Differential analog-to-digital converter

Description of equivalent WO 02067431 A1

DIFFERENTIAL ANALOG-TO-DIGITAL CONVERTER

0002Field of Invention

0003The present invention relates to analog-to-digital converters in general, and to differential analog-to-digital converters in particular.

0004Background of Invention

0005The usefulness of Analog-to-Digital Converters (ADCs) is well known. One type of ADC is known as a Successive Approximation (SSA) ADC. An SSA ADC uses a Digital-to-Analog Converter (DAC) in a feedback loop, in combination with a comparator and Successive Approximation Register (SAR). An SSA ADC first sets a Most Significant Bit (MSB) using the SAR. The comparator then compares the analog input to be converted with the DAC feedback to determine whether the input is larger or smaller than Vi the full scale reference voltage. If the input voltage is greater than Vi the reference voltage the MSB is left unchanged, otherwise it is reset to the opposite state. The analog input voltage is then reduced by the compared Vi reference voltage and compared with V2², or ^lΛ, the reference voltage to set the next MSB. The process is continued until a desired Least Significant Bit (LSB) is set.

0006Traditional SSA ADCs are undesirably prone to introducing errors though, due to the inclusion of both a comparator and DAC. To address this shortcoming, differential SSA ADCs have been proposed wherein two differential inputs are provided. However, many conventional Differential SSA ADCs are relatively costly and complicated in nature. It is an object of the present invention to provide a simplfied differential SSA ADC. Summary of Invention

0007A method for converting a plurality of input signals being indicative of a signal to be converted to a digital output including: setting a plurality of codes each being indicative of a corresponding reference level; and, for each one of the codes, converting the one code to a first analog signal, and summing the first analog signal with a first of the input signals to provide a first summed signal; complementing the one code to provide a complemented code, converting the complemented code to a second analog signal, and summing the second analog signal with a second of the input signals to provide a second summed signal corresponding to the first summed signal; comparing the corresponding first and second summed signals to provide a comparison signal; and, setting at least a portion of the digital output according to the comparison signal.

0008Brief Description of the Fi ures

0009Figure 1 illustrates a block diagram for a differential input analog-to-digital converter according to one aspect of the invention;

0010Figure 2 illustrates a block diagram of a differential input analog-to-digital converter according to another aspect of the invention;

0011Figure 3 illustrates a block diagram of a successive approximation analog-to- digital converter according to yet another aspect of the present invention;

0012Figure 4 illustrates a block diagram of a successive approximation analog-to- digital converter according to yet another aspect of the present invention;

0013Figure 5 illustrates a block diagram of a successive approximation analog-to- digital converter according to yet another aspect of the present invention; Figure 6 illustrates a block diagram of a successive approximation analog-to- digital converter according to yet another aspect of the present invention;

0014Figures 7A and 7B illustrate diagrams of circuits suitable for use as the digital- to-analog converter DAI of Figure 6 according to an aspect of the present invention; Figures 8A and 8B illustrate diagrams of circuits suitable for use as the digital- to-analog converter DA2 of Figure 6 according to an aspect of the present invention;

0015Figures 9A and 9B illustrate diagrams of alternative circuits suitable for use as the digital-to-analog converter DA2 of Figure 6 according to another aspect of the present invention; Figure 10A illustrates a diagram of circuit suitable for use as the digital-to- analog converter DA3 of Figure 6 according to an aspect of the present invention;

0016Figure 10B illustrates a diagram of a circuit suitable for use as the digital-to- analog converter DA4 of Figure 6 according to an aspect of the present invention;

0017and, Figure 11 illustrates a diagram of a circuit suitable for use as the resistor ladder of Figure 6 according to an aspect of the present invention.

