the laboratory, but also to create robust, portable "handheld" oscilloscopes for use e.g. B. on assembly and maintenance. These are floating, some are potential-free in all input channels and are often equipped with multimeter functions.

#### Subsampling

If the applied voltage (thin line) is scanned too seldom, the measuring points are combined to form a distorting image (thick line). In this simple case the frequency is obviously wrong (too low).

The scanning can no longer follow the process at higher and higher frequencies of the input voltage. If there are fewer than 2 points per period, undersampling occurs and the aliasing effect creates images that no longer have anything in common with the original gradient. Periodic signals can, however, be reassembled correctly by means of samples from many runs. A prerequisite is a very fast sample-and-hold circuit that can detect the input signal in a particularly short time. Two proven periodic sampling techniques are:

Sequential sampling: There is only one sampling per trigger. In the first run, the sampling time is a short delay time behind the trigger point. The delay time is doubled for the second run and tripled for the third - until the time window is filled. The pixels are arranged in the order in which they are scanned, one below the other at a small delay time interval.

Random sampling of a signal in several cycles

Random (independent of the triggering) sampling (random sampling): Here is taken each measurement point within the possible speed of operation, and in addition, its temporal distance is measured to the trigger point. The image points are arranged in the order of this time interval. With a sufficiently long acquisition time, the image points are so close that a closed curve appears.

With these techniques, however, no low-frequency signal components may be present, since these would show up as a blurring in the constructed curve.

#### Peak detection (glitch detection)

With digital storage oscilloscopes, there is a risk that very short events are recorded incorrectly or not at all between two sampling points due to the aliasing effect, especially with slower time base settings. So that voltage peaks (English: glitches ) are always detected, some devices have permanently available (i.e. analogue) hardware peak detectors whose positive or negative peak values ​​are temporarily stored, digitized separately and inserted into the image.

#### Differences compared to the analog oscilloscope

• The display can be larger and colored, making it easier to differentiate between the individual channels.
• Frequent sampling and averaging over successive runs result in better resolution down to below the mV / div range as well as interference suppression.
• Short-term events can be lost between samples, except for peak detection.
• With pre-trigger, the signal course can be viewed before the trigger time.
• Complicated trigger functions such as pulse width triggers or in the context of serial interfaces provide triggering on a sequence of serial bit patterns.
• Autoset and Autorange cause an automatic, in many cases optimal adjustment to the input signal. However, analog oscilloscopes of the newer design also have this function.
• Scrolling and enlarging over several saved graphs expand the display options.
• Even slow processes, e.g. B. a temperature profile over a day can be recorded.
• Instead of a one-dimensional list, the oscilloscope's memory can also contain a multi-dimensional array of the previous sampling intervals in order to simulate a phosphor screen. The previous periods are shown in different colors and thus distinguishable (e.g. colored eye diagrams ).
• Automation and remote control are possible via standardized interfaces such as B. serial interface , USB port or GPIB .
• Data or image files can be saved and integrated into other applications. This can be done via interfaces for connecting standard USB mass storage devices .
• Numerical calculations such as effective value or frequency of the displayed signal curve can be carried out and displayed in the image.
• Cursor functions make it possible to measure horizontal and vertical distances in the display. Analog oscilloscopes of the newer design also have these functions.
• Derived channels are possible; The term “math channel” also occurs. For example, the spectrum of a signal can be formed and displayed in real time by means of Fourier transformation .
• Certain device settings (setups) can be saved and called up again later. However, analog oscilloscopes of the newer design also have this function.
• The abundance of functions requires handling by setting a large number of operating elements. With digital oscilloscopes, these are only partially accessible on the front, sometimes as softkeys , otherwise by calling up menus.
• Incorrect representations due to the aliasing effect are not immediately recognizable. High quality oscilloscopes use clock dithering to suppress aliasing frequencies so that the user can immediately see that aliasing is occurring from the display.

