Coded data generation or conversion – Analog to or from digital conversion – Digital to analog conversion
Reexamination Certificate
2003-01-08
2004-03-09
Williams, Howard L. (Department: 2819)
Coded data generation or conversion
Analog to or from digital conversion
Digital to analog conversion
C341S136000
Reexamination Certificate
active
06703956
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates generally to digital-to-analog conversion of electrical signals, and more particularly to a high-precision, low-current current-steering digital-to-analog converter.
BACKGROUND OF THE INVENTION
A digital-to-analog converter (DAC) converts a digital input signal represented by a number of bits N to a substantially equivalent analog output signal. The analog output signal is typically a voltage or current signal.
A typical DAC for converting the digital input signal into an analog current output signal is a current-steering DAC. In one type of current-steering DAC, a reference voltage is converted into binary-weighted currents. For example, in one such current-steering DAC, switches couple each input of parallel coupled binary weighted resistors to either a reference voltage or ground depending on whether the digital input signal bit is a logic 1 level or a logic 0 level. The outputs of each of the binary weighted resistors are coupled to an output current node.
Another example of a current-steering DAC employs an R-2R ladder coupled to a reference voltage. Switches are coupled in parallel to outputs of the R-2R resistor ladder. The switches are controlled by digital input signal bits to couple the current from the outputs of the R-2R resistor ladder to ground when the digital input signal bit is at a logic 0 level or to a current output node when the digital input signal bit is at a logic 1 level.
SUMMARY OF THE INVENTION
The present invention provides a digital-to-analog converter (DAC) and a method for converting a digital input signal having N bits to a substantially equivalent analog current output signal. The DAC of the invention utilizes bit segmentation and shared reference current(s) to allow high precision, high linearity, and a high input signal bit count resulting from the ability to ensure CMOS FET saturation even in the submicron and deep submicron process technologies.
In accordance with the invention, the N-bit binary digital input signal is segmented into at least a first group of i bits and a second group of j bits. The DAC of the invention includes a first current-steering digital-to-analog converter configured to receive the first group of i bits and a first reference current that produces a first digital-to-analog converter output current representation based on the value of the first group of i digital bits. The DAC also includes at least a second current-steering digital-to-analog converter configured to receive the second group of j bits of the N-bit binary digital input signal and the first reference current. The second current-steering digital-to-analog converter produces a second digital-to-analog converter output current representative of a value of the second group of j bits. The second digital-to-analog converter output current is scaled down by an order of magnitude substantially equal to 2
j
by a current scaling circuit. The DAC includes a summing circuit which sums at least the first digital-to-analog converter output current and the scaled down version of the second digital-to-analog converter output current to produce an analog current signal representative of the value of the N-bit binary input signal.
In a preferred embodiment, the DAC includes a binary-to-thermometer encoder dedicated to each segmented group i, j of digital input signal bits. A first binary-to-thermometer encoder converts the first group of i bits to a thermometer code represented by 2
i
thermometer code bits, and a second binary-to-thermometer encoder converts the second group of j bits to a thermometer code represented by 2
j
thermometer code bits. Additional binary-to-thermometer encoders may be used if the digital input signal of N bits is further segmented into additional segmented groups of bits k
x
.
The DAC also includes a group of identical current source arrays dedicated to each segmented group i, j of digital input signal bits. More particularly, the DAC includes at least a first array of 2
i
identical current sources each of which is selectable by a respective one of the 2
i
thermometer code bits to contribute an identical current referenced by a first reference current to an output line. The sum of the identical currents contributed by the first array of 2
i
identical current sources is therefore summed on the output line. The DAC further includes at least a second array of 2
j
identical current sources each of which is selectable by a respective one of the 2
j
thermometer code bits to contribute an identical current referenced by a first reference current to an intermediate output line. The sum of the identical currents contributed by the second array of 2
j
identical current sources is therefore summed on the intermediate output line.
The DAC also includes a current dividing circuit which divides the total current present on the intermediate output line by 2
j
. The divided current is added to the current on the output line. Accordingly, an analog current signal representation of the digital input signal is produced on the output line of the DAC.
In a preferred embodiment, a 16-bit DAC is implemented with a 4-6-6 bit grouping, with the 4 highest significant bits encoded to generate a 2
4
-bit thermometer code, the next 6 highest significant bits encoded to generate a 2
6
-bit thermometer code, and the lowest 6 significant bits encoded to generate a 2
6
-bit thermometer code. In this embodiment, the high thermometer code bits are fed to a high current source array comprising 2
4
current sources. Each current source is selectable by a different one of the 2
4
bits of the high thermometer code. When a high thermometer code bit is asserted, its respective current source generates a current substantially equal to a high reference current.
The mid thermometer code bits are fed to a mid current source array comprising 2
6
current sources. Each current source in the mid current source array is selectable by a different one of the 2
6
bits of the mid thermometer code. When a mid thermometer code bit is asserted, its respective current source generates a current substantially equal to a mid reference current.
The low thermometer code bits are fed to a low current source array comprising 2
6
current sources. Each current source in the low current source array is selectable by a different one of the 2
6
bits of the low thermometer code. When a low thermometer code bit is asserted, its respective current source generates a current substantially equal to the mid reference current. The total current contributed by each of the low current sources is summed on an intermediate output line, which is fed to a divider circuit that divides the summed outputs from the current sources in the low current source array by 2
6
to scale down the current.
The total current contributed by each of the selected high current sources, each of the selected mid current sources, and the current output by the current divider is summed to produce an analog current signal representation of a given 16-bit digital input signal.
The segmentation and sharing of a mid-level reference current with the lowest-level segmented DAC allows the current-steering DAC of the invention to handle subthreshold current LSB's in a submicron process while maintaining very good linearity. In particular, this entails the sharing of references (Low, Mid) so the current mirror arrays are “identical” and the divide-by-64 current circuit that generates the lowest 6 bits of the 16-bit DAC.
REFERENCES:
patent: 5446455 (1995-08-01), Brooks
patent: 5703587 (1997-12-01), Clark et al.
patent: 5844511 (1998-12-01), Izumikawa
patent: 6037889 (2000-03-01), Knee
patent: 6166670 (2000-12-01), O'Shaughnessy
patent: 6204790 (2001-03-01), Jin et al.
patent: 6225929 (2001-05-01), Beck
patent: 6424283 (2002-07-01), Bugeja et al.
Teddy Loeliger and Walter Guggenbuhl, “Cascode circuits for switched current copiers”, Midwest Symposium on Circuits and Systems, Aug. 1997, pp. 256-259.
K.Ola Andersson and J.Jacob Wikner, “Characterization of a CMOS current-steering DA
Duffner Barbara J.
Mueller Brian
Agilent Technologie,s Inc.
Williams Howard L.
LandOfFree
Technique for improved linearity of high-precision,... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Technique for improved linearity of high-precision,..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Technique for improved linearity of high-precision,... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3283345