0018Detailed Description of the Preferred Embodiment

0019In a single input Analog-to-Digital Converter (ADC), a comparator compares an input signal (Vj„) and a Digital-to-Analog Converter (DAC) output indicative of a reference level to decode digital outputs. In an N-bit Successive Approximation ADC (SSA), the comparator firstly compares Vj_n with a first fixed voltage, such as Vi the Full Scale voltage (FS), to determine a first Most Significant Bit (MSB) to be output a<_N-i), secondly compares Y_m - ¹/2(a(N-i))FS with ^lΛ FS to determine a second MSB a<_N- ₂ , and so on until comparing V_ιn - ¹/2^{(N I)}(a₍N.i))FS, where I is from 1 to (N-l), to decide a Least Significant Bit (LSB) a(₀). For differential SSA ADC's, two input signals are provided, V,_n+ and V_ιn.. The relationship between these inputs is characterized by N_ιn. = FS - V_m+. If the full scale value is normalized for sake of discussion, then V,_n. = 1 - V_ιn+. In the case of analog- to-digital conversion, each input signal subtracts with a digital-to-analog feedback signal to enable the comparator to determine the decoded digital output. Assuming DA(₊) is a digital-to-analog feedback signal corresponding to the V_m+ input and DA₍.₎ is a digital-to-analog feedback signal corresponding to the N_ιn- input, the inputs to the comparator take the form of V_m+ - DA₍₊) and N_m- - DA₍.). The present invention takes advantage of the following determined relationship between DA₍₊₎ and DA₍._j.

0020Referring now to the Figures, like references identify like elements of the invention. Figure 1 illustrates a block diagram of a differential input SSA ADC 5 according to one aspect of the invention. Generally, the ADC 5 includes: a comparator 10 including a (+) input 12, (-) input 14 and output 16; a (+) input 20 for receiving V_m+ and a (-) input 30 for receiving V_ιn.. Coupled to the output 16 is a Successive Approximation Register (SAR) 40. Coupled to an output of the SAR 40 are digital outputs 50 and Digital-to-Analog Converters (DACs) 60, 70. A (+) summing circuit or summer 80 includes a (+) input 82, a (-) input 84 and an output 86. The input 82 is coupled to input 20, while input 84 is coupled to output of DAC 60 and output 86 is coupled to comparator 10 input 12. A (-) summing circuit or summer 90 includes a (+) input 92, a (-) input 94 and an output 96. The input 92 is coupled to input 30, while input 94 is coupled to an output of DAC 70 and output 96 is coupled to comparator 10 input 14. The DAC 60 produces the DA ₊₎ signal while the DAC 70 produces the DA(-) signal. Hence, comparator 10 input 12 receives V_ιn+ - DA₍₊₎ and comparator 10 input 14 receives V_ιn. - DA(.>. DA(₊) and DA(.) normalized with digital codes and voltage amplitude such that in the analog domain they do not exceed 1 in normal operation.

0021The ADC 5 receives the differential signals V_ιn+ and V_ιn. at the inputs 20, 30, i.e.

0022N ₊ - DA₍₊₎ = -[V_ιn. - DA₍.)] (1)

0023By substituting: V_ιn+ + N_ιn. = 1 into Eq. 1, due to normalization, DA₍.₎ = [1 - DA₍₊₎] = [FS - DA₍₊)] in the analog domain, or,

0024DA(.) = 2's complement of DA_(+)ι in the digital domain. (2)

0025In the digital domain, for an Ν-bit ADC,

0026N-\ f \ \^N

0027DA (+) ^a(N-\) « ^{+ a}(N-2) . ^{+ a}(N-3) „ +^{• •} -+^α (1) [p + a (0) w or

00281 = can be 0 or 1, so <img file="WO02067431A1_D0001.tif" />

0029DA₍.) = 1 - DA (+)

0030= 1 - <img file="WO02067431A1_D0002.tif" />

0031<img file="WO02067431A1_D0003.tif" /> = [inverse of DA(₊)] + <img file="WO02067431A1_D0004.tif" />

0032(3)

0033Thus, for a case as DA₍₊₎ = 0000 0000 0000 0000, the lowest possible voltage in a 12-bit ADC, DA₍._} = 1111 1111 1111 1111 + 0000 0000 0000 0001 = 1 0000 0000 0000 0000, the highest full range voltage, remembering that the 2's complement of a binary number N = (1 's complement of N) + I_LSB- And, for a case where DA₍₊₎ = 111 1 1111 1111 1111, the highest code which is one LSB below the full voltage range in a 12-bit ADC, DA₍.) = 00000000 0000 0001, one LSB above the lowest voltage.

