CA1139447A - Monolithic analog-to-digital converter - Google Patents
Monolithic analog-to-digital converterInfo
- Publication number
- CA1139447A CA1139447A CA000347177A CA347177A CA1139447A CA 1139447 A CA1139447 A CA 1139447A CA 000347177 A CA000347177 A CA 000347177A CA 347177 A CA347177 A CA 347177A CA 1139447 A CA1139447 A CA 1139447A
- Authority
- CA
- Canada
- Prior art keywords
- analog
- digital converter
- comparators
- output
- converter
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired
Links
Classifications
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
- H01L21/00—Processes or apparatus adapted for the manufacture or treatment of semiconductor or solid state devices or of parts thereof
- H01L21/70—Manufacture or treatment of devices consisting of a plurality of solid state components formed in or on a common substrate or of parts thereof; Manufacture of integrated circuit devices or of parts thereof
- H01L21/77—Manufacture or treatment of devices consisting of a plurality of solid state components or integrated circuits formed in, or on, a common substrate
- H01L21/78—Manufacture or treatment of devices consisting of a plurality of solid state components or integrated circuits formed in, or on, a common substrate with subsequent division of the substrate into plural individual devices
- H01L21/82—Manufacture or treatment of devices consisting of a plurality of solid state components or integrated circuits formed in, or on, a common substrate with subsequent division of the substrate into plural individual devices to produce devices, e.g. integrated circuits, each consisting of a plurality of components
- H01L21/822—Manufacture or treatment of devices consisting of a plurality of solid state components or integrated circuits formed in, or on, a common substrate with subsequent division of the substrate into plural individual devices to produce devices, e.g. integrated circuits, each consisting of a plurality of components the substrate being a semiconductor, using silicon technology
- H01L21/8222—Bipolar technology
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03M—CODING; DECODING; CODE CONVERSION IN GENERAL
- H03M1/00—Analogue/digital conversion; Digital/analogue conversion
- H03M1/12—Analogue/digital converters
- H03M1/34—Analogue value compared with reference values
- H03M1/36—Analogue value compared with reference values simultaneously only, i.e. parallel type
- H03M1/361—Analogue value compared with reference values simultaneously only, i.e. parallel type having a separate comparator and reference value for each quantisation level, i.e. full flash converter type
Abstract
ABSTRACT OF THE DISCLOSURE
A parallel analog-to-digital converter having high speed and high resolution, fabricated on a single integrated-circuit chip in such a manner as to avoid problems typically associated with high speed parallel converters. The converter disclosed by way of example has an eight-bit output, 256 matched comparators for quantizing an analog input signal, and encoding and latching logic for deriving digital outputs from the comparators. Problems related to comparator mismatch-ing, high comparator input capacitance and high com-parator input bias current, are effectively minimized by the use of a triple diffusion fabrication process, which substantially reduces the number of defects in the circuit and provides a relatively high component packing density.
A parallel analog-to-digital converter having high speed and high resolution, fabricated on a single integrated-circuit chip in such a manner as to avoid problems typically associated with high speed parallel converters. The converter disclosed by way of example has an eight-bit output, 256 matched comparators for quantizing an analog input signal, and encoding and latching logic for deriving digital outputs from the comparators. Problems related to comparator mismatch-ing, high comparator input capacitance and high com-parator input bias current, are effectively minimized by the use of a triple diffusion fabrication process, which substantially reduces the number of defects in the circuit and provides a relatively high component packing density.
Description
f t ! 1~3~44~ !
:
BACKGROUND OF THE INVENTION
.. . _ The present invention relates generally to analog-to-digital converters, and, more particularly, to analog-to-diqital converters capable of operating at very high speeds and suitable for processing signals having high frequency components, such as video signals.
In general, there are a variety of techniques available for conversion of an analog signal to an equivalent digital signal. An analog signal is one that can vary continuously over a range of values and can assume any value within the range, whereas a digital signal takes the form of a numerical quantity represent-ing the value of one of a number of disceete steps ln ; the signal range. The technique best suited for high speed analog-to-digital conversion is known as parallel conversion. In a parallel converter, an input analog signal is simultaneously compared with a number of different reference signals, each corresponding to a discrete step in the signal range. ~he simultane-ous comparison operations are performed in a set of ; matched comparators. Por a zero input signal, for example, all the comparators provide outputs of the same state, and as the input signal increases in value, an increasing number of the comparators provide outputs of the opposite state. In this manner, the analog lnputsignal is quantized, i.e., is recognized as falling ~ .
:~13~47 . ' . -~ '~ " -) '
:
BACKGROUND OF THE INVENTION
.. . _ The present invention relates generally to analog-to-digital converters, and, more particularly, to analog-to-diqital converters capable of operating at very high speeds and suitable for processing signals having high frequency components, such as video signals.
In general, there are a variety of techniques available for conversion of an analog signal to an equivalent digital signal. An analog signal is one that can vary continuously over a range of values and can assume any value within the range, whereas a digital signal takes the form of a numerical quantity represent-ing the value of one of a number of disceete steps ln ; the signal range. The technique best suited for high speed analog-to-digital conversion is known as parallel conversion. In a parallel converter, an input analog signal is simultaneously compared with a number of different reference signals, each corresponding to a discrete step in the signal range. ~he simultane-ous comparison operations are performed in a set of ; matched comparators. Por a zero input signal, for example, all the comparators provide outputs of the same state, and as the input signal increases in value, an increasing number of the comparators provide outputs of the opposite state. In this manner, the analog lnputsignal is quantized, i.e., is recognized as falling ~ .
:~13~47 . ' . -~ '~ " -) '
2 ~ r~' . ~i .
between two adjacent discrete levels in the signal range. If an n-bit binary output is required, the input signal is quantized into one of 2n different quantization levels, utilizing 2n _ 1 comparators.
S Typically, the outputs of the comparators are then processed to provide the desired n-bit binary code equivalent to the instantaneous value of the analog input signal.
- For many high speed analog-to-digital conver-sion applications, the resolution required is equiva-lent to eight bits or more, and the analog signal must be quantized into 28 or 256 discrete levels to attain the desired resolution. This would require 255 matched comparators and approximately 20,000 individual compo-nents. Such a device has been manufactured using discrete components, but was found to suffer from a number of significant drawbacks, not the least of which was extremely high cost. The principal disadvantage of such a converter was that the 255 discrete compara-tor circuits could not always be perectly matched intheir performance characteristics. Furthermore, the input capacitance of the comparators used was rela-tively high, and a costly buffer amplifier had to be used in many instances. ~o add further to the cost of such a device using discrete components, differential delays in the converter necessitated use of a sample-and-hold circult.
.
Although it might appear that the use of field effect transistors (FET~s) in a ~onolithic circuit would av3id these dlfficulties, this is not the case. If field-effect tr~nsistors are used, an . . ~ . .... .. ... . .. .. .. . . . .. ... .... . . .. . . . ... . . . .. .... . .
. , ` ' , ' ) ~
~ 3 ., .- '.
inherently higher mismatch in threshold leveli, usually referred to as ~Vt, can lead to ambiguity in the states of the comparators, and resultant unreliability of the device.
An alternative approach for obtaining high resolution in parallel analog-to-digital converters is to use cascaded stages of smaller parallel converters.
For example, a first four-bit converter could be used to quantize the analog input signal into one of sixteen levels, and a second four-bit converter used to provide a further four bits of resolution. Although this approach reduces the complexity of the overall conver-sion system, the two separate conversions result in a greater time delay, and necessitate the use of addi-tional components, including a sample-and-hold circuit and a digital-to-analog conversion circuit.
It is an object of the present invention to provide a parallel analog-to-digital converter with a resolution of at least five bits, which avoids all of the aforementioned disadvantages and thereby provides a practical, high resolutlon, high speed converter.
SUMMARY OF THE INVENTION
The present invention resides in a monoli-thic, parallel, high speed, anslog-to-digital converter circuit having a resolutlon of at least five bits.
Clearly, lt would be highly advantageous, both from a user standpoint and from a manufacturlng ~tandpoint, to produce an analog-to-digital converter of this type on a single lntegrated-circuit chip. However, ~ecause a `` 113~47 :
.
parallel analog-to-digital converter has such a large number of circuit elements, and therefore requires a relatively large chip area, its manufacture in inte-grated-circuit form using conventional fab~ication techniques results in a relatively large number of circuit defects in each chip. This, in turn, leads to mismatching of the comparators, and resultant inopera-tiveness or, at best, inaccuracy of the device. The circuit of the present invention is a high speed, parallel, analog-to-digital converter manufactured utilizing a triple diffusion process to avoid the problems of a more conventional epitaxial layer fabri-cation process, and to provide an acceptably high production yield.
., .
Althougb a monolithic analog-to-digital converter with a resolution as high as eight bits might ~ be manufactured using a conventional epitaxial layer ; fabrication technique, the defect density of such a circuit would be so high that the manufacturing yield, i.e., the percentage of acceptable products obtained from the process, would be very low, probably less than 1%, and the manufacturing costs would therefore be unacceptably high. Even the acceptable circuits that might be produced in this manner would have possible comparator mismatching problems. It is well known that, in the epitaxial fabrication process, there are significant defects in the form of spikes on the epitaxial layer. These spikes can tear a photoresist layer usually applied over the epitaxial layer, and can therefore produce significant circuit defects which, in turn, affect the unlformity of tbe comparator charac-teristics, as well as the production yield. If the ;epitaxial process is used, comparator ~ismatching can also arise from crystalline defects in the epitaxial .
.. , _., ... . __, ,_ __. _ _ ,, .. ; , . . .. .. .
layer. In the triple diffusion process, bipolar transistor elements are diffused into a substrate material, and no epitaxial layer is involved. Accord-ingly, there are no defe~ts due to epitaxial layer spikes, and substantially fewer defects of any kind.
Moreover, without the use of the epitaxial layer a flatter surface is obtained, and a circuit pattern of higher resolution may be used, resulting in high density packing of the components, as well as smaller transistors having lower power consumption and lower input capacitances, and having the low input leakage currents required for satisfactory comparator matching.
The manufacturing yield for an eight-bit analog-to-digital converter using the triple diffusion process is in excess of 25%, an acceptably high figure for complex circuits.
Basically, and in general terms, the circuit of the invention is a monolithic, parallel, analog-to-digital converter having an n-bit resolution, compris-j 20 ing 2n _ 1 comparators, voltage divider means to provide 2n _ 1 reference voltage levels or applica-tion to the comparators, means for coupling an analog input voltage to the comparators, and logic means for converting output signals feom the comparators to an equivalent n-bit digit signal. The converter of the invention is fabricated using a triple diffusion technique that provides a feature geometry down to approximately the two-micron level. The triple diffu-sion process results in the use of relatively low currents, so that the input bias currents to the comparators are kept as low as one microampere.
