REPORT IN ADC AND DAC NAME: ID :

GROUP:

1-Analog-to-digital converter:

Analog to Digital Converters (A/Ds)

What Are They? Some Properties Of A/D Converters Where Do You Use A/Ds?

You are at: Bits&Bytes - Interface Circuits - Analog To Digital Converters (A/Ds) Return to Table of Contents

A/D Converters
Analog-to-Digital converters - a.k.a. A/D converters - are widely used by many engineers and scientists of all types, often without their realizing it. Whenever they make a measurement of a voltage, and that measurement is taken into a computer, an A/D is used. If you're going to take measurements - and just about every engineer will do a lot of that - then you will be better off if you understand some of the basic ideas behind A/D converters. There are two simple goals for this lesson. Given an A/D converter with a given range and number of bits, To be able to calculate the resolution of the converter. Given an A/D converter in the laboratory, To be able to determine the resolution of the converter and the number of bits used in the converter.

What Are A/D Converters?

A/D converters are electrical circuits that have the following characteristics.

The input to the A/D converter is a voltage. o A/D converters may be designed for voltages from 0 to 10v, from -5 to +5v, etc., but they almost always take a voltage input. (Some rare exceptions occur with current inputs!) In any event, the input is an analog voltage signal for most cases. The output of the A/D converter is a binary signal, and that binary signal encodes the analog input voltage. So, the output is some sort of digital number.

A comparator can be used as a simple one-bit A/D converter. Although a converter with just one bit isn't particularly useful, you can begin to see how an A/D converter works by puttering with it for a moment. If you read the lesson on comparators you encountered a simluation of a comparator. That simulator is reproduced below. Click here if you want to read the lesson on comparators. Comparator Simulator Sim1 Here is the comparator simulator. You can think of a comparator as a one-bit A/D converter. The input is an analog signal, and the output is a one bit digital representation of the analog signal. In the simulator, you can control a simulated voltage source that is the input the the comparator, and the digital output bit is indicated with a simulated LED. Notice the following.

The input can range from zero (0) to ten (10) volts. When the input voltage goes above five (5) volts, the output is a binary one (1) and the LED lights. When the input voltage is less than five volts, the output is a binary zero (0) and the LED does not light.

Properties of A/D Converters

The comparator simulation reveals a few important facts about A/D converters.

An A/D converter has a range. The simulator has a range from 0v to 10v, or a total of ten (10) volts. The total range of an A/D is the difference between the highest and lowest voltages the A/D can convert. An A/D has a resolution that is determined by the largest count that the A/D's counter/register can hold. The largest count is determined by the number of bits in the counter.

o o

If there are N bits in the counter, the largest count is 2N-1. In the comparator, there is only one bit, and the largest count is 1, and there are only two different outputs that are possible.

Clearly, if you want a more accurate conversion the converter will need to have a lot more than just one bit. Let's look at another simulation. This simulation is a four-bit A/D converter. Four Bit A/D Converter Simulator Sim2 Here is a simulation of a four bit A/D converter. Note the following:

As in the comparator simulation, we assume that the input voltage can range from zero (0) to ten (10) volts. We have added some "fine control" with two buttons that "nudge" the voltage up or down, but not beyond the 0-10v range.

Try the simulator first, and then we will examine what happens in a little more detail.

Now, consider the following observations about the converter above.

There are four bits in the simulation converter. The range is from 0v to 10v, or 10v total range. With four bits, there are 16 different count values (0 through 15). Thus, 10v is divided into 16 different parts, each part being: o 10/16 = 0.625 volts wide. Now, let's consider an example question.

Example

E1 How many bits would you need to divide 10 v into .01 v intervals? To get the answer to the question consider the following.

If you divide 10 v into .01 v intervals you need 1000 intervals. If you need 1000 intervals you need to think about a power of 2 that is larger than 1000. 10 o The smallest power of 2 that is larger than 1000 is 2 which is equal to 1024. That means that you need 10 bits in the converter, and the the count in the counter/register will run from 0 to 1023. And that leads us to observe that real converters often go to 10.23v, not 10v because that gives perfect .01v increments between resolvable voltages. o And another converter might run from -5.12v to +5.11v for the same reason.

