Data driver systems with programmable modes

[0002] 1. Field of the Invention

[0003] The present invention relates generally to data drivers and, more particularly, to data drivers for signal conditioning systems.

[0004] 2. Description of the Related Art

[0005] Conventional complementary metal-oxide semiconductor (CMOS) logic circuits facilitate single-ended transmission of data. For example, a basic building block of CMOS logic is a CMOS inverter which includes first and second CMOS transistors that are serially-coupled between first and second voltage rails (e.g., between V_DDand ground). When the gates of the first and second transistors are successively coupled to the first and second rail voltages, the inverter output successively provides second and first rail voltages as output signals. Among its other advantages, CMOS logic uses substantially less power than many other systems (e.g., transistor-transistor logic (TTL) and emitter-coupled logic (ECL)). In addition, MOS logic circuits can directly drive TTL circuits.

[0006] Low voltage differential signaling (LVDS) is a data signaling technology that provides differential current signals for high-rate data transmission. Because noise is generally equally coupled onto differential signal paths, it is substantially rejected by a remote receiver which differentially receives the LVDS signals. Accordingly, LVDS differential transmission is less susceptible to common-mode noise than single-ended transmission systems. Because LVDS drivers generate low-level differential current signals, their power consumption is reduced from conventional single-ended transmission systems and is almost flat regardless of the data rate.

[0007] LVDS technology is currently standardized in the ANSI/TIA/EIA-644 Standard of the Telecommunications Industry Association/Electronics Industries Association (TIA/EIA) and in the IEEE 1596.3 Standard of the Institute for Electrical and Electronics Engineering (IEEE).

[0008] Because of their different needs, some potential users of signal conditioning circuits (e.g., analog-to-digital converters (ADCs)) prefer output signals that are compatible with CMOS/TTL circuits and others prefer output signals that are compatible with LVDS circuits. In order to enhance their attractiveness to a large number of customers, signal conditioning circuits are, therefore, preferably configured to provide both CMOS and LVDS drive signals.

[0009] However, current ADCs often provide a large number of digital bits (e.g., on the order of 16) and are generally fabricated as integrated circuits with limited package dimensions (e.g., on the order of 15 millimeters). Accordingly, it has become increasingly difficult to provide the large number of integrated-circuit pins that are required to provide both CMOS and LVDS drive signals.

[0010] The present invention is directed to data driver systems that have programmable modes of operation to thereby facilitate selection of output signal forms and reduction of output ports.

[0011] In an exemplary embodiment, rail-to-rail and LVDS drivers share output ports and are combined with a multiplexer that selectively configures them in driver and high output-impedance states.

[0012] The novel features of the invention are set forth with particularity in the appended claims. The invention will be best understood from the following description when read in conjunction with the accompanying drawings.

[0013] FIG. 1 is a block diagram of a data driver system embodiment of the present invention;

[0014] FIG. 2 is a schematic of another data driver system embodiment; and

[0015] FIG. 3 is a block diagram of ADC systems that are realized with the driver systems of FIGS. 1 and 2.

[0016] Data driver system embodiments of the present invention are shown in FIGS. 1 and 2 and ADC system embodiments are shown in FIG. 3. These embodiments effectively reduce pin count for integrated-circuit versions of communication systems (e.g., signal conditioning systems) by sharing pins between different drivers and selectively configuring the drivers in driver and high output-impedance states.

[0017] In particular, FIG. 1 illustrates a programmable driver system embodiment 20 that provides data output signals at first and second data output ports (e.g., integrated-circuit pads) 21 and 22. The system includes first and second rail-to-rail drivers 25 and 26, an LVDS driver 28 and a programmable data mulitiplexer 30. The first and second rail-to-rail drivers 25 and 26 are respectively coupled to the first and second data output ports 21 and 22 and the LVDS driver 28 is differentially coupled to the first and second data output ports 21 and 22.

[0018] The data multiplexer 30 can be programmed (e.g., in response to a command signal 31) between rail-to-rail and LVDS modes of operation. In the rail-to-rail mode, the multiplexer 30 respectively guides first and second data input signals D₁and D₂from first and second input ports 33 and 34 to the first and second rail-to-rail drivers 25 and 26 and biases the LVDS driver 28 in a high output-impedance state. In the LVDS mode, the multiplexer 30 differentially guides a third data input signal (e.g., the first data input signal D₁) to the LVDS driver 28 and biases each of the first and second rail-to-rail drivers 25 and 26 in high output-impedance states.

[0019] Essentially, the multiplexer 30 controls tristate modes of the rail-to-rail and LVDS drivers. The programmable driver system 20 is especially suited for use in high-speed integrated-circuit signal conditioning circuits (e.g., ADCs) in which a large number of output digital bits and limited package dimensions make it difficult to provide the large number of integrated-circuit pins that are required to provide more than one output-signal format.