### Analog oscilloscope

Analog oscilloscope

#### overview

Functional diagram of an electron beam oscilloscope

With analog oscilloscopes, the voltage to be measured is "projected" via a switchable amplifier onto the screen of a cathode ray tube using an electron beam. More precisely, the electron beam focused on a point is deflected in the y-direction by the input voltage. In the case of a time-dependent display, a tilting oscillation must be generated at the same time for the x deflection , which, triggered by the triggering, increases evenly with a switchable steepness and then quickly decreases again. The rise time of this sawtooth-shaped tilting oscillation gives the duration of the displayed signal segment. It can usually be set in a very wide range. The electron beam moves from left to right (during this time the image is created, which disappears again after a short afterglow period) and then immediately returns to the starting point. The beam is scanned dark so that the return of the light spot cannot be seen.

In contrast to other screens, the deflection of the electron beam in analog oscilloscopes is practically always carried out capacitively by electrical fields . This type of deflection is much easier to control over large frequency ranges; the advantages outweigh the disadvantages (light spot deformations with increasing deflection, large installation depth of the associated picture tube) in the intended area of ​​application by far.

For technical reasons, practical disadvantages (such as the size of the cathode ray tube) and economic considerations (such as the low cost of the DSO), analog oscilloscopes are only of secondary importance in practical laboratory use today.

#### Multi-channel operation

It is usually necessary to display two or more signals on the screen simultaneously in order to be able to recognize relationships. There are various methods of doing this.

• Two-beam oscilloscope: There are two electron guns, focussing and y-deflection systems in the tube, but a common x-deflection system. In this way, curves can be displayed independently at the same time. However, such devices have not been manufactured for decades. Two-beam tubes can also be combined with the following techniques in order to be able to display more than two signals.
• Multi-channel oscilloscope in chopper mode: You can switch quickly between the inputs and the sections of the gradients are displayed on the screen, for better optical separation at different heights. The display consists of a dashed (chopped) line for each channel, but at a high switching frequency (in relation to the deflection frequency) the segments move so close together that the eye sees a closed curve. When one signal triggers, the others run at the same time. This operating mode is mostly used with low deflection frequencies, for example when displaying slow signal curves below 100 Hz.
• Multi-channel oscilloscope in alternating operation: The signal of a channel is displayed once over the full width on the screen, then it is switched to the next channel and its curve is completely displayed at a different height - in continuous change. If the signal frequency is high enough, the eye can see the curves at the same time without flickering, which is why this operating mode is usually selected when displaying fast signal curves. The display of each curve is usually triggered by the same signal. As a result, the temporal relationship between the signals remains recognizable, provided that these are periodic processes. Some types of oscilloscopes can also be set so that each input signal triggers its own cycle. In this operating mode, however, the temporal relationship between the signals in the display is lost.

A multi-channel oscilloscope is more complex, since a separate vertical assembly is required for each channel.

### CCD oscilloscope

The oscilloscope has a small cathode ray tube whose electron beam generates the oscillogram on a CCD sensor located in the tube . Because the tube is very small, it can work in the GHz range. The oscilloscope has no sampling gap. An LCD monitor displays the image. A single sample can be captured by switching off the electron beam. The price for a modern device is around USD 20,000.

### Mixed signal oscilloscope

As mixed-signal Oscilloscope be called digital oscilloscopes, which not only via one or more analog inputs, but also on additional digital feature Inputs: The digital channels can usually be adjusted (to a particular logic family TTL , CMOS , etc.) and then only differentiate between the states HIGH, LOW and undefined.

### DSO as a computer accessory

Digital storage oscilloscopes are also offered as computer accessories. They are then either a plug-in card or a separate device coupled via an interface. You can also only consist of software and use a signal from an ADC card or (with limited accuracy requirements in the range between 10 Hz and 10 kHz) the audio input . However, none of these solutions achieve the parameters of autonomous DSOs, but are usually much cheaper. They can also be displayed graphically on a PC and can therefore be particularly helpful for teaching purposes.

### Waveform monitor

The waveform monitor (WFM) is a special oscilloscope that is used in professional video technology to measure analog video signals.