Almost all A/D converters use a scheme in which the analog voltage is first converted into a binary integer (a "count", as in the simulator above) when the conversion is done. In some cases, there may be some software that gives you the actual count - the binary integer. In other cases the conversion might be converted - using software - to a a numerical value or a character representation of a numerical value. Example E2 A GPIB (IEEE-488) voltmeter is used to measure a DC voltage. The instrument uses an A/D converter that generates a binary number. Within the instrument, that binary number is converted to a string of characters which is, in turn, transmitted to a computer connected to the instrument.

To illustrate the concept in the example, consider the revised simulator below. In this simulator, the conversion algorithm - from the count in the register to an analog voltage - is given by:

Vmeas = Count*0.625 o Note: .625 = 10/16 = Range/#Counts

Here is the simulation.

Four Bit A/D Converter Simulator (Revised) Sim3 Here is a revised simulation of a four bit A/D converter. Note the following:

Vmeas = Count*0.625 o Vmeas is displayed on the simulation.

Of course, the simulation only raises a few more issues.

The largest issue is that the computed voltage is always lower than the actual voltage. o The average error would probably be less if we used a different way to compute the computed voltage. o The formula for the computed voltage is: Vmeas = Count*0.625

Using this expression for the calculated voltage, we can plot the calculated voltage as a function of the count. That's shown in the figure below.

Even more interesting is the plot of the calculated voltage against the voltage input that is being measured. That's shown next. We've also included a plot (the blue line) that shows what the output would be ideally (And that might take many, many bits in the converter!). Notice that the calculated voltage is always lower than the voltage input using the calculation method we assumed above. Note also that the largest error occurs just before the converter switches when the voltage is rising. For example, as the voltage rises from 0v to 3v, there is a time when the converter output switches from 0v to 0.625v. The largest error occurs just before that point, so a bound on the error is: Error < 0.625

We can reduce the error if we calculate the voltage differently. We can examine what happens if we use a different expression for the computed voltage. We will use:

Vmeas = Count*0.625 + 0.3125

The result is shown in the next simulator.

Now, here is a plot of the calculated voltage against the voltage input that is measured. Notice that the error limit is half of what it was using the first calculation method.

The conclusion is that the way the voltage is computed can affect the average error when an A/D is used to measure voltage values. Whenever you buy an A/D converter, or a voltmeter, or a data acquisition unit, you need to be cognizant of how the data presented to the user is actually computed. That's not usually a problem in instruments, but there are A/D computer cards that use the first method above to calculate voltage, and you should be aware of that when it happens. It is less important when the number of bits in the converter is higher, but when you have high requirements for accuracty you should be thinking of what might be taking place.

The Effect of Number of Bits

The number of bits used in the counter also affects the accuracy of the conversion. Again, we will use a simulation to show the effect of the number of bits. Six Bit A/D Converter Simulator Sim4 Here is a simulation of a six bit A/D converter.

Run the simulator, and compare results with the four bit simulators above.

We can think a little bit about error in an A/D. First, note that the error limit depends upon the range and the number of bits.

If there are N bits, the number of divisions is 2N-1. If the total range of voltage is VRange, then the size of a division is: N o SmallestDivision = VRange/2 The error is related to the Division size. o If the calculated voltage is SmallestDivision*Count, the error is the same size as the smallest division. o If the calculated voltage is SmallestDivision*Count + SmallestDivision/2, the error is SmallestDivision/2, i.e. half the size of the smallest division. o See the simulations above for a better appreciation/understanding of the error.

