Skip to content

Latest commit

 

History

History
121 lines (62 loc) · 5.32 KB

Report.md

File metadata and controls

121 lines (62 loc) · 5.32 KB

4-bit Carry Lookahead Adder

The top level design of the carry lookahead adder is as follows

readme_media/Screen_Shot_2021-04-19_at_1.13.02_AM.png

The input and outputs follow the clock pulse as shown

readme_media/Screen_Shot_2021-04-19_at_1.13.22_AM.png

In a CLA adder we input the 4 bit numbers and a carry-in. In the first stage of the CLA we generate the propagate and generate signals for each bit of the input as follows

readme_media/Screen_Shot_2021-04-29_at_10.58.42_PM.png

and the carry out of the ith bit position can be generated as follows

readme_media/Screen_Shot_2021-04-29_at_10.59.05_PM.png

The Sum can be generated as follows

readme_media/Screen_Shot_2021-04-29_at_10.59.10_PM.png

Structure for adder

Derivation

We have

readme_media/Screen_Shot_2021-04-29_at_10.59.26_PM.png

Let the input carry be $C_0$.

using the equation for finding further carries we get

readme_media/Screen_Shot_2021-04-29_at_10.59.45_PM.png

We have the carries of the form

readme_media/Screen_Shot_2021-04-29_at_10.59.55_PM.png

The truth table will be as follows

readme_media/Screen_Shot_2021-04-26_at_11.00.52_PM.png

It is seen that XOR and OR differ only when both A and B are 1 but in this case i+1th carry will be independent of P as G is AND which will be 1 which will be 1 and i+1th carry is given by OR of G and AND of P and ith carry. Hence we can replace the XOR operations in P with an OR operation in the carry lookahead calculation segment of the circuit.

readme_media/Screen_Shot_2021-04-29_at_11.00.06_PM.png

We can see that this carry lookahead design would require the use of AND, OR and XOR functions. We know that complementary logic gates are inverting for CMOS static logic. Hence it will be easier and more efficient to make NAND and NOR gates in static CMOS logic than AND and OR because they would require inverting the outputs of NAND and NOR gates respectively.

We thereby modify the logic functions using the DeMorgan's theorem to replace AND and OR with NAND and NOR wherever possible.

readme_media/Screen_Shot_2021-04-29_at_11.00.13_PM.png

Using these equations we can arrive at the following structure for the circuit.

Diagram

readme_media/Screen_Shot_2021-04-27_at_11.47.34_PM.png

Design details

Gates

The width of the transistors in each of the gates have been decided to match that of the ideal inverter in order to obtain equal rise and fall times and minimize delays and uncertainties.

NOT

readme_media/Screen_Shot_2021-04-26_at_8.28.56_PM.png

readme_media/Screen_Shot_2021-04-27_at_11.52.14_PM.png

NAND

readme_media/Screen_Shot_2021-04-26_at_8.29.11_PM.png

readme_media/Screen_Shot_2021-04-27_at_11.57.23_PM.png

NOR

readme_media/Screen_Shot_2021-04-26_at_8.32.13_PM.png

readme_media/Screen_Shot_2021-04-28_at_12.07.40_AM.png

XOR

readme_media/Screen_Shot_2021-04-26_at_8.32.29_PM.png

readme_media/Screen_Shot_2021-04-28_at_12.08.13_AM.png

D-flip-flop

The D-flipflop will consist of NAND gates arranged as follows

readme_media/Screen_Shot_2021-04-28_at_6.01.47_PM.png

readme_media/Screen_Shot_2021-04-28_at_12.31.13_AM.png

Floor plan

Initial floor plan

readme_media/Screen_Shot_2021-04-28_at_2.47.12_PM.png

Initial layout

readme_media/Screen_Shot_2021-04-28_at_2.49.34_PM.png

Revised floor plan

readme_media/Screen_Shot_2021-04-28_at_2.46.59_PM.png

Revised layout

readme_media/Screen_Shot_2021-04-28_at_8.45.15_AM.png

References

  1. J. Miao and S. Li, "A novel implementation of 4-bit carry look-ahead adder," 2017 International Conference on Electron Devices and Solid-State Circuits (EDSSC), 2017, pp. 1-2, doi: 10.1109/EDSSC.2017.8126457.
  2. CMOS VLSI Design (fourth edition) by Weste and Harris.
  3. Pai, Y. and Y. Chen. “The fastest carry lookahead adder.” Proceedings. DELTA 2004. Second IEEE International Workshop on Electronic Design, Test and Applications (2004): 434-436.