## Historical development

Depiction of the hospital clerk
Early oscillogram on film
Camera attachment for recording with an analog oscilloscope
Philips oscilloscope with tube amplifier technology , 1955
Older type of oscilloscope

The first automated devices at the beginning of the 20th century to record a signal curve over time used galvanometers to move a pen over a rotating roll of paper, as is the case, for example, with the hospital clerk . In an expanded form, but with basically identical functions, such devices are still common at the beginning of the 21st century in the form of measuring recorders , although they are increasingly being replaced by data loggers . The limitation is due to the mechanical movement in the narrow bandwidth , which only allows the recording of low-frequency signal curves.

Improvements replaced the mechanical pointer of the galvanometer with a mirror galvanometer and the signal curve was recorded optically on a light-sensitive film. However, the handling including the necessary film development was complex. A significant improvement resulted from the use of cathode ray tubes . The first cathode ray tubes were developed at the end of the 19th century, the use of a measuring device for signal recording with two electron beams goes back to a development from the 1930s by the British company ACCossor , which was later bought by the Raytheon company. These mostly uncalibrated devices were used as screens for the first radar devices during World War II .

Another improvement of the oscilloscope, in addition to a calibrated time base, was created by the possibility of triggering on periodic signal curves. This made it possible to precisely align the timing of the display of repetitive signal curves and created the basic functionality of an analog oscilloscope. The triggering was developed in Germany during the Second World War and was first used in 1946 in the commercially used oscilloscope Model 511 from the American company Tektronix .

Analog oscilloscopes with cathode ray tubes with an extremely long afterglow time, a so-called storage display tube, were used to display one-off, non-periodic processes . The long afterglow time was achieved by special coatings on the luminous layer in the cathode ray tube. The storage tubes had a time-limited storage time ranging from a few seconds to less than a minute and had a comparatively low spatial resolution and a limited operating time. For a long time they were the only way to display single events with times below about 1 ms. From times of around 1 ms upwards, event-triggered photographic recordings of the image of the cathode ray tube could alternatively be made.

Another development was the non-luminous blue writing tube, also known as the Skiatron. It needs an external light source. The electron beam strikes an externally visible layer of vapor-deposited alkali halides, usually potassium chloride . The negative charge of the beam causes a discoloration of the affected areas, which appears blue to blue-violet depending on the type. This trace is very permanent, lasts from a few minutes to a few days and can be erased by warming up.

Due to the additional possibilities of digital signal processing and storage, analog oscilloscopes were increasingly replaced by digital storage oscilloscopes (DSO) from the 1980s onwards. The prerequisite for this was the availability of analog-digital converters with high bandwidth. The first digital storage oscilloscopes were brought onto the market by Walter LeCroy , the founder of the New York company LeCroy , who had previously worked at CERN on the development of high-speed analog-digital converters for recording measurement signals.

## literature

Joachim Müller: Digital Oscilloscopes - The Way to Professional Measurement . Beam-Verlag, Marburg 2017, ISBN 978-3-88976-168-2 .

Wiktionary: Oscilloscope  - explanations of meanings, word origins, synonyms, translations
Commons : Oscilloscopes  - collection of images, videos and audio files

## Individual evidence

1. data sheet, page 12
2. data sheet, page 7
3. data sheet, page 2
4. The XYZ of analog and digital oscilloscopes. 2004, page 40 or 42
5. Welec W2000a project at Sourceforge
6. Evaluating Oscilloscope Sample Rates vs. Sampling Fidelity: How to Make the Most Accurate Digital Measurements ( Memento of December 3, 2008 in the Internet Archive ), (PDF; 1.3 MB)
7. Herbert Bernstein: LF and HF measurement technology: measuring with oscilloscopes, network analyzers and spectrum analyzers. Springer Vieweg, 2015, p. 129.
8. ^ Hawkins Electrical Guide, Theo. Audel and Co., 2nd Edition 1917, Volume 6, Chapter 63: Wave Form Measurement , page 1851, Figure 2598.
9. ^ Frank Spitzer and Barry Howarth: Principles of modern Instrumentation , Rinehart and Winston, New York, 1972, ISBN 0-03-080208-3 , page 122.