We're not going to give you any more simulators with more bits. Actually, we've almost reached the resolution limit for the screen. The simulators with the "nudge buttons" just nudge the controls by a single pixel, and that is too much for a twelve bit converter, for example. Example

E3 A popular A/D board gives a count and requires you to perform the computation shown in the line of C code shown below. The variable binary is the count value that is returned. MeasuredVolts = (((binary-2048)*20)/4096); The A/D converts voltages from -10v to +10v. We can conclude the following.
o

The converter has 12 bits. You get that conclusion by noting that the count can vary from 0 to 4095. When the count is 0, MeasuredVolts = 0.

Let's sum up a few points.

More is better when it comes to bits in an A/D. The indicated voltage will be closer to the actual value of the voltage being measured when there are more bits. More is pricier when it comes to bits in an A/D. It takes more parts, and they have to be made more accurately if there are more bits.

How Does An A/D Work?

While we have discussed A/D converters above, we haven't yet given you any insight into how you could build an A/D. Here we will discuss a simple way to build an A/D. This might not be the fastest A/D possible, but it will start to give you some insight into what happens inside an A/D. The circuit is shown in the simulator below.

Simulator - Showing How You Might Construct An A/D Sim 5 This simulator is a four bit A/D, and it consists of the following components.

A pulse generator that produces a sequence of 0's and 1's. A counter which counts the pulses and produces an increasing count.

The counter has a "Stop" input. When that input goes high, the counter ceases to count. A D/A which produces an analog signal proportional to the count output from the counter. An adjustable input voltage source. A comparator which produces a "1" when the D/A output is larger than the input voltage. o The output of the comparator is used to stop the counter.
o

Here is the circuit.

Here is how the A/D Simulator works.

First, set a DC voltage from the adjustable source. Second, click the "Convert" button. o The counter begins counting, and the count is displayed on the LEDs and the numerical value of the count is displayed. o The digital count signals (four bits of signals in this case) form the input to a D/A converter which produces an output voltage proportional to the count. o The output voltage from the D/A is compared to the input voltage. o When the output voltage from the D/A becomes larger than the input voltage the comparator develops a "Stop" signal that is used to stop the counter. o The count that remains in the counter is a digital representation of the input voltage.

Try the simulator, using different voltages within the allowable range, i.e. 0-5v. The simulator lets you see the inner workings of one type of A/D converter. Note the following about any A/D converter.

The conversion takes time. In the simulator, there is a variable amount of time until the counter reaches the correct count. We have slowed down the simulator so you can see what happens, but even in the best A/Ds there is some time that must elapse between the start of a conversion and the end of a conversion. That time limits how many conversions can be done per second. People have worked on that problem and there are other kinds of converters that convert more quickly. The conversion is not exact, and the accuracy depends upon the number of bits in the counter. The result of the conversion is an integer, and that integer still must be translated into an equivalent voltage (although the conversion isn't hard to do).

Practical A/Ds
Now, consider what happens in a typical application of an A/D. We'll look at a voltmeter. In a voltmeter, this is what happens.

A voltage is applied to the voltmeter, and an A/D converts the voltage to a count. The count is converted to a floating point number. The floating point number is displayed on an LCD or LED display.

If the voltmeter is connected to a computer (say through an IEEE-488 bus) the following also takes place.

The floating point data is converted to a character representation to transmit over an IEEE-488 or internet connection.

You can see that there are a few conversions that have to take place in this process. There are a number of other considerations here as well.

The number of bits in the A/D converter will affect the accuracy of the voltmeter. There is a lot of legend and lore in that area. Click here to examine that. You need to be aware of how to make conversions. Actually, you need to become knowledgeable about the different representations as well as how to program those conversions in various programming environments like C, C++, Visual Basic and LabView.

A/D converters are found in many places - including places where you might not think that you would find them. Here are a few.

Voltmeters are digital today. To display a digital result for a voltage measurement the voltage is first converted to digital form in an A/D, and then it is changed to a decimal format to be displayed. A/Ds are used in digital thermometers, so if you've spent some time in the hospital, you've used an A/D.

And, you get the idea. However, there is an important situation where you need to dig deeper into how A/Ds operate, and how they interface with the software and hardware that is often used in measurement and control situations. Let's think about a few typical situations.