[0020] In an exemplary signal conditioning application of the programmable driver system 20, it processes output signals in either its LVDS or rail-to-rail modes to a common set of integrated-circuit output pins. In the LVDS mode, the system would generally drive a resistive load 35 (e.g., a resistor or a properly biased transistor) that is coupled across the first and second output ports 21 and 22. Because of its high-speed capabilities, the LVDS driver 28 can typically operate at the high-speed encoding rates of modern signal conditioning circuits. In response to the first data input signal D₁, the LVDS driver 28 would therefore generate differential currents 41 and 42 through the load 35 to realize an output signal in which N, N+1 - - - N+n data signals occur at the system encoding rate as indicated by data signal 44 in FIG. 1.

[0021] In contrast, the single-ended rail-to-rail voltage signals of the first and second rail-to-rail drivers 25 and 26 are typically generated across high-impedance loads and generally cannot follow high-speed encoding rates. In response to first and second data input signals D₁and D₂, they would therefore typically generate parallel data signals 45 and 46 at the first and second output ports 21 and 22 in which data signals N, N+2 and so on are generated at the first data output port 21 and N+1, N+3 and so on are generated at the second data output port 22.

[0022] That is, the rail-to-rail data signals are generated at one half the speed of the LVDS signals but are paralleled to compensate for this speed difference. Alternatively, first and second data input signals D₁and D₂could be altered to generate interleaved data signals 45 and 47 of FIG. 1 wherein the signal 47 is delayed respective to the signal 45. It is apparent therefore, that the programmable driver system 20 can process a respective digital bit in a high-speed integrated-circuit signal conditioning circuit and provide this bit across shared integrated-circuit pins in both rail-to-rail and LVDS operational modes.

[0023] FIG. 2 illustrates another programmable driver system embodiment 60 which is coupled between the first and second input ports 33 and 34 and first and second data output ports 21 and 22 of FIG. 1. The system 60 includes first and second rail-to-rail drivers 65 and 66, an LVDS driver 68 and a programmable data mulitiplexer 70 that provides rail-to-rail and LVDS operational modes in response to a command signal 31. The first and second rail-to-rail drivers 65 and 66 are respectively coupled to the first and second data output ports 21 and 22 and the LVDS driver 68 is differentially coupled to the first and second data output ports 21 and 22.

[0024] In the driver system 60, the rail-to-rail drivers 65 and 66 are realized as CMOS inverters in which a p-channel MOS transistor 71 and an n-channel MOS transistor 72 are serially-coupled drain-to-drain between bias voltages V_DDand ground. The gates of the transistors 71 and 72 are preferably respectively buffered by inverters 73 and 74. In successive response to V_DDand ground signals from the multiplexer 70, successive output signals at output ports 21 and 22 will have substantially V_DDand ground signal levels (into high-impedance loads). Thus, the CMOS inverters 65 and 66 provide rail-to-rail output signals at the output ports 21 and 22.

[0025] The LVDS driver 68 includes first, second, third and fourth current-steering switches in the form of n-channel MOS transistors 76, 77, 78 and 79 that are serially-connected to form a circuit bridge 80 with the first and third transistors 76 and 78 respectively coupled to the top and bottom of a first side of the circuit bridge and the second and fourth transistors 77 and 79 respectively coupled to the top and bottom of a second side of the circuit bridge.

[0026] Gates of the first and fourth transistors 76 and 79 are coupled together and gates of the second and third transistors 77 and 78 are coupled together to thereby form a differential input port 81. Current sources 83 and 84 couple (through switches 85 and 86) the top and bottom of the circuit bridge 80 between V_DDand ground. The resistive load 35 is coupled (via output ports 21 and 22) across the circuit bridge 80.

[0027] In response to one state of differential signals at its input, transistors 76 and 79 will direct current 41 through the load 35 and, in response to a different state of the differential signals, transistors 77 and 78 will direct current 42 through the load. When the circuit bridge 80 is not in operation, current drains can be reduced by opening the switches 85 and 86 (or by applying low signals to the gates of transistors 71 and 72 with the multiplexer 70).

[0028] The multiplexer 70 includes a first switch network 91, a second switch network 92 and a third or LVDS switch network 93. As indicated by the realization arrow 94, the first switch network 91 can be realized with switches 95 and 96 that, when closed, respectively couple inverters 73 and 74 to ground and V_DD. The first switch network 91 also includes switches 97 and 98 that, when closed, couple the data input signal D₁to the inverters 73 and 74.