Moreover, by avoiding the epitaxial layer process, tbe high resolution patterns that can be employed result in .
'` ` li3!i~4~ ) transistors that are very ~uch smaller than in the epitaxial device, and that have much lower input capacitance characteristics.
It will be apparent from the foeegoing that S the present invention represents a significant advance ln the field of analog-to-digital converters, in that it provides a very high speed converter on a single integrated-circuit chip, which can be produced at relatively high production yields without any of the inherent disadvantages of di~crete-component circuits designed for the same purpose. Other aspects and advantages of the present invention will become apparent from the following more detailed desc~iption, taken in conjunction with the accompanying drawings.
BRIEF DESCRIPTION OF THE DRAWINGS
; .. - -- FIGURE 1 is a simplified schematic and block diagram of the analog-to-digital converter of the present invention;
;
FIGS. 2a-2c together constitute a timing diagram showing the relationship between a clock signal, an analog input signal, and corresponding digital output signals;
FIG. 3 is a detailed schematic diagram o a comparator circuit utilized in the converter of the invention;
FIG. 4 is a simplified block diagram of the eight-bit encoding circuitry used in the converter of the lnvention;
'` 113~447 FIG. 5 is a fragmentary schematic diagram of a portion of a set of 32 column OR gates shown in FIG.
4;
FIG. 6 is a diagrammatic view showing how all of the set of column OR gates of FIG. 5 are connected;
FIG. 7 is a schematic diagram showing a ~- sensor circuit for use with the column OR gates shown in FIGS. 5 and 6;
FIG. 8 is a schematic diagram showing how reference voltages are generated for buffer circuits used in conjunction with the column OR gates of FIGS. 5 and 6;
FIG. 9 is a schematic diagram showing a ' clocked buffer used to store outputs from the column OR
gates of FIGS. 5 and 6;
FIG. 10 is a fragmentary schematic diagram . showing portions of a set of output OR gates used to combine the outputs from the column OR gates of FIGS. 5 and 6;
20FIG. 11 is a diagrammatic view showing how all of the .output OR gates are connected to provide .digital outputs;
FIG. 12 is a schematic diagram showing a . sensing circuit for the output signals from the output - 25 OR gates of FIGS. 10 and 11;
FIG. 13 is a schematic diagram showing ln detail the latch circuit used to ~tore the outputs from - the output OR gates of FIGS. 10 and 11;
.. ., ,. . . , .. .. . . . _ . _ .
~`` ~` `) 113~4~7 FIG. 13a is a schematic diagram of an input buffer used to process an inversion control signal applied to the latch circuit of FIG. 13;
- .
FIG. 14 is a schematic diagram showing a clock buffer circuit for generating clock signals for use with the output buffer circuit shown in FIG.
... .
FIG. 14a is a schematic diagram of a clock buffer circuit used to generate internal clock signals;
FIG. 14b is a schematic diagram of a clock signal translator circuit for generating an internal clock signal from an externally supplied clock signal;
FIG. 15 is a simplified diagrammatic plan view, not drawn to scale, showing a transistor manufac-tured in accordance with the advanced triple diffusionprocess, which forms an essential part of the present invention; and FIGS. 16 - 25 are sectional views illustrat-ing the sequence of process steps performed in the advanced triple diffusion process.
DESCRIPTION OF THE PREFERRED EMBODIMENT
. .
As shown in the drawings for purposes of illustration, the present invention is concerned with a monolithic parallel, analog-to-digital converter capable of operation at very high speed. As mentioned earlier ln this specification, the basic circuit elements needed to construct a high resolution converter of this type are well known, at least conceptually.
.. ____ ... . .. , __ , . _ .. . _ . . , ., _. __ , . . . . . .. _ ., . , ~
~-"
447 ?
.
g Basically, and as shown in FIG. 1, a plurality of comparators, indicated by reference numeral 10, are employed to compare the instantaneous value of an input analog signal, on line 12, with a graduated plurality .. 5 of reference signals obtained from a precision resist-; ance voltage divider 14.
.
Each of the comparators 10 provides an output signal of one particular state when the analog input signal exceeds the reference signal level applied to 10 the particular comparator, and provides an output of - the opposite state when the analog input signal does ; not exceed the reference signal. Thus, when the analog input signal is at or near zero, all of the comparator outputs have the same state, but as the analog input signal increases in value, more and more of the co~pa-rators change the state of their output signals. In this manner, the arrangement of comparators quan-tizes the analog signal into one of a plurality of discrete levels, and this level can be further encoded 20 in the form of a binary output signal. Unfortunately, however, for converters having binary outputs of more than four bits, i.e. for resolutions better than 1 part in 16, there are practical difficulties of circuit design that render such a converter virtually impos-sible to produce using conventional integrated-circuit techniques. Moreover, equivalent circuits using discrete components have thus far proved to be very costly, and not always reliable.
In accordance with the present invention, a 30 parallel, high speed, analog-to-digital converter is provided, utilizing a triple diffusion fabrication process that avoids the inherent disadvantages of the prior art, and thereby provides a reliable and rela-f` ~ J4~7 tively low-cost converter on a single integrated-cir-cuit chip. Many of the circuit design details of the analog-to-digital converter disclosed herein are not critical to the invention. Rather, it is believed that the invention in its broadest sense encompasses the use of an advanced triple diffusion process, which will be discussed in more detail, to fabricate a parallel analog-to-digital converter having a resolution of five or more bits, such resolution being previously unobtain-;- 10 able in a low-cost, sinqle-stage converter. A detailed circuit design for an eight-bit converter is presented herein by way of example, but it will be understood that the invention is not limited to an eight-bit converter, or to any other design details of the disclosed circuit design.
~ .
As shown in FIG. 1, a reference voltage indicated as VREF, is applied across the precision voltage divider 14 to provide a plurality of reference voltage levels, which are applied to the non-inverting terminals of the respective comparators 10. The analog input voltage on the line 12, is applied to the inverting terminals of the comparators 10, and a clock signal is applied to enable the comparators, as shown at 16. The clock signal is depicted as being supplied from an external source over line 18. Internal clock signals are shown as being generated by an amplifier 20, which has line 18 as its input, and has an inverted output on line 22 and a non-inverted output on line 24.
As later described, the clock signals on these lines are employed to control gating logic in various por-tions of the converter circuitry.
Each of the comparators 10 produces an output signal of one state, ~ logical zero, when the analog input voltage exceeds the reference voltage applied to .
) il~'3447 11 , the comparator, and produces an output of the opposite state, a logical one, when the analog input voltage is less than the reference voltage. Each of the con~para-tor outputs is applied as one inputr over line 25, to a corresponding ANI) gate, indicated at 26, and each AND
gate has two other inputs. For each AND gate 26 except the uppermost one, i.e. the one associated with the highest reference voltage, a second input is supplied by a line 28 from the next highest comparator output.
Similarly, for each AND gate except the lowermost one, a third input is supplied by a line 30 from the next lowest comparator output. This third input is inverted. For the uppermost AND gate, the second input signal is a permanently supplied logical one, as shown at 32, and for the lowermost AND gate, the third, inverted input is a permanently supplied logical zero, as shown at 34. The AND gates 26 are all enabled by a clock signal on line 36, obtained from line 22.
In the embodiment illustrated in detail in the drawings, there are 256 comparators 10, although it will be appreciated that only 255 comparators are needed to quantize an analog signal into one of 256 discrete levels. Using 256 comparators and related components preserves the symmetry of the circui'cry, and helps achieve the desired objective of matching the electrical characteristics of the comparators.
The purpose of the AND gates 26 is to provide a signal on only one of 256 lines, the output lines 38 from the AND gates. It will be apparent from inspec-tion of the AND gate logic that, when all the compara-tor outputs are logical ones, the lowermost AND gate will have a logical one output, but all higher AND
gates will have zero outputs. If the lowermost compa-( ( :113~4~7 :. ` ' '`
.
. . .
rator output is a zero, and all the other comparator outputs are ones, the lowermost AND g~te will provide a zero output, the second AND gate will provide a one output, and all of the higher AND gates will provide zero outputs. Thus, it will be seen that, in general, the output of the AND gate at the same level as the lowest comparator having a one output will also be one, while all other AND gate outputs will be zero. When all the comparator outputs are zero, all of the AND
gates will have zero outputs.
:, .
The outputs of the AND gates 26 are next gated over the output lines 38 into an encoder 40, which is operative to convert the l-of-256 input to an eight-bit output code on lines 42. The output code is then gated into an eight-bit latch 44, from which the desired output signal is derived on lines 46. Opera-tion of the encoder 40 is controlled by clock signals on llne 24, and operation of the latch 44 by clock signals on line 22.
FIGS. 2a, 2b and 2c illustrate the basic timing relationships of the converter. As shown in FIG. 2a, a periodic clock signal is employed to time sampling of the analog input signal shown by way of example in FIG. 2b. FIG. 2c shows the timing of poss1ble single-bit digital output changes from zero to one or one to zero. It will be appreciated from FIGS.
2a-2c that there is a time delay of somewhat more than one clock period between the sampling of analog data and the generation of corresponding digital data, the delay corresponding to the propagation time through the AND gates 26 and encoder logic 40.
.... _ .. .. . .. . ... _ . . . _ .
(` i 11394~ ) -., .
Any of a number of different comparator designs could be used in practicing the present inven-` tion. However, by way of illustration, the specific comparator circuit module design presently preferred is shown in detail in FIG. 3. This circuit moduleincludes both a comparator 10 and an AND gate 26. It will be noted that both the comparator and AND gate utilize, for the most part, differential circuitry and cuerent mode logic. Although this arrangement requires a large number of component parts, it minimizes the necessity for a very accurate reference voltage and for close tracking of comparator output levels, both of which would be essential if a single-ended circuit scheme were used. The comparator and AND gate circuit module shown in FIG. 3 has a VRE~ input line 50, an input line 12 for the analog input voltage VIN, and an AND gate output line 38, the output signal being indicated by the signal name COMPOUT.
The comparator circuit 10 provides two pairs of output signals, designated AOUT, AOUT' and BOUT, B~UT', respectively. These supply the output slgnals for cross-connection to adjacent ones of the AND gates 26, as indicated by lines 28 and 30 in FIG. 1. Simi-larly, the AND gate circuit 26 has two pairs of input 2S signals, designated AIN, AIN', and BIN, BIN' for connection of the AND gate inputs. It will be appre-~ciated that different AND gate circuitry ~ust beemployed for processinq outputs from the first and last comparators in the system, since these require a logical one and a logical zero to be permanently - supplied to the respective AND gates.
The comparator circuit module of FIG. 3 also includes inverted and non-inverted clock ter~inals, indicated at 52 and 52', to which signals CLK and CLK' ._ .... . _. _. .__. .. ._ .. . . . .......... .. .. ... , . _ _ __ . _ . ..
