PARALLEL DIGITAL-ANALOG CONVERTERS

Transcription

1 CMOS Analog IC Design Page PARALLEL DIGITAL-ANALOG CONVERTERS CLASSIFICATION OF DIGITAL-ANALOG CONVERTERS

2 CMOS Analog IC Design Page CURRENT SCALING DIGITAL-ANALOG CONVERTERS GENERAL CURRENT SCALING DACS The output voltage can be expressed as V OUT = -R F (I 1 + I 2 + I I N ) where the currents I 1, I 2, I 3,... are binary weighted currents.

3 CMOS Analog IC Design Page BINARY-WEIGHTED RESISTOR DAC Circuit: Comments: 1.) R F can be used to scale the gain of the DAC. If R F = KR/2, then b 1 v OUT = - R F I O = -KR 2 R + b 2 2R + b 3 4R + + b N 2 N-1 R V REF v OUT = -K 2 + b b b N 2 N V REF where b i is 1 if switch S i is connected tov REF or 0 if switch S i is connected to ground. 2.) Component spread value = R MSB R = R LSB 2 N-1 R = 1 2 N-1 3.) Positive attributes: Insensitive to parasitics fast 4.) Negative attributes: Large component spread value Trimming required for large values of N b 1

4 CMOS Analog IC Design Page Nonmonotonic

5 CMOS Analog IC Design Page R-2R LADDER IMPLEMENTATION OF THE BINARY WEIGHTED RESISTOR DAC Use of the R-2R concept to avoid large element spreads: How does the R-2R ladder work? The resistance seen to the right of any of the vertical 2R resistors is 2R. Attributes: Not sensitive to parasitics (the currents through the resistors never changes as S i is varied) Small element spread Resistors made from the same unit (2R consist of two in series or R consists of two in parallel) Not monotonic

6 CMOS Analog IC Design Page CURRENT SCALING USING BINARY WEIGHTED MOSFET CURRENT SINKS Circuit: Operation: v OUT = R 2 (b N I + b N-1 2I + b N-2 4I + + b 1 2 N-1 I) If I = I REF = V REF 2 N R 2, then v OUT = b b b b N-2 2 N-2 + b N-1 2 N-1 + b N 2 N V REF Attributes: Fast (no floating nodes) and not monotonic

7 CMOS Analog IC Design Page VOLTAGE SCALING DIGITAL-ANALOG CONVERTERS GENERAL VOLTAGE SCALING DIGITAL ANALOG CONVERTER Operation: Creates all possible values of the analog output then uses a decoding network to determine which voltage to select based on the digital input word.

8 CMOS Analog IC Design Page BIT VOLTAGE SCALING DIGITAL-ANALOG CONVERTER The voltage at any tap can be expressed as: v OUT = V REF 8 (n 0.5) = V REF (2n 1) 16 Attributes:

9 CMOS Analog IC Design Page Guaranteed monotonic, compatible with CMOS technology, large area if N is large, sensitive to parasitics, requires a buffer, large current can flow through the resistor string.

10 CMOS Analog IC Design Page ALTERNATE REALIZATION OF THE 3-BIT VOLTAGE SCALING DAC

11 CMOS Analog IC Design Page INL AND DNL OF THE VOLTAGE SCALING DAC Uses a worst-case approach. For an n-bit voltage scaling DAC, assume there are 2 n resistors between V REF and ground. Also assume that the resistors are numbered from 1 to 2 n beginning with the resistor connected to V REF and ending with the resistor connected to ground. Integral Nonlinearity The voltage at the i-th resistor from the top is, v i = (2 n -i)r (2 n -i)r + ir V REF where there are i resistors above v i and 2 n -i resistors below v i. For worst case, assume that i = 2 n-1 (midpoint). Define R max = R + R and R min = R - R. The worst case INL is Therefore, INL = v 2 n-1(actual) - v 2 n-1(ideal) Differential Nonlinearity The worst case DNL can be found as DNL = v step (actual) - v step (ideal) Substituting the actual and ideal steps gives, = (R± R)V REF 2 n R = R± R R - R R Therefore, DNL = ± R R - R V REF 2 n R V REF 2 n = ± R V REF R 2 n LSBs INL = or 2 n-1 (R+ R)V REF 2 n-1 (R+ R) + 2 n-1 (R- R) - V REF 2 = R 2R V REF

12 CMOS Analog IC Design Page V REF INL= 2n 2 n R 2R V REF =2n-1 R R 2 n =2 n-1 R R LSBs

13 CMOS Analog IC Design Page EXAMPLE Accuracy Requirements of a Voltage-Scaling digital-analog Converter If the resistor string of a voltage scaling digital-analog converter is a 5 µm wide polysilicon strip having a relative accuracy of ±1%, what is the largest number of bits that can be resolved and keep the worst case INL within ±0.5 LSB? For this number of bits, what is the worst case DNL? Solution From the previous page, we can write that 2 n-1 R R = 2n This inequality can be simplified 2 n 100 which has a solution of n = 6. The value of the DNL for n = 6 is found from the previous page as DNL = ±1 LSBs = ±0.01LSBs 100 (This is the reason the resistor string is monotonic.)