Assume that you are recording temperature at four points in a heat-treating oven. You need to do the following. o You need to get the data into a computer and you need to compute and record the average temperature (the four temperature data points). You have three tanks used for chemical processing. You need to control the level of the liquid in the three tanks. Each tank has a small computer with an A/D board that measures and controls the liquid level in each tank and also measures the temperature. Each computer has a network card that connects it to an Ethernet LAN. There is a central computer on the LAN and you need to send information from each of the control computers to the central computer.

You need to take the data at each computer and you need to transmit it over the network to the central computer so that it can be recorded there and be available for any computations that might be necessary.

In both of these situations you need to consider how to manipulate the data that you measure and want to store. Here are some of the considerations.

If you take data with an instrument connected to a computer (like a data acquisition unit or a voltmeter) the instrument may be connected to the computer with an IEEE-488 (GPIB) connection. (Click here to learn more about IEEE-488.) IEEE-488 connections have some considerations. o Most IEEE-488 instruments convert data into strings of characters and those strings are transmitted to the computer. If you send data over a network, you must send strings of characters.

The conclusion that you need to reach is that you need to be aware of how data has to be converted in these kinds of situations. Let's look at the sequence of operations you might face.

The A/D converter gets some raw data or information and it is in the form of an integer count. Often it is simply a binary number represented with some fixed number of bits. o The count could be a number like 10001000 (or 129 in decimal numbers). o If the A/D converter has eight (8) bits or less, the data will take a single byte. o If the A/D converter has more than eight bits (which is typical) the data will take more than one byte. Usually, for most converters, the data will take two bytes, since most converters are 10, 12 or more bits. The count is converted to a number that represents the voltage (or possibly some other physical variable). That takes some computer code (And there will probably be some sort of computing chip in the instrument.) and the process results in a number that is probably a floating point number. o The floating point number might be something like 3.1416v. If you transmit that floating point number you need to generate a string of characters.

o o

The string of characters might be something like a "3", a period ("."), a "1", a "4", etc. In many cases you may need to add one or more characters to signal the end of a string. That could be a carriage return and a line feed, for example.

Representing Instrument Data Taking data with an instrument immediately leads you to consider several things about data. Here is a sample.

If you take a lot of data, you need to know how the data is represented. If you take a lot of data, you need to know how the data is transmitted from the instrument to a computer. If you take a lot of data, you need to know how the data you get will be stored if you store it in a computer file.

Digital Voltmeters
Digital Voltmeters are a special case of A/Ds. Obviously, if voltage measurements are taken and the results are displayed digitally with LED or LCD displays, the instrument has to contain an A/D converter. Digital voltmeters have some characteristics that you might need to understand. You can click here for more material on digits in digital voltmeters.

Digital voltmeters usually have scales that are 0-0.3v, 0-3v, 0-30v, 0-300v, etc.

You need to learn a little terminology and the reasons for the terminology. Let's take a look at a sample voltmeter. Example E4 Consider a voltmeter built around a 10 bit A/D converter. We will assume the following.

The range of the voltmeter is from 0-3v, and it does DC voltage measurements. It does not measure negative voltages.

Then, with 10 bits we can draw these inferences.

Ten bits will produce 210 intervals. That's 1024 intervals. If there are 1024 intervals over a range of 3v, each interval will be 3/1024 = .00293v. It is easier to compute the displayed voltage if the interval is adjusted to .003v. o That would make the range 0-3.072v. (That's .003 x 1024.) o If you are measuring a voltage that varies around 3v, that would allow you to keep the range the same, but still change the range when the voltage got large enough. Manufacturers like to build in a little "hysteresis" to prevent constant range changes in situations like that and it might be especially hard on auto-ranging meters. If you wanted to measure negative voltages and have the range be from -3v to +3v, you would have intervals of .006v, and the meter would measure from -3.072v to +3.072v. If you wanted to measure voltages on a 0-30v scale, you would probably use a voltage divider or some other way to reduce the voltage by a factor of (exactly) 10 (i.e., multiply it by exactly 0.1) and then use the same converter as on the 0-3v scale.