.
f. -(` il~i447 - . .
are applied. These internal clock signals are gen-erated from the externally applied clock signal, by the circuits illustrated in FIGS. 14a and 14b. The circuit module of FIG. 3 also has a power supply terminal , 5 54, and a ground terminal 56. The ohmic values of the resistance elements in the circuit are qiven directly on the circuit diagram, and the NPN transistor elements are fabricated as will now be described in more detail.
.
The triple diffusion process by which the analog-to-digital converter of the present invention is fabricated is illustrated in FIGS. 15-25. As shown in FIG. 16, a silicon substrate 60 is coated with a first oxide layer 62. Then, as shown in FIG. 17, a rectangu-lar area or window, indicated by reference numeral 64, of the first oxide layer 62 is re~oved, and an N type material, in this case phosphorous, is implanted in what is to become the collector region of the transis-tor, indicated at 66 in FIGS. 19 and 20.
The area of the collector region 66, and the areas of the other diffusion regions of the transistor, are all defined geometrically by means of a conven-tional photoresist process, referred to as PR in the legendæ accompanying the drawings, but not illustrated in detail. Basically, in such a process a photosensi-tive coating known as photoresist is deposited on theoxide, and is then selectively exposed to ultraviolet light through a photolith mask (not shown). In the particular case of the collector region 66, the mask defines the rectangular window 64. The exposed window area of the photoresist is ~ubseguently washed away, and then the oxide in the window area is etched with acid to expose the underlying silicon substrate, leaving the window 64, as ~hown in FIG. 17. The N type , ' .
~` ~
~:13~4'7 , collector material, indicated at 68, can then be implanted in the collector window region, and the remaining oxide layer 62 stripped off, to leave the structure as indicated in-FIG. 18. In a subsequent diffusion operation, a second oxide layer 70 is for~ed as indicated in FIG. 19, and the N type collector material 68 is diffused into the substrate 60 to form the collector region 66.
.
The collector diffusion region 66 is approxi-mately 3.5 microns deep, has a surface concentration of 2 x 1017 cm 3 and a sheet resistance of approxi-mately 500 ohms per square. Sheet resistance, as is well known, is a ter~ usually applied to the electrical resistance of a conductive or semiconductive layer.
The resistance to current flow from one edge of a square sheet of such a layer, to the opposite edge, is independent of its area.
.~. .
Next, as shown in FIG. 20, the second oxide layer 70 is selectively removed to form a base-region window, indicated at 72. Portions of the oxide layer 70 situated outside of the collector region 66 are also removed. Then, a P type material, in this case boron, is implanted in the base-region window 72, as indicated at 74, and is also implanted in the exposed regions of the substrate outside of the collector region 66.
Then, as shown in FIG. 21, a third oxide layer 76 is added, and the boron is diffused tc a depth of approxi-mately 1.2 microns, to form the base region 78. The base region has a depth of approximately 1.2 microns, a surface concentration of 1 x 1019 cm 3 and a sheet resistance of 150-200 ohms per square. The P type boron material is also diffused into the surround-ing field region outside of the collector region 66.
, . . .. ~ . .. _ _ . _ .. , _ .. _ , . ,, . . .. . _ . _ . .. . . . . . ~ _ . .. _ _ _ _ ~
S 44'7 .. ..
This field diffusion region, indicated at 80, increases the surface concentration of the substrate 60, and effectively isolates adjacent devices from each other.
Next, as shown in FIG. 22, the third oxide layer 76 is selectively removed, again using the photoresist process, to expose an emitter-region window 82 over the base region 78, and also to expose a continuous rectangular strip 84 (FIG. 15) over the collector region 66. An N+ type phosphorous emitter material is deposited in the exposed areas, and is diffused to a depth of approximately 0.9 micron, to form the emitter region 86, and also to form a contin-uous N+ type region beneath the rectangular strip 84 around the collector region, for purposes of making lS contact with the collector. The surface concentration of the emitter region 86 is approximately 1 x 1021 cm 3 and the sheet res~stance is approximately 18-25 ohms per square.
As shown in FIG. 23, a fourth oxide layer 90 is applied over the already formed layers, and is selectively removed to expose contact holes to be used for making electrical contact with the various semicon-ductor regions. Then mètal contact strips are formed, again using a conventional photoresist process that is not shown ln detail. First, a metalized layer is ; applied over the entire structure, then a photoresist layer is applied over the metal, selectively exposed through a photolith mask, and washed away to leave photoresist material over those areas of metal to be retained. Then the remaining, unmasked areas of metal are etched away to leave the interconnecting metal strips shown in the drawings. In particular, as shown in FIG. 24, there ls a collector contact strip 92, i:~39447 l7 which makes contact with the N+ field diffusion layer - 84 around the outer periphery of the collector area 66, a base contact strip 94, which makes direct contact with the base region 78, and, finally, an emitter contact strip 96, which makes direct contact with the center of the emitter region 86. The final step in the fabrication process is the application of a passi-vating oxide layer 98 over the entire structure.
- The metal contact strips 92, 94 and 96 are single layers approximately 8,000-12,000 angstroms thick (0.8-1.2 microns). Each metalized layer com-prises a thin first coating of titanium, approximately 50 angstroms thick, with the remainder of the layer comprising copper and aluminum. The metal strips are spaced by a minimum of 2 microns from each other, and are a minimum of 5 microns wide, after shrinkage during the fabrication process. All other features o the device have a minimum geometry of 2 microns, which applies to both width and spacing. Thus, the circuit is said to have a atwo-micron feature geome-try~, meaning that, in plan view, all semiconductor regions have a minimum dimension of two microns, measured in the plane of the device, and are spaced rom each other by a minimum dimension o two microns, measured in the same plane. The areas of the semicon-- ductor regions can be dimensioned as desired, within this geometric limitation. For example, the emitter region 86 in one presently preerred embodiment is approximately 5-6 microns wide, and the contact holes are 3 microns wide.
The encoder 40 and latch 44 shown in FIG. ~
can follow any appropriate circuit design. However, an exemplary design is illustrated in FIGS. 4-14b, for an ... .. _ _.. .. . .. _ _ . . . .. _ .. __ _ .. . .. . .. .. . ... _.. . .. _ _ .. _ _ ._ _ _ .
eight-bit encoder. As shown in FIG. ~, the encoder logic comprises four OR gate circuit modules 100, referred to as colu~n O~ gates, and a set of output OR
gates 102, which functions to combine the outputs from the column OR gates. Each of the column OR gate modules 100 has 64 binary inputs and generates a six-bit coded output on six data lines, together with a zero-bit output signal~ which is employed to indicate when the six data bits are all zero. The four 1~ sets of seven-bit outputs are fed over lines 104 to the output OR gates 102, where they are logically combined to produce an eight-bit digital output for transmission to the latch 44.
As shown in FIG. 5, each colu~n-OR gate circuit module 100 comprises a plurality of NPN tran-sistors. More specifically, there are 64 rows, each containing eight such transistors. The collector terminals of all the transistors in the module are commonly connected to ground, as indicated at 106, and each of the 64 input lines is connected to the base terminals of one row of transistors. Four such input lines are shown by way of example in FIG. 5, and these are designated by the signal names COMPOUT/NO. 0, 31, 32 and 63. Also included in a column OR gate are seven output lines, comprising the six data lines 108 and the zero line 110, to which the emitter terminals of the transistors are selectively connected. An eighth parallel line 112, and a corresponding ~column~ of transistors shown in FIG. 5 are not used in the illu-strative eight-bit configuration~
Selective connection of the output lines 108 ~nd 110 to the emitter terminals of the transistors provides for the conversion of a d iscrete voltage 11399~
19 - .
level, indicated by a signal on one of the 64 input lines, to a corresponding binary code. For example, row ~0 of the transistors has the "zero" column tran-sistor emitter ter~inal connected to the "zero~ output line ll0, but none of the other transistors in row t0 has its emitter terminal connected to an output line.
Thus, when input line t0 to the column OR gate module supplies an input signal to the module, only the ~zero~
output line 110 is affected. By way of further example, row ~31 of transistors has five transistors connected to the output data lines 108, to impress on the data lines the pattern 011111, which will be recognized as the binary code for the decimal number 31.
.
15~he complete areangement of transistor connections for each of the column OR gate modules 100 - is shown diagrammatically in ~IG. 6. The vertical lines in FIG. 6 are the output lines 108 and zero line 110, and the horizontal lines numbered 0-63 represent the input lines. The dots represent connections between emitter terminals and output lines. It will be appreciated that not all of the transistors in the column OR gate modules are utilized, and that the unnecessary transistors need not be present at all.
However, for purposes of manufacturing efficiency, it is convenient to fabricate the column OR gate modules all in identical form, each with a full complement of transistors, and then to ~program~ each of the modules, in a metalization step, by appropriately connecting the emitter terminals to the output data lines.
Each of the output lines 108 and 110 from the column OR gates l00 is connected to a sensing circuit of the type shown by way of example in FIG. 7. Thus, .
... .
~, ' .
when a transistor in one of the column OR gates 100 is rendered conductive by a signal impressed at its base terminal, its emitter terminal is connected to one of the outpu~ lines 108, and the resul~ant signal current will be reflected in a change in state in the sensed output signal on line 120 in FIG. 7. The sensed outpùt signals representing all of the column OR gate outputs are then passed to clocked buffers, of which one is shown in FIG. 8, and fro~ which output signals are generated on lines 122, under control of internal clock signals on iines 52 and 52'. Reference voltages on lines 124 and 126, utilized by the clocked buffer circuit of FIG. 8, are generated from a power supply voltage by the circuit shown in FIG. 9.
The buffered data outputs on lines 122 from the clocked buffer circuits are then supplied to the output OR gates 102, shown in detail in FIGS~ 10 and 11. The output OR gates comprise a plurality of NPN
transistors arranged in eight rows, each of 28 transis-tors. The collector terminals of all of the transis-tors are co~monly connec~ed to ground, and the base terminals of ~each of the 28 columns of transistors are connected, respectively, to the 28 input lines 122 from the buffered outputs of the four column OR gates 100.
Eight binary data output lines 130, which appear as horizontal lines in ~IG~ 10, are selectively connected to the emitter terminals of the transistors to provide an eight-bit digital output. More specifically, and as is apparent from FIG. 11, each set of six data lines,
between two adjacent discrete levels in the signal range. If an n-bit binary output is required, the input signal is quantized into one of 2n different quantization levels, utilizing 2n _ 1 comparators.
S Typically, the outputs of the comparators are then processed to provide the desired n-bit binary code equivalent to the instantaneous value of the analog input signal.
