A proposed successive approximation analog-to-digital converter (ADC) would contain a capacitively terminated chain of identical capacitor cells (see figure). Like a conventional successive approximation ADC containing a bank of binary scaled capacitors, the proposed ADC would store an input voltage on a sample and hold capacitor and would digitize the stored input voltage by finding the closest match between this voltage and a capacitively generated sum of binary fractions of a reference voltage (Vref). However, the proposed capacitor-chain ADC would offer two major advantages over a conventional binary scaled-capacitor ADC:

  • In a conventional ADC that digitizes to n bits, the largest capacitor (representing the most significant bit) must have 2n–1 times as much capacitance, and hence, approximately 2n–1 times as much area as does the smallest capacitor (representing the least significant bit), so that the total capacitor area must be 2n times that of the smallest capacitor. In the proposed capacitor-chain ADC, there would be three capacitors per cell, each approximately equal to the smallest capacitor in the conventional ADC, and there would be one cell per bit. Therefore, the total capacitor area would be only about 3n times that of the smallest capacitor. The net result would be that the proposed ADC could be considerably smaller than the conventional ADC.
  • Because of edge effects, parasitic capacitances, and manufacturing tolerances, it is difficult to make capacitor banks in which the values of capacitance are scaled by powers of 2 to the required precision. In contrast, because all the capacitors in the proposed ADC would be identical, the problem of precise binary scaling would not arise.

In the proposed ADC, as in the conventional capacitor-chain ADC, a sampled version of the input voltage is sent together with the capacitively generated sum of binary fractions of Vref into the two input terminals of a comparator. Also as in the conventional ADC, the sum of binary fractions of Vref is generated by electronically switching capacitor connections between ground and Vref. The comparator determines whether the generated voltage is higher or lower than the input voltage, and by switching in or out successively smaller binary fractions of Vref, the ADC brackets the input voltage between successively converging binary values.

A Capacitor-Chain Successive-Approximation ADC, shown here in simplified fom, would contain a capacitively terminated chain of identical capacitor cells. A sum of binary-weighted voltages would be generated by throwing switches, starting in the leftmost cell for the most-significant bit and proceeding rightward through successive cells to the least-significant bit.
Each cell could be characterized as containing two or three capacitors, depending on one's perspective: One capacitor would have the minimum capacitance, C, while the other would have a capacitance of 2C. In practice, the 2C capacitor could be fabricated as two 1C capacitors in parallel, so that a cell would contain three identical capacitors, of which two would be tied together.

Only the first cell would be tied directly to the inverting terminal of the comparator. The second cell would be connected to the first cell, the third cell connected to the second, and so forth, forming a chain of identical cells. The first and last cells of the chain would be terminated to ground via capacitors of 2C and C, respectively. The chain would constitute a capacitive voltage-divider network containing switched parallel and series connections. By an algebraic derivation from basic circuit theory, it can be shown that switching from the ground connection to the Vref connection in each successive cell of this network would contribute the next less-significant binary fraction of Vref to the input terminal of the comparator. More specifically, the voltage contribution obtained by such switching in the mth cell would be given by

ÆVn = 2–m(Vref/3).

This work was done by Thomas Cunningham of Caltech for NASA's Jet Propulsion Laboratory. For further information, access the Technical Support Package (TSP) free on-line at www.techbriefs.com/tsp under the Computers/Electronics category.

In accordance with Public Law 96-517, the contractor has elected to retain title to this invention. Inquiries concerning rights for its commercial use should be addressed to

Intellectual Property group

Mail Stop 202-233
4800 Oak Grove Drive
Pasadena, CA 91109
(818) 354-2240

Refer to NPO-21145, volume and number of this NASA Tech Briefs issue, and the page number.

This Brief includes a Technical Support Package (TSP).
Capictor-Chain Successive-Approximation ADC

(reference NPO-21145) is currently available for download from the TSP library.

Don't have an account? Sign up here.

NASA Tech Briefs Magazine

This article first appeared in the July, 2003 issue of NASA Tech Briefs Magazine.

Read more articles from the archives here.