0034Thus in any N-bit ADC according to the present invention, using Eq. 2, the ADC circuit of Figure 1 can be realized as is illustrated in Figure 2 using a simple 2's complementary conversion circuit at the output of SAR 40 in combination with DAC 70 to provide the DA(.) signal according to an aspect of the present invention. Referring now also to Figure 2, there is shown a block diagram for a differential input SSA ADC 100 according to an aspect of the invention. Still referring to Figure 2, there is shown the ADC of Figure 1 now also illustrating a 2's complementary conversion circuit 110 coupled between the SAR 40 and DAC 70. Thus, DA which is the 2's complement of DA(₊), is provided as an input to the DAC 70. The 2's complementary circuit 110 has a carry out bit as the MSB, i.e. is has (N+l) bits. The relationship between DA₍₊₎ and DA₍.) in Eq. 2 is valid for differential

0035ADC's using different types of digital-to-analog conversion schemes. According to another aspect of the present invention, the digital codes supplied at the output of the SAR 40 are decomposed into groups of DACs and then summed through a network of ratio capacitors and resistor ladders to achieve a correct weighting for the individual components and analog voltage level. Referring now also to Figure 3, there is shown a 10-bit SSA ADC 200 block diagram according to an aspect of the present invention. Like elements to those described with reference to the previous figures will not be again described. The ADC 200 illustrated therein uses pseudo differential inputs Ni

003620 and V_INR 30', where Vι_N 20 is a real analog input signal to be converted and V_{J R} 30' is a DC or zero value at the lowest analog input voltage.

0037Referring still to Figure 3, the DAC 60 is decomposed into four (4) smaller DACs 62: DAj, DA , DA₃ and DA₄. DAj processes four bits, DA processes four bits plus one offset bit, DA₃ processes 1 bit and DA₄ processes 1 bit according to another aspect of the invention. Further, adder 210 is interposed between SAR 40 and digital outputs 50.

0038Referring now to Figure 4, there is shown a 10-bit SSA ADC 220 according to another aspect of the present invention. Again, like elements to those described with reference to the previous figures will not be again described. Separate decoding circuits 62 and 72" are provided for the differential inputs 20, 30, respectively. DA₍₊₎ is again decomposed into four DACs 62: DAι₍₊₎, DA₂₍₊₎, DA ₍₊₎ and DA₄₍₊₎, analogously to the SSA 200 of Figure 3. DA₍.) is also decomposed into four DACs 72: DAi(-), DA₂(-), DA₃(.) and DA₄(-). Again, DA₍.) is the 2's complementary input of DA₍₊) (See EQ.2).

0039Still referring to Figure 4, an offset bit D_4S is added to DA₍₊₎, such that DA'₊₎ = DA(₊) + D₄s. Equation (2) remains applicable, such that DA -) = 1 - DA'₍₊₎ = 1 - [DA(₊) + D₄s] = [1 - DA(₊)] - D₄s. DAi(₊₎ is a four bit decoder, DA₂(₊) is a five bit decoder, while DA₃₍₊) and DA₄(₊) are one bit decoders. DA₁₍₊) receives the four MSBs: D₉, D₈, D₇, D₆ output from SAR 40. DA₂₍₊₎ receives the next four MSBs: D₅, D , D₃, D and the offset bit D₄s output from SAR 40. DA₃₍₊₎ receives the second LSB Di, while DA ₍₊₎ receives the LSB Do output from SAR 40. DA₁₍-₎ is a 5 bit decoder, DA₂(.) is a 4 bit decoder, while DA₃₍.) and DA₄₍-₎ are one bit decoders. DAi₍-₎ receives the four MSBs: D₉, D₈, D₇, D₆ and bit D₎₀ output from SAR 40. DA₂₍.₎ receives the next four MSBs: D₅, D₄, D₃, D₂ output from SAR 40. DA₃₍.₎ receives the second LSB Di, while DA₄₍.₎ receives the LSB D₀ output from SAR 40. Adder 210 serves to account for offset bit D _S being parsed from the output of SAR 40.

0040Referring now to Figure 5, there is shown an SSA ADC 230 according to yet another aspect of the present invention. Again, like elements to those described with reference to the previous figures will not be again described. A single decoder circuit is used therein for driving the DA₍₊₎ and DA₍.) DACs 62, 72. Basically, the DA₍₊> DAC is decomposed into four DACs 62: DAι_W, DA₂(₊), DA₃₍₊₎ and DA₄₍₊₎ 62'. In mathematical expression this yields:

0041DA₍₊₎ = (39,88^7,86,0,0,0,0,0,0) + (0,0,0,0,85,-14,8₃,82,0,0) + (0,0,0,0,0,0,0,0,8,, 0) +

0042(0,0,0,0,0,0,0,0,0,8o)