- For many high speed analog-to-digital conver-sion applications, the resolution required is equiva-lent to eight bits or more, and the analog signal must be quantized into 28 or 256 discrete levels to attain the desired resolution. This would require 255 matched comparators and approximately 20,000 individual compo-nents. Such a device has been manufactured using discrete components, but was found to suffer from a number of significant drawbacks, not the least of which was extremely high cost. The principal disadvantage of such a converter was that the 255 discrete compara-tor circuits could not always be perectly matched intheir performance characteristics. Furthermore, the input capacitance of the comparators used was rela-tively high, and a costly buffer amplifier had to be used in many instances. ~o add further to the cost of such a device using discrete components, differential delays in the converter necessitated use of a sample-and-hold circult.
.
Although it might appear that the use of field effect transistors (FET~s) in a ~onolithic circuit would av3id these dlfficulties, this is not the case. If field-effect tr~nsistors are used, an . . ~ . .... .. ... . .. .. .. . . . .. ... .... . . .. . . . ... . . . .. .... . .
. , ` ' , ' ) ~
~ 3 ., .- '.
inherently higher mismatch in threshold leveli, usually referred to as ~Vt, can lead to ambiguity in the states of the comparators, and resultant unreliability of the device.
An alternative approach for obtaining high resolution in parallel analog-to-digital converters is to use cascaded stages of smaller parallel converters.
For example, a first four-bit converter could be used to quantize the analog input signal into one of sixteen levels, and a second four-bit converter used to provide a further four bits of resolution. Although this approach reduces the complexity of the overall conver-sion system, the two separate conversions result in a greater time delay, and necessitate the use of addi-tional components, including a sample-and-hold circuit and a digital-to-analog conversion circuit.
It is an object of the present invention to provide a parallel analog-to-digital converter with a resolution of at least five bits, which avoids all of the aforementioned disadvantages and thereby provides a practical, high resolutlon, high speed converter.
SUMMARY OF THE INVENTION
The present invention resides in a monoli-thic, parallel, high speed, anslog-to-digital converter circuit having a resolutlon of at least five bits.
Clearly, lt would be highly advantageous, both from a user standpoint and from a manufacturlng ~tandpoint, to produce an analog-to-digital converter of this type on a single lntegrated-circuit chip. However, ~ecause a `` 113~47 :
.
parallel analog-to-digital converter has such a large number of circuit elements, and therefore requires a relatively large chip area, its manufacture in inte-grated-circuit form using conventional fab~ication techniques results in a relatively large number of circuit defects in each chip. This, in turn, leads to mismatching of the comparators, and resultant inopera-tiveness or, at best, inaccuracy of the device. The circuit of the present invention is a high speed, parallel, analog-to-digital converter manufactured utilizing a triple diffusion process to avoid the problems of a more conventional epitaxial layer fabri-cation process, and to provide an acceptably high production yield.
., .
Althougb a monolithic analog-to-digital converter with a resolution as high as eight bits might ~ be manufactured using a conventional epitaxial layer ; fabrication technique, the defect density of such a circuit would be so high that the manufacturing yield, i.e., the percentage of acceptable products obtained from the process, would be very low, probably less than 1%, and the manufacturing costs would therefore be unacceptably high. Even the acceptable circuits that might be produced in this manner would have possible comparator mismatching problems. It is well known that, in the epitaxial fabrication process, there are significant defects in the form of spikes on the epitaxial layer. These spikes can tear a photoresist layer usually applied over the epitaxial layer, and can therefore produce significant circuit defects which, in turn, affect the unlformity of tbe comparator charac-teristics, as well as the production yield. If the ;epitaxial process is used, comparator ~ismatching can also arise from crystalline defects in the epitaxial .
.. , _., ... . __, ,_ __. _ _ ,, .. ; , . . .. .. .
layer. In the triple diffusion process, bipolar transistor elements are diffused into a substrate material, and no epitaxial layer is involved. Accord-ingly, there are no defe~ts due to epitaxial layer spikes, and substantially fewer defects of any kind.
Moreover, without the use of the epitaxial layer a flatter surface is obtained, and a circuit pattern of higher resolution may be used, resulting in high density packing of the components, as well as smaller transistors having lower power consumption and lower input capacitances, and having the low input leakage currents required for satisfactory comparator matching.
The manufacturing yield for an eight-bit analog-to-digital converter using the triple diffusion process is in excess of 25%, an acceptably high figure for complex circuits.
Basically, and in general terms, the circuit of the invention is a monolithic, parallel, analog-to-digital converter having an n-bit resolution, compris-j 20 ing 2n _ 1 comparators, voltage divider means to provide 2n _ 1 reference voltage levels or applica-tion to the comparators, means for coupling an analog input voltage to the comparators, and logic means for converting output signals feom the comparators to an equivalent n-bit digit signal. The converter of the invention is fabricated using a triple diffusion technique that provides a feature geometry down to approximately the two-micron level. The triple diffu-sion process results in the use of relatively low currents, so that the input bias currents to the comparators are kept as low as one microampere.
Moreover, by avoiding the epitaxial layer process, tbe high resolution patterns that can be employed result in .
'` ` li3!i~4~ ) transistors that are very ~uch smaller than in the epitaxial device, and that have much lower input capacitance characteristics.
It will be apparent from the foeegoing that S the present invention represents a significant advance ln the field of analog-to-digital converters, in that it provides a very high speed converter on a single integrated-circuit chip, which can be produced at relatively high production yields without any of the inherent disadvantages of di~crete-component circuits designed for the same purpose. Other aspects and advantages of the present invention will become apparent from the following more detailed desc~iption, taken in conjunction with the accompanying drawings.
BRIEF DESCRIPTION OF THE DRAWINGS
; .. - -- FIGURE 1 is a simplified schematic and block diagram of the analog-to-digital converter of the present invention;
;
FIGS. 2a-2c together constitute a timing diagram showing the relationship between a clock signal, an analog input signal, and corresponding digital output signals;
FIG. 3 is a detailed schematic diagram o a comparator circuit utilized in the converter of the invention;
FIG. 4 is a simplified block diagram of the eight-bit encoding circuitry used in the converter of the lnvention;
'` 113~447 FIG. 5 is a fragmentary schematic diagram of a portion of a set of 32 column OR gates shown in FIG.
4;
FIG. 6 is a diagrammatic view showing how all of the set of column OR gates of FIG. 5 are connected;
FIG. 7 is a schematic diagram showing a ~- sensor circuit for use with the column OR gates shown in FIGS. 5 and 6;
FIG. 8 is a schematic diagram showing how reference voltages are generated for buffer circuits used in conjunction with the column OR gates of FIGS. 5 and 6;
FIG. 9 is a schematic diagram showing a ' clocked buffer used to store outputs from the column OR
gates of FIGS. 5 and 6;
FIG. 10 is a fragmentary schematic diagram . showing portions of a set of output OR gates used to combine the outputs from the column OR gates of FIGS. 5 and 6;
20FIG. 11 is a diagrammatic view showing how all of the .output OR gates are connected to provide .digital outputs;
FIG. 12 is a schematic diagram showing a . sensing circuit for the output signals from the output - 25 OR gates of FIGS. 10 and 11;
FIG. 13 is a schematic diagram showing ln detail the latch circuit used to ~tore the outputs from - the output OR gates of FIGS. 10 and 11;
.. ., ,. . . , .. .. . . . _ . _ .
~`` ~` `) 113~4~7 FIG. 13a is a schematic diagram of an input buffer used to process an inversion control signal applied to the latch circuit of FIG. 13;
- .
FIG. 14 is a schematic diagram showing a clock buffer circuit for generating clock signals for use with the output buffer circuit shown in FIG.
... .
FIG. 14a is a schematic diagram of a clock buffer circuit used to generate internal clock signals;
FIG. 14b is a schematic diagram of a clock signal translator circuit for generating an internal clock signal from an externally supplied clock signal;
FIG. 15 is a simplified diagrammatic plan view, not drawn to scale, showing a transistor manufac-tured in accordance with the advanced triple diffusionprocess, which forms an essential part of the present invention; and FIGS. 16 - 25 are sectional views illustrat-ing the sequence of process steps performed in the advanced triple diffusion process.
DESCRIPTION OF THE PREFERRED EMBODIMENT
. .
As shown in the drawings for purposes of illustration, the present invention is concerned with a monolithic parallel, analog-to-digital converter capable of operation at very high speed. As mentioned earlier ln this specification, the basic circuit elements needed to construct a high resolution converter of this type are well known, at least conceptually.
.. ____ ... . .. , __ , . _ .. . _ . . , ., _. __ , . . . . . .. _ ., . , ~
~-"
447 ?
.
g Basically, and as shown in FIG. 1, a plurality of comparators, indicated by reference numeral 10, are employed to compare the instantaneous value of an input analog signal, on line 12, with a graduated plurality .. 5 of reference signals obtained from a precision resist-; ance voltage divider 14.
.
Each of the comparators 10 provides an output signal of one particular state when the analog input signal exceeds the reference signal level applied to 10 the particular comparator, and provides an output of - the opposite state when the analog input signal does ; not exceed the reference signal. Thus, when the analog input signal is at or near zero, all of the comparator outputs have the same state, but as the analog input signal increases in value, more and more of the co~pa-rators change the state of their output signals. In this manner, the arrangement of comparators quan-tizes the analog signal into one of a plurality of discrete levels, and this level can be further encoded 20 in the form of a binary output signal. Unfortunately, however, for converters having binary outputs of more than four bits, i.e. for resolutions better than 1 part in 16, there are practical difficulties of circuit design that render such a converter virtually impos-sible to produce using conventional integrated-circuit techniques. Moreover, equivalent circuits using discrete components have thus far proved to be very costly, and not always reliable.
In accordance with the present invention, a 30 parallel, high speed, analog-to-digital converter is provided, utilizing a triple diffusion fabrication process that avoids the inherent disadvantages of the prior art, and thereby provides a reliable and rela-f` ~ J4~7 tively low-cost converter on a single integrated-cir-cuit chip. Many of the circuit design details of the analog-to-digital converter disclosed herein are not critical to the invention. Rather, it is believed that the invention in its broadest sense encompasses the use of an advanced triple diffusion process, which will be discussed in more detail, to fabricate a parallel analog-to-digital converter having a resolution of five or more bits, such resolution being previously unobtain-;- 10 able in a low-cost, sinqle-stage converter. A detailed circuit design for an eight-bit converter is presented herein by way of example, but it will be understood that the invention is not limited to an eight-bit converter, or to any other design details of the disclosed circuit design.
~ .
As shown in FIG. 1, a reference voltage indicated as VREF, is applied across the precision voltage divider 14 to provide a plurality of reference voltage levels, which are applied to the non-inverting terminals of the respective comparators 10. The analog input voltage on the line 12, is applied to the inverting terminals of the comparators 10, and a clock signal is applied to enable the comparators, as shown at 16. The clock signal is depicted as being supplied from an external source over line 18. Internal clock signals are shown as being generated by an amplifier 20, which has line 18 as its input, and has an inverted output on line 22 and a non-inverted output on line 24.
As later described, the clock signals on these lines are employed to control gating logic in various por-tions of the converter circuitry.