If we could use a 12 bit A/D, then some conclusions would change.

Twelve bits will produce 212 intervals. That's 4096 intervals. If there are 4096 intervals over a range of 3v, each interval will be 3/4096 = .000732v. It is easier to compute the displayed voltage if the interval is adjusted to .0075v. o That would make the range 0-3.072v - just as it was in the case of the 10 bit converter, o That produces the same advantages as you had with the 10 bit converter. If you wanted to measure negative voltages and have the range be from -3v to +3v, you would have intervals of .0015v, and the meter would measure from -3.072v to +3.072v.

At this point, you should have a good idea about how the number of bits in an A/D converter determines the accuracy of the converter - i.e. the resolution of the converter.

You also need to learn how to determine the resolution of an instrument in the laboratory, and relate the resolution of lab instruments to what you know about the number of bits in a converter. You can click here for an laboratory exercise which leads you through getting the number of bits in the A/D converter inside a voltmeter.

2-Digital-to-analog converter:
Overview:

Ideally sampled signal. A DAC converts an abstract finite-precision number (usually a fixed-point binary number) into a physical quantity (e.g., a voltage or a pressure). In particular, DACs are often used to convert finite-precision time series data to a continually varying physical signal. A typical DAC converts the abstract numbers into a concrete sequence of impulses that are then processed by a reconstruction filter using some form of interpolation to fill in data between the impulses. Other DAC methods (e.g., methods based on delta-sigma modulation) produce a pulsedensity modulated signal that can then be filtered in a similar way to produce a smoothly varying signal. As per the NyquistShannon sampling theorem, a DAC can reconstruct the original signal from the sampled data provided that its bandwidth meets certain requirements (e.g., a baseband signal with bandwidth less than the Nyquist frequency). Digital sampling introduces quantization error that manifests as low-level noise added to the reconstructed signal.

Practical operation:

Piecewise constant output of a conventional practical DAC. Instead of impulses, usually the sequence of numbers update the analogue voltage at uniform sampling intervals.

These numbers are written to the DAC, typically with a clock signal that causes each number to be latched in sequence, at which time the DAC output voltage changes rapidly from the previous value to the value represented by the currently latched number. The effect of this is that the output voltage is held in time at the current value until the next input number is latched resulting in a piecewise constant or 'staircase' shaped output. This is equivalent to a zero-order hold operation and has an effect on the frequency response of the reconstructed signal. The fact that DACs output a sequence of piecewise constant values (known as zero-order hold in sample data textbooks) or rectangular pulses causes multiple harmonics above the Nyquist frequency. Usually, these are removed with a low pass filter acting as a reconstruction filter in applications that require it.

Applications:

A simplified functional diagram of an 8-bit DAC

Audio:
Most modern audio signals are stored in digital form (for example MP3s and CDs) and in order to be heard through speakers they must be converted into an analog signal. DACs are therefore found in CD players, digital music players, and PC sound cards. Specialist standalone DACs can also be found in high-end hi-fi systems. These normally take the digital output of a compatible CD player or dedicated transport (which is basically a CD player with no internal DAC) and convert the signal into an analog line-level output that can then be fed into an amplifier to drive speakers. Similar digital-to-analog converters can be found in digital speakers such as USB speakers, and in sound cards. In VoIP (Voice over IP) applications, the source must first be digitized for transmission, so it undergoes conversion via an Analog-to-Digital Converter, and is then reconstructed into analog using a DAC on the receiving party's end.

Top-loading CD player and external digital-to-analog converter.