0043= GP_u+) + GP₂₍₊₎ + GP₃₍₊₎ +GP₄₍₊₎ The 2's complement of DA(₊) is DA₍.) (See EQ.2), and DA₍.) = 1 - DA₍₊₎. Hence, DA(-) = (l-a₉,l-a₈,l-a₇, 1-86,0,0,0,0,0,0) + (0,0,0,0, l-a₅,l -a l-a₃,l-a₂,0,0) + (0,0,0,0,0,0,0,0, 1 -8_! ,0) + (0,0,0,0,0,0,0,0,0, 1 -ao) + (0,0,0,0,0,0,0,0,0, 1 ) (4)

0044= GP_I(.) + GP₂₍.) + GP₃₍.) +GP₄₍.)

0045Where ao = 0, GP₄₍.) = (0,0,0,0,0,0,0,0,1,0). So, GP₄₍.₎ has a range of (0,0,0,0,0,0,0,0,0,1) to (0,0,0,0,0,0,0,0,1,0). In the DA₄ decoder, the ao = 0 selects V = 0 for DA₄₍₊₎ and V = V4(l,0) for D- ). Further, ao = 1 selects V = V4(0,l) for DA4(+) and DA4(-). Thus, the same decoder can be used for both DA4(+) and DA4(- ). It should be noted that V4(l,0) is twice the value of V4(0,l), and V4(0,0) is 0V. Further, in this case V4(0,l) is 1/1024 the full voltage range of the ADC.

0046Where ao=0 and a₁₌0, GP₃₀ = (0,0,0,0,0,0,0,1,0,0). Where ao=0, a₁₌l, GP₃₍.₎ = (0,0,0,0,0,0,0,0,1,0). Accordingly, GP₃₍.) ranges from (0,0,0,0,0,0,0,0,1,0) to

0047(0,0,0,0,0,0,0,1,0,0). In the DA₃ decoder, the aι=0 selects V=0 for DA₃₍₊₎ and V=V3(1,0) for DA₃₍.); while a,=l selects V=V3(0,1) for DA₃₍₊₎ and DA₃₍.,. Thus, a same decoder can be used for both DA₃₍₊₎ and DA₃₍.). V3(l,0) is twice N3(0,l), and V3(0,0) is 0V and V3(0,l) is 1/512 the full voltage range of the ADC. Where ao=aι=0, and a₂=a₃=a4=a₅=0, GP_2ω = (0,0,0,1,0,0,0,0,0,0). Where ao=0, aj=0, and a₂=a₃=a₄=a₅=l, GP₂₍.) = (0,0,0,0,0,0,0,1,0,0). Hence, GP₂₍.₎ ranges from (0,0,0,0,0,0,0,1,0,0) to (0,0,0,1,0,0,0,0,0,0). As DA₂ is a four bit decoder, at a₂=a₃=a =a₅=0, the decoded output selects V=0 for DA₂₍₊₎ and V=V2( 1,0,0,0,0) for DA ₍.₎. While at a₂=a₃=a₄=a₅=l, the decoded output selects V=V2(1,1,1,1) for DA₂₍₊₎ and V=V2(0,0,0,1) for DA₂(.). Thus, a same decoder can be used for both DA₂₍₊₎ and

0048DA₂₍.₎. V2( 1,0,0,0,0) is 16 times the value of V2(0,0,0,l). V2( 1,1, 1,1) is 15 times the value of V2(0,0,0,l), and V2(0,0,0,0) is 0V. V2(0,0,0,l) is 1/256 the full scale voltage range of the ADC.

0049It should be noted that when an offset bit a₄s is added into DA₂₍₊₎, the carry bit from this summation does not change the original DAι₍₊₎, So, the new DA_2'(+) has the same maximum value of (1,0,0,1,1) if a₄s=(0, 1,0,0). The new range of new DA₂-₍₊₎ in this particular case is ( 1 ,0, 1 ,0,0) instead of ( 1 ,0,0,0,0). Hence, DA₂χ.) = [(1,0,1,0,0) - DA₂₍₊₎ - ∑ s]