Each of the comparators 10 produces an output signal of one state, ~ logical zero, when the analog input voltage exceeds the reference voltage applied to .
) il~'3447 11 , the comparator, and produces an output of the opposite state, a logical one, when the analog input voltage is less than the reference voltage. Each of the con~para-tor outputs is applied as one inputr over line 25, to a corresponding ANI) gate, indicated at 26, and each AND
gate has two other inputs. For each AND gate 26 except the uppermost one, i.e. the one associated with the highest reference voltage, a second input is supplied by a line 28 from the next highest comparator output.
Similarly, for each AND gate except the lowermost one, a third input is supplied by a line 30 from the next lowest comparator output. This third input is inverted. For the uppermost AND gate, the second input signal is a permanently supplied logical one, as shown at 32, and for the lowermost AND gate, the third, inverted input is a permanently supplied logical zero, as shown at 34. The AND gates 26 are all enabled by a clock signal on line 36, obtained from line 22.
In the embodiment illustrated in detail in the drawings, there are 256 comparators 10, although it will be appreciated that only 255 comparators are needed to quantize an analog signal into one of 256 discrete levels. Using 256 comparators and related components preserves the symmetry of the circui'cry, and helps achieve the desired objective of matching the electrical characteristics of the comparators.
The purpose of the AND gates 26 is to provide a signal on only one of 256 lines, the output lines 38 from the AND gates. It will be apparent from inspec-tion of the AND gate logic that, when all the compara-tor outputs are logical ones, the lowermost AND gate will have a logical one output, but all higher AND
gates will have zero outputs. If the lowermost compa-( ( :113~4~7 :. ` ' '`
.
. . .
rator output is a zero, and all the other comparator outputs are ones, the lowermost AND g~te will provide a zero output, the second AND gate will provide a one output, and all of the higher AND gates will provide zero outputs. Thus, it will be seen that, in general, the output of the AND gate at the same level as the lowest comparator having a one output will also be one, while all other AND gate outputs will be zero. When all the comparator outputs are zero, all of the AND
gates will have zero outputs.
:, .
The outputs of the AND gates 26 are next gated over the output lines 38 into an encoder 40, which is operative to convert the l-of-256 input to an eight-bit output code on lines 42. The output code is then gated into an eight-bit latch 44, from which the desired output signal is derived on lines 46. Opera-tion of the encoder 40 is controlled by clock signals on llne 24, and operation of the latch 44 by clock signals on line 22.
FIGS. 2a, 2b and 2c illustrate the basic timing relationships of the converter. As shown in FIG. 2a, a periodic clock signal is employed to time sampling of the analog input signal shown by way of example in FIG. 2b. FIG. 2c shows the timing of poss1ble single-bit digital output changes from zero to one or one to zero. It will be appreciated from FIGS.
2a-2c that there is a time delay of somewhat more than one clock period between the sampling of analog data and the generation of corresponding digital data, the delay corresponding to the propagation time through the AND gates 26 and encoder logic 40.
.... _ .. .. . .. . ... _ . . . _ .
(` i 11394~ ) -., .
Any of a number of different comparator designs could be used in practicing the present inven-` tion. However, by way of illustration, the specific comparator circuit module design presently preferred is shown in detail in FIG. 3. This circuit moduleincludes both a comparator 10 and an AND gate 26. It will be noted that both the comparator and AND gate utilize, for the most part, differential circuitry and cuerent mode logic. Although this arrangement requires a large number of component parts, it minimizes the necessity for a very accurate reference voltage and for close tracking of comparator output levels, both of which would be essential if a single-ended circuit scheme were used. The comparator and AND gate circuit module shown in FIG. 3 has a VRE~ input line 50, an input line 12 for the analog input voltage VIN, and an AND gate output line 38, the output signal being indicated by the signal name COMPOUT.
The comparator circuit 10 provides two pairs of output signals, designated AOUT, AOUT' and BOUT, B~UT', respectively. These supply the output slgnals for cross-connection to adjacent ones of the AND gates 26, as indicated by lines 28 and 30 in FIG. 1. Simi-larly, the AND gate circuit 26 has two pairs of input 2S signals, designated AIN, AIN', and BIN, BIN' for connection of the AND gate inputs. It will be appre-~ciated that different AND gate circuitry ~ust beemployed for processinq outputs from the first and last comparators in the system, since these require a logical one and a logical zero to be permanently - supplied to the respective AND gates.
The comparator circuit module of FIG. 3 also includes inverted and non-inverted clock ter~inals, indicated at 52 and 52', to which signals CLK and CLK' ._ .... . _. _. .__. .. ._ .. . . . .......... .. .. ... , . _ _ __ . _ . ..
.
f. -(` il~i447 - . .
are applied. These internal clock signals are gen-erated from the externally applied clock signal, by the circuits illustrated in FIGS. 14a and 14b. The circuit module of FIG. 3 also has a power supply terminal , 5 54, and a ground terminal 56. The ohmic values of the resistance elements in the circuit are qiven directly on the circuit diagram, and the NPN transistor elements are fabricated as will now be described in more detail.
.
The triple diffusion process by which the analog-to-digital converter of the present invention is fabricated is illustrated in FIGS. 15-25. As shown in FIG. 16, a silicon substrate 60 is coated with a first oxide layer 62. Then, as shown in FIG. 17, a rectangu-lar area or window, indicated by reference numeral 64, of the first oxide layer 62 is re~oved, and an N type material, in this case phosphorous, is implanted in what is to become the collector region of the transis-tor, indicated at 66 in FIGS. 19 and 20.
The area of the collector region 66, and the areas of the other diffusion regions of the transistor, are all defined geometrically by means of a conven-tional photoresist process, referred to as PR in the legendæ accompanying the drawings, but not illustrated in detail. Basically, in such a process a photosensi-tive coating known as photoresist is deposited on theoxide, and is then selectively exposed to ultraviolet light through a photolith mask (not shown). In the particular case of the collector region 66, the mask defines the rectangular window 64. The exposed window area of the photoresist is ~ubseguently washed away, and then the oxide in the window area is etched with acid to expose the underlying silicon substrate, leaving the window 64, as ~hown in FIG. 17. The N type , ' .
~` ~
~:13~4'7 , collector material, indicated at 68, can then be implanted in the collector window region, and the remaining oxide layer 62 stripped off, to leave the structure as indicated in-FIG. 18. In a subsequent diffusion operation, a second oxide layer 70 is for~ed as indicated in FIG. 19, and the N type collector material 68 is diffused into the substrate 60 to form the collector region 66.
.
The collector diffusion region 66 is approxi-mately 3.5 microns deep, has a surface concentration of 2 x 1017 cm 3 and a sheet resistance of approxi-mately 500 ohms per square. Sheet resistance, as is well known, is a ter~ usually applied to the electrical resistance of a conductive or semiconductive layer.
The resistance to current flow from one edge of a square sheet of such a layer, to the opposite edge, is independent of its area.
.~. .
Next, as shown in FIG. 20, the second oxide layer 70 is selectively removed to form a base-region window, indicated at 72. Portions of the oxide layer 70 situated outside of the collector region 66 are also removed. Then, a P type material, in this case boron, is implanted in the base-region window 72, as indicated at 74, and is also implanted in the exposed regions of the substrate outside of the collector region 66.
Then, as shown in FIG. 21, a third oxide layer 76 is added, and the boron is diffused tc a depth of approxi-mately 1.2 microns, to form the base region 78. The base region has a depth of approximately 1.2 microns, a surface concentration of 1 x 1019 cm 3 and a sheet resistance of 150-200 ohms per square. The P type boron material is also diffused into the surround-ing field region outside of the collector region 66.
, . . .. ~ . .. _ _ . _ .. , _ .. _ , . ,, . . .. . _ . _ . .. . . . . . ~ _ . .. _ _ _ _ ~
S 44'7 .. ..
This field diffusion region, indicated at 80, increases the surface concentration of the substrate 60, and effectively isolates adjacent devices from each other.
Next, as shown in FIG. 22, the third oxide layer 76 is selectively removed, again using the photoresist process, to expose an emitter-region window 82 over the base region 78, and also to expose a continuous rectangular strip 84 (FIG. 15) over the collector region 66. An N+ type phosphorous emitter material is deposited in the exposed areas, and is diffused to a depth of approximately 0.9 micron, to form the emitter region 86, and also to form a contin-uous N+ type region beneath the rectangular strip 84 around the collector region, for purposes of making lS contact with the collector. The surface concentration of the emitter region 86 is approximately 1 x 1021 cm 3 and the sheet res~stance is approximately 18-25 ohms per square.
As shown in FIG. 23, a fourth oxide layer 90 is applied over the already formed layers, and is selectively removed to expose contact holes to be used for making electrical contact with the various semicon-ductor regions. Then mètal contact strips are formed, again using a conventional photoresist process that is not shown ln detail. First, a metalized layer is ; applied over the entire structure, then a photoresist layer is applied over the metal, selectively exposed through a photolith mask, and washed away to leave photoresist material over those areas of metal to be retained. Then the remaining, unmasked areas of metal are etched away to leave the interconnecting metal strips shown in the drawings. In particular, as shown in FIG. 24, there ls a collector contact strip 92, i:~39447 l7 which makes contact with the N+ field diffusion layer - 84 around the outer periphery of the collector area 66, a base contact strip 94, which makes direct contact with the base region 78, and, finally, an emitter contact strip 96, which makes direct contact with the center of the emitter region 86. The final step in the fabrication process is the application of a passi-vating oxide layer 98 over the entire structure.
- The metal contact strips 92, 94 and 96 are single layers approximately 8,000-12,000 angstroms thick (0.8-1.2 microns). Each metalized layer com-prises a thin first coating of titanium, approximately 50 angstroms thick, with the remainder of the layer comprising copper and aluminum. The metal strips are spaced by a minimum of 2 microns from each other, and are a minimum of 5 microns wide, after shrinkage during the fabrication process. All other features o the device have a minimum geometry of 2 microns, which applies to both width and spacing. Thus, the circuit is said to have a atwo-micron feature geome-try~, meaning that, in plan view, all semiconductor regions have a minimum dimension of two microns, measured in the plane of the device, and are spaced rom each other by a minimum dimension o two microns, measured in the same plane. The areas of the semicon-- ductor regions can be dimensioned as desired, within this geometric limitation. For example, the emitter region 86 in one presently preerred embodiment is approximately 5-6 microns wide, and the contact holes are 3 microns wide.