Video:
Video sampling tends to work on a completely different scale altogether thanks to the highly nonlinear response both of cathode ray tubes (for which the vast majority of digital video foundation work was targeted) and the human eye, using a "gamma curve" to provide an appearance of evenly distributed brightness steps across the display's full dynamic range - hence the need to use RAMDACs in computer video applications with deep enough colour resolution to make engineering a hardcoded value into the DAC for each output level of each channel impractical (e.g. an Atari ST or Sega Genesis would require 24 such values; a 24-bit video card would need 768...). Given this inherent distortion, it is not unusual for a television or video projector to truthfully claim a linear contrast ratio (difference between darkest and brightest output levels) of 1000:1 or greater, equivalent to 10 bits of audio precision even though it may only accept signals with 8-bit precision and use an LCD panel that only represents 6 or 7 bits per channel. Video signals from a digital source, such as a computer, must be converted to analog form if they are to be displayed on an analog monitor. As of 2007, analog inputs were more commonly used than digital, but this changed as flat panel displays with DVI and/or HDMI connections became more widespread.[citation needed] A video DAC is, however, incorporated in any digital video player with analog outputs. The DAC is usually integrated with some memory (RAM), which contains conversion tables for gamma correction, contrast and brightness, to make a device called a RAMDAC. A device that is distantly related to the DAC is the digitally controlled potentiometer, used to control an analog signal digitally.

Mechanical:
An unusual application of digital-to-analog conversion was the whiffletree electromechanical digital-to-analog converter linkage in the IBM Selectric typewriter.[citation needed]

DAC types:
The most common types of electronic DACs are:

The pulse-width modulator, the simplest DAC type. A stable current or voltage is switched into a low-pass analog filter with a duration determined by the digital input code. This technique is often used for electric motor speed control, but has many other applications as well. Oversampling DACs or interpolating DACs such as the delta-sigma DAC, use a pulse density conversion technique. The oversampling technique allows for the use of a lower resolution DAC internally. A simple 1-bit DAC is often chosen because the oversampled result is inherently linear. The DAC is driven with a pulse-density modulated signal, created with the use of a low-pass filter, step nonlinearity (the actual 1-bit DAC), and negative feedback loop, in a technique called delta-sigma modulation. This results in an effective high-pass filter acting on the quantization (signal processing) noise, thus steering this noise out of the low frequencies of interest into the megahertz frequencies of little interest, which is called noise shaping. The quantization noise at these high frequencies is removed or greatly attenuated by use of an analog low-pass filter at the output (sometimes a simple RC low-pass circuit is sufficient). Most very high resolution DACs (greater than 16 bits) are of this type due to its high linearity and low cost. Higher oversampling rates can relax the specifications of the output low-pass filter and enable further suppression of quantization noise. Speeds of greater than 100 thousand samples per second (for example, 192 kHz) and resolutions of 24 bits are attainable with deltasigma DACs. A short comparison with pulse-width modulation shows that a 1-bit DAC with a simple first-order integrator would have to run at 3 THz (which is physically unrealizable) to achieve 24 meaningful bits of resolution, requiring a higher-order lowpass filter in the noise-shaping loop. A single integrator is a low-pass filter with a frequency response inversely proportional to frequency and using one such integrator in the noise-shaping loop is a first order delta-sigma modulator. Multiple higher order topologies (such as MASH) are used to achieve higher degrees of noise-shaping with a stable topology. The binary-weighted DAC, which contains one resistor or current source for each bit of the DAC connected to a summing point. These precise voltages or currents sum to the correct output value. This is one of the fastest conversion methods but suffers from poor accuracy because of the high precision required for each individual voltage or current. Such high-precision resistors and current sources are expensive, so this type of converter is usually limited to 8-bit resolution or less. The R-2R ladder DAC which is a binary-weighted DAC that uses a repeating cascaded structure of resistor values R and 2R. This improves the precision due to the relative ease of producing equal valued-matched resistors (or current sources). However, wide converters perform slowly due to increasingly large RC-constants for each added R-2R link. The thermometer-coded DAC, which contains an equal resistor or current-source segment for each possible value of DAC output. An 8-bit thermometer DAC would have 255 segments, and a 16-bit thermometer DAC would have 65,535 segments. This is perhaps the fastest and highest precision DAC architecture but at the expense of high

