n analog-to-digital converter (ADC) acceptsan analog input—a voltage or a current—and converts it to a digital value that can beread by a microprocessor. Figure 1 shows asimple voltage-input ADC. This hypotheti-cal part has two inputs: a reference and thesignal to be measured. It has one output, an 8-bit digitalword that represents the input value. The reference voltage is the maximum value that theADC can convert. Our example 8-bit ADC can convert valuesfrom 0V to the reference voltage. This voltage range is divid-ed into 256 values, or steps. The size of the step is given by:Vref/256where Vrefis the reference voltage. The step size of the con-verter defines the converter’s resolution. For a 5V refer-ence, the step size is:5V/256 = 0.0195V or 19.5mVOur 8-bit converter represents the analog input as a digitalword. The most significant bit of this word indicateswhether the input voltage is greater than half the reference(2.5V, with a 5V reference). Each succeeding bit representshalf the range of the previous bit. Table 1 illustrates this point. Adding the voltages corre-sponding to each set bit in 0010 1100, we get:.625 + .156 + .078 = .859 voltsThe resolution of an ADC is determined by the refer-ence input and by the word width. The resolution definesthe smallest voltage change that can be measured by theADC. As mentioned earlier, the resolution is the same as thesmallest step size, and can be calculated by dividing the ref-erence voltage by the number of possible conversion values. 10 MAY 2001 Embedded Systems ProgrammingSTUART BALLfeatureAnalog-to-DigitalConvertersThe usual method of bringing analog inputs into a microprocessor is to use an analog-to-digital converter (ADC). Here are some tips for selecting such a part and calibrating it to fit your needs.ARETURNEmbedded Systems Programming MAY 2001 11For the example we’ve been usingso far, the resolution is 19.5mV. Thismeans that any input voltage below19.5mV will result in an output of 0.Input voltages between 19.5mV and39mV will result in an output of 1.Between 39mV and 58.6mV, the out-put will be 2. Resolution can be improved byreducing the reference input.Changing that from 5V to 2.5V gives aresolution of 2.5/256, or 9.7mV.However, the maximum voltage thatcan be measured is now 2.5V insteadof 5V.The only way to increase resolutionwithout reducing the range is to usean ADC with more bits. A 10-bit ADChas 210, or 1,024 possible outputcodes. So the resolution is 5V/1,024,or 4.88mV; a 12-bit ADC has a 1.22mVresolution for this same reference. Types of ADCsADCs come in various speeds, use dif-ferent interfaces, and provide differ-ing degrees of accuracy. The mostcommon types of ADCs are flash, suc-cessive approximation, and sigma-delta.Flash ADCThe flash ADC is the fastest type avail-able. A flash ADC uses comparators,one per voltage step, and a string ofresistors. A 4-bit ADC will have 16 com-parators, an 8-bit ADC will have 256comparators. All of the comparatoroutputs connect to a block of logicthat determines the output based onwhich comparators are low and whichare high.The conversion speed of the flashADC is the sum of the comparatordelays and the logic delay (the logicdelay is usually negligible). FlashADCs are very fast, but consume enor-mous amounts of IC real estate. Also,because of the number of compara-tors required, they tend to be powerhogs, drawing significant current. A10-bit flash ADC may consume half anamp. A variation on the flash converter isthe half-flash, which uses an internaldigital-to-analog converter (DAC) andsubtraction to reduce the number ofinternal comparators. Half-flash con-verters are slower than true flash con-verters but faster than other types ofADCs. We’ll lump them into the flashconverter category.Successive approximation converter A successive approximation convert-er uses a comparator and countinglogic to perform a conversion. Thefirst step in the conversion is to see ifthe input is greater than half the ref-erence voltage. If it is, the most sig-nificant bit (MSB) of the output isset. This value is then subtractedfrom the input, and the result ischecked for one quarter of the refer-ence voltage. This process continuesuntil all the output bits have been setADCs come in various speeds, use different interfaces, and provide differing degrees of accuracy. The most common types of ADCs are flash, successive approximation, and sigma-delta.RUPERT ADLEYFIGURE 1 Simple ADCTABLE 1 Example conversion, on an 8-bit ADCBit: Bit 7 Bit 6 Bit 5 Bit 4 Bit 3 Bit 2 Bit 1 Bit 0Volts: 2.5 1.25 .625 .3125 .156 .078 .039 .0195OutputValue: 0 0101100.4.6.81Reference Voltage (Vref).232 64 96 128Output Value (Decimal)160 192 224 255InputOutput Bits (8)Reference Voltage (Vref)or reset. A successive approximationADC takes as many clock cycles asthere are output bits to perform aconversion. Sigma-deltaA sigma-delta ADC uses a 1-bit DAC, fil-tering, and oversampling to achievevery accurate conversions. The conver-sion accuracy is controlled by the inputreference and the input clock rate.The primary advantage of a sigma-delta converter is high resolution. Theflash and successive approximationADCs use a resistor ladder or resistorstring. The problem with these is thatthe accuracy of the resistors directlyaffects the accuracy of the conversionresult. Although modern ADCs usevery precise, laser-trimmed resistornetworks, some inaccuracies still per-sist in the resistor ladders. The sigma-delta converter does not have a resis-tor ladder but instead takes a numberof samples to converge on a result.The primary disadvantage of thesigma-delta converter is speed.Because the converter works by over-sampling the input, the conversiontakes many clock cycles. For a givenclock rate, the sigma-delta converter isslower than other converter types. Or,to put it another way, for a given con-version rate, the sigma-delta converterrequires a faster clock.Another disadvantage of the sigma-delta converter is the complexity ofthe digital filter that converts the dutycycle information to a digital outputword. The sigma-delta converter hasbecome more commonly availablewith the ability to add a digital filter orDSP to the IC die.ADC comparisonFigure 2 shows the range of resolu-tions available for sigma-delta, succes-sive approximation, and flash convert-ers. The maximum conversion speedfor each type is shown as well. As youcan see, the speed of available sigma-delta ADCs reaches into the range ofthe successive approximation
View Full Document
Unlocking...