The encoder 40 and latch 44 shown in FIG. ~
can follow any appropriate circuit design. However, an exemplary design is illustrated in FIGS. 4-14b, for an ... .. _ _.. .. . .. _ _ . . . .. _ .. __ _ .. . .. . .. .. . ... _.. . .. _ _ .. _ _ ._ _ _ .
eight-bit encoder. As shown in FIG. ~, the encoder logic comprises four OR gate circuit modules 100, referred to as colu~n O~ gates, and a set of output OR
gates 102, which functions to combine the outputs from the column OR gates. Each of the column OR gate modules 100 has 64 binary inputs and generates a six-bit coded output on six data lines, together with a zero-bit output signal~ which is employed to indicate when the six data bits are all zero. The four 1~ sets of seven-bit outputs are fed over lines 104 to the output OR gates 102, where they are logically combined to produce an eight-bit digital output for transmission to the latch 44.
As shown in FIG. 5, each colu~n-OR gate circuit module 100 comprises a plurality of NPN tran-sistors. More specifically, there are 64 rows, each containing eight such transistors. The collector terminals of all the transistors in the module are commonly connected to ground, as indicated at 106, and each of the 64 input lines is connected to the base terminals of one row of transistors. Four such input lines are shown by way of example in FIG. 5, and these are designated by the signal names COMPOUT/NO. 0, 31, 32 and 63. Also included in a column OR gate are seven output lines, comprising the six data lines 108 and the zero line 110, to which the emitter terminals of the transistors are selectively connected. An eighth parallel line 112, and a corresponding ~column~ of transistors shown in FIG. 5 are not used in the illu-strative eight-bit configuration~
Selective connection of the output lines 108 ~nd 110 to the emitter terminals of the transistors provides for the conversion of a d iscrete voltage 11399~
19 - .
level, indicated by a signal on one of the 64 input lines, to a corresponding binary code. For example, row ~0 of the transistors has the "zero" column tran-sistor emitter ter~inal connected to the "zero~ output line ll0, but none of the other transistors in row t0 has its emitter terminal connected to an output line.
Thus, when input line t0 to the column OR gate module supplies an input signal to the module, only the ~zero~
output line 110 is affected. By way of further example, row ~31 of transistors has five transistors connected to the output data lines 108, to impress on the data lines the pattern 011111, which will be recognized as the binary code for the decimal number 31.
.
15~he complete areangement of transistor connections for each of the column OR gate modules 100 - is shown diagrammatically in ~IG. 6. The vertical lines in FIG. 6 are the output lines 108 and zero line 110, and the horizontal lines numbered 0-63 represent the input lines. The dots represent connections between emitter terminals and output lines. It will be appreciated that not all of the transistors in the column OR gate modules are utilized, and that the unnecessary transistors need not be present at all.
However, for purposes of manufacturing efficiency, it is convenient to fabricate the column OR gate modules all in identical form, each with a full complement of transistors, and then to ~program~ each of the modules, in a metalization step, by appropriately connecting the emitter terminals to the output data lines.
Each of the output lines 108 and 110 from the column OR gates l00 is connected to a sensing circuit of the type shown by way of example in FIG. 7. Thus, .
... .
~, ' .
when a transistor in one of the column OR gates 100 is rendered conductive by a signal impressed at its base terminal, its emitter terminal is connected to one of the outpu~ lines 108, and the resul~ant signal current will be reflected in a change in state in the sensed output signal on line 120 in FIG. 7. The sensed outpùt signals representing all of the column OR gate outputs are then passed to clocked buffers, of which one is shown in FIG. 8, and fro~ which output signals are generated on lines 122, under control of internal clock signals on iines 52 and 52'. Reference voltages on lines 124 and 126, utilized by the clocked buffer circuit of FIG. 8, are generated from a power supply voltage by the circuit shown in FIG. 9.
The buffered data outputs on lines 122 from the clocked buffer circuits are then supplied to the output OR gates 102, shown in detail in FIGS~ 10 and 11. The output OR gates comprise a plurality of NPN
transistors arranged in eight rows, each of 28 transis-tors. The collector terminals of all of the transis-tors are co~monly connec~ed to ground, and the base terminals of ~each of the 28 columns of transistors are connected, respectively, to the 28 input lines 122 from the buffered outputs of the four column OR gates 100.
Eight binary data output lines 130, which appear as horizontal lines in ~IG~ 10, are selectively connected to the emitter terminals of the transistors to provide an eight-bit digital output. More specifically, and as is apparent from FIG. 11, each set of six data lines,
3~ derived from the four column OR gates 100, is essent-~ally ORed to the six corresponding least significant data output lines 130~ In addition, each data input line 122 is ~electively ORed to the two most ~ignifi-cant of the output lines 130, to provide a two-bit code ~, .
4~;' ) indica~ing to which set the input data line belongs.
In addition, the "zero" lines from the column OR gates are also selectively ORed to the two most significant output data lines. The output OR gates effectively
In addition, the "zero" lines from the column OR gates are also selectively ORed to the two most significant output data lines. The output OR gates effectively
5 combin~ the six-bit outputs o~ the column OR gates 100 to provide the desired eight-bit output. It will be appreciated from FIG. 11 that the four sets of input data lines are connected to the two most significant output lines to form the codes 00, 01, 10 and 11, 10 respectively, thus providing the appropriate eight bits of output data.
A sensing circuit, such as that shown in FIG.
12, is required to sense the output on each the data lines 130 from the output OR gates 102, and to provide 15 a sensed output on lines 42. The sensing circuit also supplies a reference voltage level, designated VR2/ORMS
on line 132.
FIG. 13 shows by way of example the detailed design of an output data buffer circuit that serves as 20 the latch 44 (FIG. 1). Input to each of these buffer circuits is derived over line 42 from a corresponding sensing circuit (FIG. 12). Internal clock signals are derived from the circuit shown in FIG. 14, over lines 134 and 134', and a further reerence signal is 25 supplied over line 136. Also supplied to the latch circuit of FIG. 13 are two complementary inversion signals, INV and INV', on lines 138 and 138', respec-tively. The state of these inversion signals provide the means for selective inversion of the signal on line 30 42 prior to storage in the latch for output on line 46.
. ~
4~
The state of the inveesion signals supplied to the ~IG. 13 circuitry is derived either from an external signal NMINV on line 140 (FIG. 13a), for the most siqnificant digit, or from an external signal NLINV on line 142 (FIG. 14). The inversion signals can be utilized for two's complement operations, which necessitate selective inversion of the most significant - bit or the remaining bits.
FIGS. 14a and 14b show a clock signal trans-lation circuit and column clock buffer used to generatevarious clock signal used in the converter, from an external clock signal supplied on line 18. The FIG.
14b circuit generates a related clock signal named ECLK
on line 144, and this is utilized in the FIG. 14a circuit to generate the internal clock signals on lines 52 and 52', and in the FIG. 14 circuit to generate clock signals for the latch, on lines 134 and 134'. It will also be noted that the circuits shown in FIGS. 13, 13a, 14 and 14b all utilize a data-ground signal DGND, also supplied from an external source.
It will be appreciated from the foregoing that the present invention provides a sign~ficant advance in the field of analog-to-digital converters.
In particular, it provides a hitherto unavailable monolithic, parallel, high speed, analog-to-digital converter with accurately matched comparators having relatively low input capacitance and low input bias current. The converter of the invention can be fabri-cated at relatively high production yields, to provide a relatively low cost converter suitable for use in many high speed applications. ~he eight-bit converter disclosed herein by way of example is capable o digitizing an analog signal at rates from zero to 30 megasamples per 6econd, and can process input ~ignals 4~
with frequency components up to 7 MHz. It will also be appreciated that, although this specific embodiment has been illustrated and described in detail by way of example, various modifications may be made without departing from the spirit and scope of the invention.
- Accordingly, the invention is not to be limited except as by the appended claims.
A sensing circuit, such as that shown in FIG.
12, is required to sense the output on each the data lines 130 from the output OR gates 102, and to provide 15 a sensed output on lines 42. The sensing circuit also supplies a reference voltage level, designated VR2/ORMS
on line 132.
FIG. 13 shows by way of example the detailed design of an output data buffer circuit that serves as 20 the latch 44 (FIG. 1). Input to each of these buffer circuits is derived over line 42 from a corresponding sensing circuit (FIG. 12). Internal clock signals are derived from the circuit shown in FIG. 14, over lines 134 and 134', and a further reerence signal is 25 supplied over line 136. Also supplied to the latch circuit of FIG. 13 are two complementary inversion signals, INV and INV', on lines 138 and 138', respec-tively. The state of these inversion signals provide the means for selective inversion of the signal on line 30 42 prior to storage in the latch for output on line 46.
. ~
4~
The state of the inveesion signals supplied to the ~IG. 13 circuitry is derived either from an external signal NMINV on line 140 (FIG. 13a), for the most siqnificant digit, or from an external signal NLINV on line 142 (FIG. 14). The inversion signals can be utilized for two's complement operations, which necessitate selective inversion of the most significant - bit or the remaining bits.
FIGS. 14a and 14b show a clock signal trans-lation circuit and column clock buffer used to generatevarious clock signal used in the converter, from an external clock signal supplied on line 18. The FIG.
14b circuit generates a related clock signal named ECLK
on line 144, and this is utilized in the FIG. 14a circuit to generate the internal clock signals on lines 52 and 52', and in the FIG. 14 circuit to generate clock signals for the latch, on lines 134 and 134'. It will also be noted that the circuits shown in FIGS. 13, 13a, 14 and 14b all utilize a data-ground signal DGND, also supplied from an external source.
It will be appreciated from the foregoing that the present invention provides a sign~ficant advance in the field of analog-to-digital converters.
In particular, it provides a hitherto unavailable monolithic, parallel, high speed, analog-to-digital converter with accurately matched comparators having relatively low input capacitance and low input bias current. The converter of the invention can be fabri-cated at relatively high production yields, to provide a relatively low cost converter suitable for use in many high speed applications. ~he eight-bit converter disclosed herein by way of example is capable o digitizing an analog signal at rates from zero to 30 megasamples per 6econd, and can process input ~ignals 4~
with frequency components up to 7 MHz. It will also be appreciated that, although this specific embodiment has been illustrated and described in detail by way of example, various modifications may be made without departing from the spirit and scope of the invention.
- Accordingly, the invention is not to be limited except as by the appended claims.
Claims (9)
1. A monolithic parallel analog-to-digital converter having an n-bit binary output, where n is at least 5, said conhvlerter comprising:
2n-1comparators;
voltage divider means to provide 2n-1 uniformly graduated reference voltage levels for application to one terminal of each of said compara-tors;
means for coupling an analog input voltage to the other terminal of each of said comparators; and encoding logic means for converting output signals from said comparators to an equivalent n-bit binary signal; and wherein said converter is a monolithic integrated circuit fabricated in accordance with a high-density triple diffusion process providing a high resolution feature geometry and correspondingly small junction depths, to provide high packing density and high production yield.
2n-1comparators;
voltage divider means to provide 2n-1 uniformly graduated reference voltage levels for application to one terminal of each of said compara-tors;
means for coupling an analog input voltage to the other terminal of each of said comparators; and encoding logic means for converting output signals from said comparators to an equivalent n-bit binary signal; and wherein said converter is a monolithic integrated circuit fabricated in accordance with a high-density triple diffusion process providing a high resolution feature geometry and correspondingly small junction depths, to provide high packing density and high production yield.