0050= [(1,0,0,0,0) - DA₂₍₊₎] + (0,0,0,0) - a-js = DA₂₍.) + (0,1,0,0) - a-js, where -14s is a simplified expression for (0^,0,0). Thus, where a^ = (0,1,0,0), DA₂χ.) = DA₂₍.), DA₂χ₊₎ = DA₂₍₊₎ + (0,1,0,0). And, where - _s = (0,0,0,0), DA₂χ.) = DA₂₍.) +(0,1,0,0), and DA_{2 +)} = DA₂₍₊₎. Thus, a decoding circuit used in connection with DA₂(₊) with an offset bit can be used for DA₂'(₊) and DA₂χ.) as well. In this case, the decoded output of (0,0,0,0) in DA₂χ₊> selects V=0, and selects V=V2( 1,0, 1,0,0) in DA₂χ_>. The decoded output of (1,0,0,1,1) selects V=N2( 1,0,0, 1,1) for DA_2χ+) and selects V=V2(0,0,0,1) for DA₂χ->. V2(l,0,l,0,0) is 20 times the value of V2(0,0,0,l). V2(l,0,0,l,l) is 19 times the value of V2(0,0,0,l) and V2(0,0,0,0) is ON. V2(0,0,0,l) is 1/256 the full scale voltage range of the ADC.

0051Where a0=al=a2=a3=a4=a5=0, and a6=a7=a8=a9=0, GPl(-) = (1,1,1,1,1,1,1,1,1,1) + (0,0,0,0,0,0,0,0,0,1) = (1,0,0,0,0,0,0,0,0,0,0), an 1 1 -bit code. Where a0=al=a2=a3=a4=a5=0 and a6=a7=a8=a9=l, GPl(-) = (0,0,0,1,0,0,0,0,0,0). Accordingly, Gpl(-) ranges from (0,0,0,1,0,0,0,0,0,0) to (1,0,0,0,0,0,0,0,0,0,0). DAI uses a four-bit decoder, where a6=a7=a8=a9=0 the decoded output selects V=0 for

0052DA1(+) and V=V 1(1, 0,0,0,0) for DAl(-). Where a6=a7=a8=a9=l, the decoded output selects V=N1(1, 1,1,1) for DA1(+) and V=V1(0,0,0,1) for DAl(-). Thus, a same decoder can be used for both DA1(+) and DAl(-). VI (1,0,0,0,0) is 1/16 the full scale voltage range of the ADC. Referring now also to Figure 6, there is shown a 10-bit SSA ADC 250 according to yet another aspect of the present invention. Again, the ADC 250 generally includes comparator 10, (+) differential analog input 20, (-)differential analog input 30, SAR 40, adder 210 and digital outputs 50. In response to the comparator 10, SAR 40 provides a 10-bit digital code, having an MSB D_9, LSB D₀ and offset bit D_4S. Digital-to-analog conversion is performed by four DACs 62: DA_t, DA₂, DA₃ and DA . DAj provides DA_{1 +}) and DAj₍.) using the four MSBs supplied by the SAR 40, i.e. D₉, D₈, D₇ and D₆. DA₂ provides DA₂₍₊₎ and DA₂₍.) using the next four MSBs supplied by the SAR 40, i.e. D₅, D₄, D₃ and D₂, and the offset bit D_4S. DA₃ and DA₄ provide DA₃(₊>, DA₃₍.) and DA₄(₊) and DA₄(.) using the two LSBs supplied by the SAR 40, i.e. Di, D₀. A resistor ladder 260 is provided and uses a reference voltage V_LHF and span voltage V_RHF to supply a plurality of voltages Va₍o-i₆₎ and Vb(o-₂o> as will be described in greater detail with regard to Figure 11. Still referring to Figure 6, capacitors 252 serve to appropriately weight DA^_+j, DA₂₍₊₎, DA₃₍₊₎, DA_{4 +}), DA|₍.), DA₂(.), DA₃₍.) and DA₄(.) as they are supplied to the comparator

005310. In the illustrated 10-bit SSA ADC of Figure 6, if the capacitors 252 corresponding to DA₃(₊), DA₄(₊), DA₃₍.) and DA₄₍.₎ have a given capacitance C, the capacitors 252 corresponding to DA₂₍₊) and DA₂(.) should each have a capacitance fours time the given value, or 4C while the capacitors 252 corresponding to DAι₍₊₎ and DA^-) should each have a capacitance eight times the given value of 8C.