[0029] The first switch network is shown in the rail-to-rail mode wherein the data input signal D₁is processed through the first rail-to-rail driver 65. When programmed to the LVDS mode, the switches 95, 96, 97 and 98 all change state so that the inverter transistors 71 and 72 are biased to high output-impedance states. Although not shown, the second switch network 92 is similarly realized and performs similar operational modes with respect to the output port 22.

[0030] As indicated by the realization arrow 104, the third or LVDS switch network 93 can be realized with switches 105 and 106 that, when closed, couple the input port 81 to ground and thereby bias the transistors of the circuit bridge 80 to high output-impedance states. The switch network 93 also includes switches 107 and 108 and an inverter 109 that is in series with the switch 108. When closed, the switches differentially couple the data input signal D₂to the input port 81 of the circuit bridge 80. In a feature of the invention, the inverter 109 converts the single-ended data input signal D₂to a differential input signal. Essentially, the switches 107 and 108 form first and second signal paths and the inverter 109 is positioned in one to these signal paths.

[0031] The third or LVDS switch network is shown in the LVDS mode wherein the data input signal D₂is differentially processed through the LVDS driver 68. When programmed to the rail-to-rail mode, the switches 105, 106, 107 and 108 all change state so that the bridge transistors 76, 77, 78 and 79 are biased to high output-impedance states. Although the third or LVDS switch network 93 could be coupled to a respective third input port to receive data input signals, it is exemplarily coupled in the driver embodiment 60 of FIG. 2 to the first input port 33. That is, the first and third data input ports may be separate input ports or may be a common input port.

[0032] Driver systems of the present invention are especially suited for use in signal conditioning systems such as the ADC system 120 of FIG. 3 in which a driver system 122 supplies the output signals of an ADC 121. In particular, the driver system provides pairs of output pins such as the exemplary pair of pins 21 and 22 (also shown in FIGS. 1 and 2) and couples a driver system 123 to the pair wherein the driver system 123 is exemplarized by the systems 20 and 60 of FIGS. 1 and 2.

[0033] The ADC 121 can be realized as various specific ADC structures which include a successive approximation ADC 124, a flash ADC 126 and a pipelined ADC 128 as respectively indicated in FIG. 3 by realization arrows 125, 127 and 129.

[0034] In the successive approximation ADC 124, an analog input signal is sampled with a sample-and-hold amplifier (SHA) 130 and the resulting samples passed to one input of a comparator 132. In response to digital command signals 133 from a successive-approximation register (SAR) 134, a DAC 136 provides analog approximation signals to another input of the comparator 132.

[0035] With feedback decisions from the comparator 132, the SAR typically determines the most significant bit (MSB) initially and then proceeds to vary its digital command signals 133 to successively determine the remaining bits. At the end of this successive process, the final digital command signals 133 are processed to the converted digital signal by an encoder 138 and passed to the programmable drivers 122.

[0036] In the flash ADC 126, an analog input signal is applied to a string of comparators 140 which each receive a respective reference signal from a ladder of serially-connected resistors 142. The decisions of the comparators 140 is coupled to an encoder 144 that provides the converted digital signal to the programmable drivers 122.

[0037] The pipelined ADC 128 samples analog signals with a SHA 150 and converts the sampled signals to MSBs with an ADC 152 (e.g., a flash ADC). These MSBs are coupled to a correction logic 154 and are also converted to a corresponding analog signal in a DAC 156. The corresponding analog signal is subtracted from the sampled signals to produce an analog residue signal that is gained up in an amplifier 158 and passed to a succeeding conversion stage 160. At the end of this process, a final ADC 162 generates the least significant bits and passes them to the correction logic 154 which time aligns the converted signals of all stages to thereby generate the output digital signal. The residue signals between conversion stages are typically sampled before processing by succeeding stages. The outputs of the correction logic 154 are coupled to the programmable drivers 122.

[0038] The switches of the switch networks 91, 92 and 93 of FIG. 2 can be realized in various conventional forms, e.g., as a parallel an n-channel MOS transistor and a n-channel MOS transistor coupled in parallel with their common sources and their common drains forming opposite switch terminals that are coupled and decoupled in response to command signals applied to their gates.

[0039] Although data bits in signal conditioning systems are typically designated as D₀, D₁- - - D_n, exemplary data bits at first and second input ports have been shown as D₁and D₂in FIGS. 1 and 2 to enhance clarity and simplicity of the corresponding description.

[0040] The embodiments of the invention described herein are exemplary and numerous modifications, variations and rearrangements can be readily envisioned to achieve substantially equivalent results, all of which are intended to be embraced within the spirit and scope of the invention as defined in the appended claims.