2. An analog-to-digital converter as set forth in claim 1, wherein said triple diffusion process provides apporximately a two-micron feature geometry.
3. An analog-to-digital converter as set forth in claim 2, wherein said comparators and said encoding logic means include a plurality of NPN tran-sistors.
4. An analog-to-digital converter as set forth in claim 3, wherein each of said transistors has an N type collector region diffused to a depth of approximately 3.5 microns, a P type base region dif-fused to a depth of approximately 1.2 microns into said collector region, and an N+ type emitter region dif-fused to a depth of approximately 0.9 micron into said base region.
5. A monolithic parallel analog-to-digital converter having a resolution of one part in thirty-two or better, said converter comprising:
a substrate on which various circuit elements may be formed by a triple diffusion fabrication process;
a plurality of comparators comprising integrated circuits formed on said substrate by the triple diffusion fabrication process, to provide an acceptably high production yield in spite of a large number of required components;
a precision voltage divider also formed on said substrate, to provide a plurality of discrete reference voltage levels for connection to said com-parators for comparison with an analog voltage level to be converted; and encoding logic also formed on said substrate, to convert output signals from said comparators to an equivalent digital code.
a substrate on which various circuit elements may be formed by a triple diffusion fabrication process;
a plurality of comparators comprising integrated circuits formed on said substrate by the triple diffusion fabrication process, to provide an acceptably high production yield in spite of a large number of required components;
a precision voltage divider also formed on said substrate, to provide a plurality of discrete reference voltage levels for connection to said com-parators for comparison with an analog voltage level to be converted; and encoding logic also formed on said substrate, to convert output signals from said comparators to an equivalent digital code.
6. An analog-to-digital converter as set forth in claim 5, wherein said encoding logic includes:
first logic means for generating a l-in-n code indicative of the analog voltage level, where n is at least 32; and second logic means for generating a multiple-hit binary code equivalent to the l-in-n code.
first logic means for generating a l-in-n code indicative of the analog voltage level, where n is at least 32; and second logic means for generating a multiple-hit binary code equivalent to the l-in-n code.
7. An analog-to-digital converter as set forth in claim 5, wherein said triple diffusion process provides approximately a two-micron feature geometry.
8. An analog-to-digital converter as set forth in claim 7, wherein said comparators and encoding logic include a plurality of NPN transistors.
9. An analog-to-digital converter as set forth in claim 8, wherein each of said transistors has an N type collector region diffused to a depth of approximately 3.5 microns, a P type base region dif-fused to a depth of approximately 1.2 microns into said collector region, and an N+ type emitter region dif-fused to a depth of approximately 0.9 micron into said base region.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US021,577 | 1979-03-19 | ||
| US06/021,577 US4276543A (en) | 1979-03-19 | 1979-03-19 | Monolithic triple diffusion analog to digital converter |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CA1139447A true CA1139447A (en) | 1983-01-11 |
Family
ID=21805013
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CA000347177A Expired CA1139447A (en) | 1979-03-19 | 1980-03-06 | Monolithic analog-to-digital converter |
Country Status (7)
| Country | Link |
|---|---|
| US (1) | US4276543A (en) |
| JP (1) | JPS55159626A (en) |
| CA (1) | CA1139447A (en) |
| DE (1) | DE3009434A1 (en) |
| FR (1) | FR2452206B1 (en) |
| GB (1) | GB2047997A (en) |
| SE (1) | SE8002074L (en) |
Families Citing this family (92)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4441096A (en) * | 1980-06-23 | 1984-04-03 | California Institute Of Technology | Optical analog-to-digital converter |
| US4590463A (en) * | 1980-09-29 | 1986-05-20 | Rca Corporation | Digital control of color in CRT display |
| JPS59107629A (en) * | 1982-12-10 | 1984-06-21 | Matsushita Electric Ind Co Ltd | Analog-digital converter |
| JPS60102027A (en) * | 1983-11-09 | 1985-06-06 | Nec Corp | Digital-analog converter |
| JPS6188619A (en) * | 1984-09-28 | 1986-05-06 | シーメンス、アクチエンゲゼルシヤフト | Direct current-alternating current converter |
| US4633226A (en) * | 1984-12-17 | 1986-12-30 | Black Jr William C | Multiple channel analog-to-digital converters |
| KR930001724B1 (en) * | 1985-03-25 | 1993-03-12 | 가부시기가이샤 히다찌세이사꾸쇼 | Resistor and electronic device |
| JPH0758910B2 (en) * | 1985-05-27 | 1995-06-21 | 松下電器産業株式会社 | Parallel A / D converter |
| JPH0681050B2 (en) * | 1985-08-21 | 1994-10-12 | 株式会社日立製作所 | Parallel AD converter |
| JPS6387022A (en) * | 1986-09-30 | 1988-04-18 | Toshiba Corp | Analog/digital converter |
| US4749984A (en) * | 1987-06-29 | 1988-06-07 | Rca Corporation | Subranging A/D converter with converging digitally controlled voltages |
| JPH0773213B2 (en) * | 1987-11-20 | 1995-08-02 | 三菱電機株式会社 | A / D converter |
| JPH0787371B2 (en) * | 1988-02-09 | 1995-09-20 | 松下電器産業株式会社 | Parallel A / D converter |
| US4866001A (en) * | 1988-07-01 | 1989-09-12 | Bipolar Integrated Technology, Inc. | Very large scale bipolar integrated circuit process |
| DD278237A1 (en) * | 1988-12-16 | 1990-04-25 | Halbleiterwerk Kom Mikro Elekt | FLASH ANALOG / DIGITAL CONVERTER |
| US5119098A (en) * | 1989-06-20 | 1992-06-02 | Sony Corporation | Full flash analog-to-digital converter |
| US5012246A (en) * | 1990-01-31 | 1991-04-30 | International Business Machines Corporation | BiCMOS analog-to-digital converter with minimized metastability |
| US5157397A (en) * | 1991-01-28 | 1992-10-20 | Trw Inc. | Quantizer and related method for improving linearity |
| FR2756101B1 (en) * | 1996-11-19 | 1999-02-12 | Sgs Thomson Microelectronics | METHOD FOR MANUFACTURING AN NPN TRANSISTOR IN BICMOS TECHNOLOGY |
| FR2756100B1 (en) | 1996-11-19 | 1999-02-12 | Sgs Thomson Microelectronics | BIPOLAR TRANSISTOR WITH INHOMOGENEOUS TRANSMITTER IN A BICMOS INTEGRATED CIRCUIT |
| FR2756099B1 (en) * | 1996-11-19 | 1999-02-12 | Sgs Thomson Microelectronics | PROCESS FOR MANUFACTURING AN NPN MINIMUM SURFACE TRANSISTOR |
| FR2756103B1 (en) * | 1996-11-19 | 1999-05-14 | Sgs Thomson Microelectronics | MANUFACTURE OF BIPOLAR / CMOS INTEGRATED CIRCUITS AND A CAPACITOR |
| AU2001225485A1 (en) * | 2000-01-11 | 2001-07-24 | Hamamatsu Photonics K.K. | X-ray image sensor |
| JP4360010B2 (en) | 2000-04-27 | 2009-11-11 | ソニー株式会社 | Parallel type analog-digital converter |
| US6710733B2 (en) | 2001-06-29 | 2004-03-23 | Sony Corporation | Comparator circuit |
| US7132972B2 (en) * | 2003-06-27 | 2006-11-07 | The Trustees Of Columbia University In The City Of New York | Continuous-time digital signal generation, transmission, storage and processing |
| WO2006063192A1 (en) * | 2004-12-07 | 2006-06-15 | The Trustees Of Columbia University In The City Of New York | Systems and methods for continuous-time digital modulation |
| FR2893611B1 (en) * | 2005-11-23 | 2007-12-21 | Commissariat Energie Atomique | PROCESS FOR PRODUCING A COATING BASED ON OXIDE CERAMIC COMPRISING THE GEOMETRY OF A SUBSTRATE HAVING RELIEF PATTERNS |
| US9288089B2 (en) | 2010-04-30 | 2016-03-15 | Ecole Polytechnique Federale De Lausanne (Epfl) | Orthogonal differential vector signaling |
| US9450744B2 (en) | 2010-05-20 | 2016-09-20 | Kandou Lab, S.A. | Control loop management and vector signaling code communications links |
| US9106238B1 (en) * | 2010-12-30 | 2015-08-11 | Kandou Labs, S.A. | Sorting decoder |
| US9362962B2 (en) | 2010-05-20 | 2016-06-07 | Kandou Labs, S.A. | Methods and systems for energy-efficient communications interface |
| US9596109B2 (en) | 2010-05-20 | 2017-03-14 | Kandou Labs, S.A. | Methods and systems for high bandwidth communications interface |
| US9288082B1 (en) | 2010-05-20 | 2016-03-15 | Kandou Labs, S.A. | Circuits for efficient detection of vector signaling codes for chip-to-chip communication using sums of differences |
| US9564994B2 (en) | 2010-05-20 | 2017-02-07 | Kandou Labs, S.A. | Fault tolerant chip-to-chip communication with advanced voltage |
| US9985634B2 (en) | 2010-05-20 | 2018-05-29 | Kandou Labs, S.A. | Data-driven voltage regulator |
| US9077386B1 (en) | 2010-05-20 | 2015-07-07 | Kandou Labs, S.A. | Methods and systems for selection of unions of vector signaling codes for power and pin efficient chip-to-chip communication |
| US9251873B1 (en) | 2010-05-20 | 2016-02-02 | Kandou Labs, S.A. | Methods and systems for pin-efficient memory controller interface using vector signaling codes for chip-to-chip communications |
| US8593305B1 (en) | 2011-07-05 | 2013-11-26 | Kandou Labs, S.A. | Efficient processing and detection of balanced codes |
| US9246713B2 (en) | 2010-05-20 | 2016-01-26 | Kandou Labs, S.A. | Vector signaling with reduced receiver complexity |
| US9401828B2 (en) | 2010-05-20 | 2016-07-26 | Kandou Labs, S.A. | Methods and systems for low-power and pin-efficient communications with superposition signaling codes |
| US9275720B2 (en) | 2010-12-30 | 2016-03-01 | Kandou Labs, S.A. | Differential vector storage for dynamic random access memory |