0054Referring now to Figure 7A also, there is shown a circuit 255 for generating DA₍₁₊). Generally, the circuit 255 includes a 4-to-16 decoder 270 which receives the four MSBs provided by the SAR 40 (FIG. 6), i.e. D₉, D₈, D₇ and D₆ and provides 16 pairs of outputs, T₍₀-ι₅), TN(₀-ι₅) in response thereto. The circuit 255 further includes a plurality of switches 256 responsive to the pairs of signals T and TN output from the 4

0055- 16 decoder 270 to supply corresponding ones of supplied voltages Vao - Vai₆ as DAi₍₊₎. The following Table- 1 illustrates which of the voltages Va o-ι₅₎ are provided responsively to T(o-ι₅) and TN(o-ιs) by the switches 256.

0056TABLE-1 - DAι₍₊₎<img file="WO02067431A1_D0005.tif" />

0057Thus, a first switch 256', uses signals T₍₀) and TN(o) to selectively supply voltage Va₍₀₎

0058as DA₁₍₊).

0059Referring now also to Figure 7B, there is shown a circuit 280 for generating signal DA^.₎. The circuit 280 includes a plurality of switches 285 also responsive to the pairs of signals T and TN output from the 4 - 16 decoder 270 (Fig. 7 A) to supply corresponding ones of voltages Vao - Vaj₆ as DAj₍.). The following Table-2 illustrates which of voltages Va(o.₁₅> are provided responsively to T₍₀-ι₅) and TN(o-ι₅) by the switches 285.

TABLE-2 - DA K-)

0061<img file="WO02067431A1_D0006.tif" /><img file="WO02067431A1_D0007.tif" />

0062Thus, one of the switches 280', uses signals T₍o> and TN(o> to selectively supply voltage Va₍₁₆₎ as signal DA₁₍-).

0063Referring now to Figure 8A, there is shown a circuit 290 for generating DA ₍₊₎. Generally, the circuit 290 includes a 5-to-20 decoder 300 which receives the four next MSBs provided by the SAR 40 (FIG. 6), i.e. D₅, D₄, D₃ and D₂, as well as the offset bit D₄s, and provides 20 pairs of outputs, T o-ι₉ TN₍₀.₁₉). The circuit 290 includes a plurality of switches 295 responsive to these pairs of signals T and TN output from the 5 - 20 decoder 300 to supply corresponding ones of voltages Vbo - Vb₁₉ as DA₂₍₊). The following Table-3 illustrates which of voltages Vb(o-₂o), are provided responsively to T(o_₁ ) and TN ₀-ι₉) by the switches 295.

0064TABLE-3 - DA 2(+)

0065<img file="WO02067431A1_D0008.tif" /><img file="WO02067431A1_D0009.tif" />

0066Thus, a first switch 295' uses signals T₍o) and TN(o> to selectively supply voltage Vb₍₀)

0067as DA₂₍₊₎.

0068Referring now to Figure 8B, there is shown a circuit 310 for generating DA₂₍.₎. Generally, the circuit 310 includes a plurality of switches 315 responsive to these pairs of signals T and TN output from the 5 - 20 decoder 300 to supply corresponding ones of voltages Vb₀ - Vb₂o as DA₂₍.). The following Table-4 illustrates which of voltages Vb₍o-ι ) are provided as DA₂(.) responsively to T(o-₁₉) and TN(o-ι₉) by the switches 315.

0069TABLE-4 - DA_{2( )}

0070<img file="WO02067431A1_D0010.tif" /><img file="WO02067431A1_D0011.tif" />

0071Thus, a switch 315' uses signals T₍o> and TN₍o) to selectively supply voltage Vb₍₂₀₎.

0072Referring now to Figures 9A and 9B, therein is illustrated circuits 290' and 310' according to another aspect of the present invention. Therein a 5 - 20 decoder 300' is used to provide signals T₍₀-ι₉) in response to input of the next four MSBs, i.e. D₅, D₄, D₃ and D₂, and the offset bit D₄s. A plurality of transistors 296 are used to selectively provide voltages Vb(o-ι₉) as DA₂₍₊) and <img file="WO02067431A1_D0012.tif" /> in response to T₍₀-₂o). The following Tables 5 and 6 show which ones of signals T(₀.₂o) are used to selectively activate the transistors 296 to provide DA₂₍₊) and DA₂(-), respectively.

0073TABLE S - DA 2(+)

0074<img file="WO02067431A1_D0013.tif" /><img file="WO02067431A1_D0014.tif" />

0075TABLE-6 - DA₂₍.)