14 CMOS Analog IC Design Page CHARGE SCALING DIGITAL-ANALOG CONVERTERS GENERAL CHARGE SCALING DIGITAL-ANALOG CONVERTER General principle is to capacitively attenuate the reference voltage. Capacitive attenuation is simply: Calculate as if the capacitors were resistors. For example, V out = 1 C V REF = C 1 C 2 C 1 C 1 + C 2 V REF

15 CMOS Analog IC Design Page BINARY-WEIGHTED, CHARGE SCALING DAC Circuit: Operation: 1.) All switches connected to ground during φ 1. 2.) Switch S i closes to V REF if b i = 1 or to ground if b i = 0. Equating the charge in the capacitors gives, which gives V REF C eq = V REF b 1 C + b 2C 2 + b 3C b NC 2 N 1 = C tot v OUT = 2C v OUT v OUT = [b b b b N 2 -N ]V REF Equivalent circuit of the binary-weighted, charge scaling DAC is: Attributes: Accurate Sensitive to parasitics

16 CMOS Analog IC Design Page Not monotonic Charge feedthrough occurs at turn on of switches

17 CMOS Analog IC Design Page INTEGRAL NONLINEARITY OF THE CHARGE SCALING DAC Again, we use a worst case approach. Assume an n-bit charge scaling DAC with the MSB capacitor of C and the LSB capacitor of C/2 n-1 and the capacitors have a tolerance of C/C. The ideal output when the i-th capacitor only is connected to V REF is v OUT (ideal) = C/2i-1 2C V REF = V REF 2 i 2n 2 n = 2n 2 i LSBs The maximum and minimum capacitance is C max = C + C and C min = C - C. Therefore, the actual worst case output for the i-th capacitor is v OUT (actual) = (C± C)/2i-1 2C Now, the INL for the i-th bit is given as V REF = V REF 2 i ± C V REF 2 i = 2n C 2 i ± 2n C 2 i C LSBs INL(i) = v OUT (actual) - v OUT (ideal) = ±2n C 2 i C = 2n-i C C LSBs Typically, the worst case value of i occurs for i = 1. Therefore, the worst case INL is INL = ± 2 n-1 C C LSBs

18 CMOS Analog IC Design Page DIFFERENTIAL NONLINEARITY OF THE CHARGE SCALING DAC The worst case DNL for the binary weighted capacitor array is found when the MSB changes. The output voltage of the binary weighted capacitor array can be written as C eq. v OUT = V (2C-C eq. ) + C REF eq. where C eq represents capacitors whose bits are 1 and (2C - C eq ) represents capacitors whose bits are 0. The worst case DNL can be expressed as DNL = v step (worst case) v OUT ( ) - v OUT ( ) - 1 = - 1 LSBs v step (ideal) LSB The worst case choice for the capacitors is, C 1 =C+ C, C 2 = 1 2 (C- C), C 3 = 1 4 (C- C),...,C n-1 = 1 2 n-2(c- C), C n = 1 2 n-1(c- C), and C term = 1 2 n-1(c- C) n Note that ΣC i + C term = C 2 + C C n-1 + C n + C term = C- C i=2 (v OUT ( ) = C+ C (C+ C)+(C- C) V REF = C+ C 2C V REF and 1 v OUT ( ) = (C- C) -C (C- C) - term (C+ C)+(C- C) V REF = 2 n-1(c- C) V (C+ C)+(C- C) REF = C- C 2C n V REF v OUT ( ) - v OUT ( ) - 1 LSBs = = 2 LSB n C+ C 2C - 2n C- C 2C n - 1 = (2n - 1) C C LSBs

19 CMOS Analog IC Design Page Therefore, DNL = (2 n - 1) C C LSBs

20 CMOS Analog IC Design Page EXAMPLE DNL and INL of a Binary Weighted Capacitor Array DAC If the tolerance of the capacitors in an 8-bit, binary weighted, charge scaling DAC are ±0.5%, find the worst case INL and DNL. Solution For the worst case INL, we get from above that INL = (2 7 )(±0.005) = ±0.64 LSBs For the worst case DNL, we can write that DNL = (2 8-1)(±0.005) = ±1.275 LSBs

21 CMOS Analog IC Design Page EXAMPLE Influence of Capacitor Ratio Accuracy on Number of Bits Use the data of Fig to estimate the number of bits possible for a charge scaling DAC assuming a worst case approach for INL and that the worst conditions occur at the midscale (1 MSB). Solution Assuming an INL of ±0.5 LSB, we can write that INL = ±2 n-1 C C = ± 1 2 C C = 1 2 n. From the data presented in Fig , it is reasonable to assume that the relative accuracy of the capacitor ratios will decrease with the number of bits. Let us assume a unit capacitor of 50 µm by 50 µm and a relative accuracy of approximately ±0.1%. Solving for N in the above equation gives approximately 10 bits. However, the ±0.1% figure corresponds to ratios of 16:1 or 4 bits. In order to get a solution, we estimate the relative accuracy of capacitor ratios as C C N Using this approximate relationship, a 9-bit digital-analog converter should be realizable.

22 CMOS Analog IC Design Page BINARY WEIGHTED, CHARGE AMPLIFIER DAC Attributes: No floating nodes which implies insensitive to parasitics and fast No terminating capacitor required With the above configuration, charge feedthrough will be V error -(C OL /2C N ) V Can totally eliminate parasitics with parasitic-insensitive switched capacitor circuitry but not the charge feedthrough

23 CMOS Analog IC Design Page SUMMARY OF THE PARALLEL DAC PERFORMANCE DAC Type Advantage Disadvantage Current Scaling Fast, insensitive to switch parasitics Large element spread, nonmonotonic Voltage Scaling Monotonic, equal resistors Large area, sensitive to parasitic capacitance Charge Scaling Fast, good accuracy Large element spread, nonmonotonic