| US9268683B1 (en) | 2012-05-14 | 2016-02-23 | Kandou Labs, S.A. | Storage method and apparatus for random access memory using codeword storage |
| EP2926260B1 (en) | 2013-01-17 | 2019-04-03 | Kandou Labs S.A. | Methods and systems for chip-to-chip communication with reduced simultaneous switching noise |
| CN105122758B (en) | 2013-02-11 | 2018-07-10 | 康杜实验室公司 | High bandwidth interchip communication interface method and system |
| EP2979388B1 (en) | 2013-04-16 | 2020-02-12 | Kandou Labs, S.A. | Methods and systems for high bandwidth communications interface |
| CN105393512B (en) | 2013-06-25 | 2019-06-28 | 康杜实验室公司 | Vector signaling with low receiver complexity |
| US9106465B2 (en) | 2013-11-22 | 2015-08-11 | Kandou Labs, S.A. | Multiwire linear equalizer for vector signaling code receiver |
| US9806761B1 (en) | 2014-01-31 | 2017-10-31 | Kandou Labs, S.A. | Methods and systems for reduction of nearest-neighbor crosstalk |
| US9369312B1 (en) | 2014-02-02 | 2016-06-14 | Kandou Labs, S.A. | Low EMI signaling for parallel conductor interfaces |
| US9100232B1 (en) | 2014-02-02 | 2015-08-04 | Kandou Labs, S.A. | Method for code evaluation using ISI ratio |
| WO2015131203A1 (en) | 2014-02-28 | 2015-09-03 | Kandou Lab, S.A. | Clock-embedded vector signaling codes |
| US9509437B2 (en) | 2014-05-13 | 2016-11-29 | Kandou Labs, S.A. | Vector signaling code with improved noise margin |
| US9148087B1 (en) | 2014-05-16 | 2015-09-29 | Kandou Labs, S.A. | Symmetric is linear equalization circuit with increased gain |
| US9852806B2 (en) | 2014-06-20 | 2017-12-26 | Kandou Labs, S.A. | System for generating a test pattern to detect and isolate stuck faults for an interface using transition coding |
| US9112550B1 (en) | 2014-06-25 | 2015-08-18 | Kandou Labs, SA | Multilevel driver for high speed chip-to-chip communications |
| WO2016007863A2 (en) | 2014-07-10 | 2016-01-14 | Kandou Labs, S.A. | Vector signaling codes with increased signal to noise characteristics |
| US9432082B2 (en) | 2014-07-17 | 2016-08-30 | Kandou Labs, S.A. | Bus reversable orthogonal differential vector signaling codes |
| US9444654B2 (en) | 2014-07-21 | 2016-09-13 | Kandou Labs, S.A. | Multidrop data transfer |
| EP3175592B1 (en) | 2014-08-01 | 2021-12-29 | Kandou Labs S.A. | Orthogonal differential vector signaling codes with embedded clock |
| US9674014B2 (en) | 2014-10-22 | 2017-06-06 | Kandou Labs, S.A. | Method and apparatus for high speed chip-to-chip communications |
| KR102517583B1 (en) | 2015-06-26 | 2023-04-03 | 칸도우 랩스 에스에이 | High speed communications system |
| US9557760B1 (en) | 2015-10-28 | 2017-01-31 | Kandou Labs, S.A. | Enhanced phase interpolation circuit |
| US9577815B1 (en) | 2015-10-29 | 2017-02-21 | Kandou Labs, S.A. | Clock data alignment system for vector signaling code communications link |
| US10055372B2 (en) | 2015-11-25 | 2018-08-21 | Kandou Labs, S.A. | Orthogonal differential vector signaling codes with embedded clock |
| US10003315B2 (en) | 2016-01-25 | 2018-06-19 | Kandou Labs S.A. | Voltage sampler driver with enhanced high-frequency gain |
| WO2017185070A1 (en) | 2016-04-22 | 2017-10-26 | Kandou Labs, S.A. | Calibration apparatus and method for sampler with adjustable high frequency gain |
| CN115051705A (en) | 2016-04-22 | 2022-09-13 | 康杜实验室公司 | High performance phase locked loop |
| US10003454B2 (en) | 2016-04-22 | 2018-06-19 | Kandou Labs, S.A. | Sampler with low input kickback |
| CN109417521B (en) | 2016-04-28 | 2022-03-18 | 康杜实验室公司 | Low power multi-level driver |
| WO2017189931A1 (en) | 2016-04-28 | 2017-11-02 | Kandou Labs, S.A. | Vector signaling codes for densely-routed wire groups |
| US10153591B2 (en) | 2016-04-28 | 2018-12-11 | Kandou Labs, S.A. | Skew-resistant multi-wire channel |
| US9906358B1 (en) | 2016-08-31 | 2018-02-27 | Kandou Labs, S.A. | Lock detector for phase lock loop |
| US10411922B2 (en) | 2016-09-16 | 2019-09-10 | Kandou Labs, S.A. | Data-driven phase detector element for phase locked loops |
| US10200188B2 (en) | 2016-10-21 | 2019-02-05 | Kandou Labs, S.A. | Quadrature and duty cycle error correction in matrix phase lock loop |
| US10200218B2 (en) | 2016-10-24 | 2019-02-05 | Kandou Labs, S.A. | Multi-stage sampler with increased gain |
| US10372665B2 (en) | 2016-10-24 | 2019-08-06 | Kandou Labs, S.A. | Multiphase data receiver with distributed DFE |
| US10116468B1 (en) | 2017-06-28 | 2018-10-30 | Kandou Labs, S.A. | Low power chip-to-chip bidirectional communications |
| US10686583B2 (en) | 2017-07-04 | 2020-06-16 | Kandou Labs, S.A. | Method for measuring and correcting multi-wire skew |
| US10203226B1 (en) | 2017-08-11 | 2019-02-12 | Kandou Labs, S.A. | Phase interpolation circuit |
| US10326623B1 (en) | 2017-12-08 | 2019-06-18 | Kandou Labs, S.A. | Methods and systems for providing multi-stage distributed decision feedback equalization |
| US10554380B2 (en) | 2018-01-26 | 2020-02-04 | Kandou Labs, S.A. | Dynamically weighted exclusive or gate having weighted output segments for phase detection and phase interpolation |
| US10931249B2 (en) | 2018-06-12 | 2021-02-23 | Kandou Labs, S.A. | Amplifier with adjustable high-frequency gain using varactor diodes |
| EP3808044A1 (en) | 2018-06-12 | 2021-04-21 | Kandou Labs, S.A. | Passive multi-input comparator for orthogonal codes on a multi-wire bus |
| CN113169714A (en) | 2018-09-10 | 2021-07-23 | 康杜实验室公司 | Programmable continuous time linear equalizer with stable high frequency peaking function for controlling slicer working current |
| US10574487B1 (en) | 2019-04-08 | 2020-02-25 | Kandou Labs, S.A. | Sampler offset calibration during operation |
| US10608849B1 (en) | 2019-04-08 | 2020-03-31 | Kandou Labs, S.A. | Variable gain amplifier and sampler offset calibration without clock recovery |
| US10680634B1 (en) | 2019-04-08 | 2020-06-09 | Kandou Labs, S.A. | Dynamic integration time adjustment of a clocked data sampler using a static analog calibration circuit |
| US10721106B1 (en) | 2019-04-08 | 2020-07-21 | Kandou Labs, S.A. | Adaptive continuous time linear equalization and channel bandwidth control |
| US11303484B1 (en) | 2021-04-02 | 2022-04-12 | Kandou Labs SA | Continuous time linear equalization and bandwidth adaptation using asynchronous sampling |
| US11374800B1 (en) | 2021-04-14 | 2022-06-28 | Kandou Labs SA | Continuous time linear equalization and bandwidth adaptation using peak detector |
| US11456708B1 (en) | 2021-04-30 | 2022-09-27 | Kandou Labs SA | Reference generation circuit for maintaining temperature-tracked linearity in amplifier with adjustable high-frequency gain |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US3885994A (en) * | 1973-05-25 | 1975-05-27 | Trw Inc | Bipolar transistor construction method |
| US3981072A (en) * | 1973-05-25 | 1976-09-21 | Trw Inc. | Bipolar transistor construction method |
| CA1105143A (en) * | 1976-02-12 | 1981-07-14 | Robert A. Nordstrom | Parallel analog-to-digital converter |
-
1979
- 1979-03-19 US US06/021,577 patent/US4276543A/en not_active Expired - Lifetime
-
1980
- 1980-03-05 GB GB8007403A patent/GB2047997A/en not_active Withdrawn
- 1980-03-06 CA CA000347177A patent/CA1139447A/en not_active Expired
- 1980-03-12 DE DE19803009434 patent/DE3009434A1/en active Granted
- 1980-03-17 SE SE8002074A patent/SE8002074L/en unknown
- 1980-03-18 JP JP3468280A patent/JPS55159626A/en active Pending
- 1980-03-18 FR FR8006091A patent/FR2452206B1/en not_active Expired
Also Published As
| Publication number | Publication date |
|---|---|
| GB2047997A (en) | 1980-12-03 |
| SE8002074L (en) | 1980-11-03 |
| FR2452206B1 (en) | 1986-03-28 |
| JPS55159626A (en) | 1980-12-11 |
| DE3009434A1 (en) | 1980-10-02 |
| DE3009434C2 (en) | 1989-12-07 |
| US4276543A (en) | 1981-06-30 |
| FR2452206A1 (en) | 1980-10-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CA1139447A (en) | Monolithic analog-to-digital converter | |
| Van De Plassche et al. | A monolithic 14-bit D/A converter | |
| US4533903A (en) | Analog-to-digital converter | |
| US5952952A (en) | Switched-capacitor array | |
| US5949362A (en) | Digital-to-analog converter including current cell matrix with enhanced linearity and associated methods | |
| EP0070734B1 (en) | Analog-to-digital converters | |
| US4087812A (en) | Digital-to-analog and analog-to-digital converter circuit | |
| Maio et al. | An untrimmed D/A converter with 14-bit resolution | |
| JP2007142379A (en) | Analog/digital converter | |
| Nejime et al. | An 8-b ADC with over-Nyquist input at 300-Ms/s conversion rate | |
| Weiss | A 1 Gs/s 8-bit GaAs DAC with on-chip current sources | |
| Kolluri | A 12-bit 500-ns subranging ADC | |
| EP0082736B1 (en) | Analogue to digital converter | |
| US20040075116A1 (en) | Transistor array and method of layout | |
| US4831281A (en) | Merged multi-collector transistor | |
| US4866444A (en) | Analog-to-digital converter | |
| JP3028420B2 (en) | Semiconductor integrated device | |
| Emmert et al. | A versatile bipolar monolithic 6-bit A/D converter for 100 MHz sample frequency | |
| JPH0119474Y2 (en) | ||
| Tanner et al. | An 8-bit low-power ADC array for CMOS image sensors | |
| JP2692955B2 (en) | Parallel A / D converter | |
| Kuo | Superconducting A/D converters based on Josephson binary counters | |
| US5317312A (en) | Analog-to-digital converter of an annular configuration | |
| US3786493A (en) | Analog to digital converter using a drift transistor | |
| Watanabe et al. | A 400-MHz DA converter with a 4-bit color map for 2000-line display |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| MKEX | Expiry |