0076<img file="WO02067431A1_D0015.tif" />

0077Referring now to Figure 10A, there is shown a circuit 320 for providing DA (₊) and DA₃₍-). Still referring to Figure 10A, the circuit 320 receives bit Dj via an input. The input is coupled to an inverter 321 and a first input of a NOR gate 322. The inverter 321 outputs to a first input of a second NOR gate 323. The output of the NOR gate 322 is provided as a second input for NOR gate 323. Likewise, the output of NOR gate 323 is provided as a second input for NOR gate 322. In other words, the NOR gates 322, 323 are cross-coupled similarly as for a conventional S-R latch. The output of NOR gate 322 is also coupled to a gate input for a transistor 324 and a gate input for a transistor 326. The output of NOR gate 323 is coupled to a gate input of a transistor 325 and a gate input of a transistor 327. The transistor 324 is provided on a source input with Vb₍o> and transistor 325 is provided on a source input with Vb₍₂₎. The drains of transistors 324 and 325 are coupled to a common node to provide DA₃₍₊₎. Accordingly, Vb(o> and Vb(₂) are selectively provided as DA₃₍₊) dependently upon Di. Similarly, transistor 326 is provided on a source input with Vb₍₄> and transistor 327 is provided on a source input with Vb(₂). The drains of transistors 326 and 327 are coupled to a common node to provide DA₃9. Accordingly, Vb₍₄₎ and Vb₍₂₎ are selectively provided as DA₃₍.) dependently upon Dj. Referring now to Figure 10B, there is shown a circuit 330 for providing DA₄₍₊₎ and DA-K-₎. Still referring to Figure 10B, the circuit 330 receives Do via an input. The input is coupled to an inverter 331 and a first input of a NOR gate 332. The inverter

0078331 outputs to a first input of a second NOR gate 333. The output of the NOR gate

0079332 is provided as a second input for NOR gate 333. Likewise, the output of NOR gate 333 is provided as a second input for NOR gate 332. In other words, the NOR gates 332, 333 are cross-coupled similarly as for a conventional S-R latch. The output of NOR gate 332 is coupled to a gate input for a transistor 334 and a gate input for a transistor 336. The output of NOR gate 333 is coupled to a gate input of a transistor 335 and a gate input of a transistor 337. The transistor 334 is provided on a source input with Vbrø and transistor 335 is provided on a source input with Vb₍i₎. The drains of transistors 334 and 335 are coupled to a common node to provide DA₄₍₊₎. Accordingly, Vb(₀> and Vb(D are selectively provided as DA_{4 +)} dependently upon Do. Similarly, transistor 336 is provided on a source input with Vb₍₂₎ and transistor 337 is provided on a source input with Vb₍₁₎. The drains of transistors 336 and 337 are coupled to a common node to provide DA-κ-). Accordingly, Vb₍₂₎ and Vb₍i₎ are selectively provided as DA ₍.) dependently upon D₀.

0080Referring finally to Figure 11, there is shown the resistor ladder 260 discussed with regard to Figure 6. The resistor ladder 260 provides the voltages Va₍₀-i₆₎ and Vb(o-₂₀) as have been discussed. The resistor ladder 260 includes two serially- connected resistor ladders 261, 262. These ladders 261, 262 are cross-connected to reduce resistivity non-uniformity due to fabrication, for example. Each resistor ladder 261, 262, is divided into 16 main sections Va₀-i₆) by resistors 263. Between Vao and Vai, each ladder is subdivided into 8 sections Vb(o-₈). Between Vaj and Va₂, each ladder is subdivided into another 8 sections Vb( -j,₆). Between Va₂ and Va₃, each ladder is subdivided into another 5 sections Vb(₁₇-₂₀). Each Vb section is 1/128 of the full range for the analog signal input to an ADC utilizing the ladder 260. The resistors that make up the ladder 260 are preferably all about the same value, for example Ik ohm, within a degree of accuracy of about 1 ohm, for example. Although the invention has been described and pictured in a preferred form with a certain degree of particularity, it is understood that the present disclosure of the preferred form, has been made only by way of example, and that numerous changes in the details of construction and combination and arrangement of parts may be made without departing from the spirit and scope of the invention as hereinafter claimed. It is intended that the patent shall cover by suitable expression in the appended claims, whatever features of patentable novelty exist in the invention disclosed.