cost. Conversion speeds of >1 billion samples per second have been reached with this type of DAC. Hybrid DACs, which use a combination of the above techniques in a single converter. Most DAC integrated circuits are of this type due to the difficulty of getting low cost, high speed and high precision in one device. o The segmented DAC, which combines the thermometer-coded principle for the most significant bits and the binary-weighted principle for the least significant bits. In this way, a compromise is obtained between precision (by the use of the thermometer-coded principle) and number of resistors or current sources (by the use of the binary-weighted principle). The full binary-weighted design means 0% segmentation, the full thermometer-coded design means 100% segmentation.

DAC performance:
DACs are very important to system performance. The most important characteristics of these devices are:

Resolution: This is the number of possible output levels the DAC is designed to reproduce. This is usually stated as the number of bits it uses, which is the base two logarithm of the number of levels. For instance a 1 bit DAC is designed to reproduce 2 (21) levels while an 8 bit DAC is designed for 256 (28) levels. Resolution is related to the Effective number of bitswhich is a measurement of the actual resolution attained by the DAC. Resolution determines color depth in video applications and audio bit depth in audio applications. Maximum sampling rate: This is a measurement of the maximum speed at which the DACs circuitry can operate and still produce the correct output. As stated in the Nyquist Shannon sampling theorem defines a relationship between the sampling frequency and bandwidth of the sampled signal. Monotonicity: This refers to the ability of a DAC's analog output to move only in the direction that the digital input moves (i.e., if the input increases, the output doesn't dip before asserting the correct output.) This characteristic is very important for DACs used as a low frequency signal source or as a digitally programmable trim element. THD+N: This is a measurement of the distortion and noise introduced to the signal by the DAC. It is expressed as a percentage of the total power of unwanted harmonic distortion and noise that accompany the desired signal. This is a very important DAC characteristic for dynamic and small signal DAC applications. Dynamic range: This is a measurement of the difference between the largest and smallest signals the DAC can reproduce expressed in decibels. This is usually related to resolution and noise floor.

Other measurements, such as phase distortion and jitter, can also be very important for some applications, some of which (e.g. wireless data transmission, composite video) may even rely on accurate production of phase-adjusted signals. Linear PCM audio sampling usually works on the basis of each bit of resolution being equivalent to 6 decibels of amplitude (a 2x increase in volume or precision).

Non-linear PCM encodings (A-law / mu-law, ADPCM, NICAM) attempt to improve their effective dynamic ranges by a variety of methods - logarithmic step sizes between the output signal strengths represented by each data bit (trading greater quantisation distortion of loud signals for better performance of quiet signals)

DAC figures of merit:

Static performance: o Differential nonlinearity (DNL) shows how much two adjacent code analog values deviate from the ideal 1 LSB step.[1] o Integral nonlinearity (INL) shows how much the DAC transfer characteristic deviates from an ideal one. That is, the ideal characteristic is usually a straight line; INL shows how much the actual voltage at a given code value differs from that line, in LSBs (1 LSB steps). o Gain o Offset o Noise is ultimately limited by the thermal noise generated by passive components such as resistors. For audio applications and in room temperatures, such noise is usually a little less than 1 V (microvolt) of white noise. This limits performance to less than 20~21 bits even in 24-bit DACs. Frequency domain performance o Spurious-free dynamic range (SFDR) indicates in dB the ratio between the powers of the converted main signal and the greatest undesired spur. o Signal-to-noise and distortion ratio (SNDR) indicates in dB the ratio between the powers of the converted main signal and the sum of the noise and the generated harmonic spurs o i-th harmonic distortion (HDi) indicates the power of the i-th harmonic of the converted main signal o Total harmonic distortion (THD) is the sum of the powers of all HDi o If the maximum DNL error is less than 1 LSB, then the D/A converter is guaranteed to be monotonic. However, many monotonic converters may have a maximum DNL greater than 1 LSB. Time domain performance: o Glitch energy o Response uncertainty o Time nonlinearity (TNL)