Introduction to VLSI Design:

 

Definitions: VLSI = Very Large Scale Integration

VLSI > 10k tranzistors on a chip

LSI > 1K tranzistors on a chip

MSI > 100 tranzistors on a chip

SSI > 10 tranzistors on a chip

Nota bene: Different sources = different definitions!

 

 

Terminology differences:

TRANSISTOR COUNT

versus

GATE COUNT

 

What is the difference:

Silicon

NMOS 1+FANIN

PMOS 2*FANIN

ECL 3+FANIN

GaAs

E/D-MESFET 1+FANIN

Nota bene: The term DEVICE has two meanings

TRANSISTORS

Or

GATE

 

The VLSI Design Methodologies:

 

  1. Geometric symbols

    2. FC or FULL CUSTOM (designing N mask levels)

     

  2. Logic symbols

    3. SC or STANDARD CELL (designing N mask levels)

    Semi-custom versus macro-cell

    GA or GATE ARRAYS (designing only 1 or 2 wiring

    related mask levels; other mask levels

    prefabricated)

    GTL elements versus RTL elements

    PL or PROGRAMMBLE LOGIC

    (all mask levels prefabricated)

    GTL elements versus RTL elements

     

  3. Behavior symbols

ST or SILICOM TRANSLATION

(translation to any of the above)

 

Figure 4.1. Sketch of the semiconductor wafer containing a large number of chips designed using the SC (Standard Cells) VLSI methodology. The channels containing standard cells have the same width, and are not equidistant. The channels containing connections have varying width-their widths are governed by the number and the structure of connections that have to be made. They take up a relatively small proportion of the chip area (around 33% in this sketch). The total chip area is as big as the implemented logic requires (the chip utilization is always close to 100%)

 

 

Figure 4.2. Sketch of the chip designed using GA (Gate Array) VLSI methodology. The channels that contain the gates are of the same width, and are equidistant. The channels that contain the connection are also of the same width-their widths are predetermined during the prefabrication process. They take up more chip area (around 50% of the sketch). The total chip area is always bigger then the area required by the logic, because the prefabricated chips of standard dimensions are used (the chip utilization is virtually always below 100%-the designers are usually very happy if they reach 90%)

 

 

Figure 4.3. Sketch of the chip designed using the PL (Programmable Logic) VLSI methodology. The areas containing the macrocells (shaded) take up only the fraction of the chip area (around 25% in the sketch). The channels containing the connections take up the rest of the area (around 75% in the sketch), and their intersections contain the interconnection networks that are controlled using either RAM (XILINX) or ROM (ALTERA) memory elements. The sketch shows one of the many possible internal architectures that are frequently implemented. The total chip area is always bigger than area required by logic, because the prefabricated chip of standard dimensions are used (the chip utilization is virtually always below 100%-the designers are usually very happy if they reach 80%)

 

The Programmable Logic VLSI:

 

1. Altera(Untel)

2. Xilinx

3. Cypress + Actel + Lattice

4. Plessey + Ferranti

 

The ALTERA Approach to PL VLSI:

 
    1. Hardware

      2. 4 families

       

    2. Software

4 corresponding families, all of the same structure
      1. Schematic entry
      2. Logic and timing testing
      3. Placements and routing (trivial)

 

    1. Programmation (JEDEC)
    2. Board population and testing

 

 

Figure 4.4. The dependency of the cost with respect to the number of chips manufactured. The C-axis represents the total cost of the design and manufacturing processes, for the series comprising of Nsp VLSI chips. The value C0 that corresponds to Nsp = 0 represents the sum of the two components. They are: the cost of the design, and the cost of all the masks that have to be made after the design is complete.

 

The FULL CUSTOM Approach to VLSI

 

 

Each device technology and each design technique are characterized with a certain number of VLSI mask levels.

Different VLSI mask levels correspond to different technological actions in the fabrication processes

Each mask level is defined via a set of non-overlapping poligons. Poligons from different mask levels do overlap, according to some min/max rules

With the structured design approaches, the minimal geometry (lambda) is defined. The min/max values of overlapping are defined via the lambda rules

 

 

CROSS SECTION OF PHYSICAL STRUCTURE

(SIDE VIEW)

MASK (TOP VIEW)

PTUB MASK

 

Figure 4.5a. The processes of implementation of a silicon CMOS transistor, and geometric structure of the corresponding VLSI masks: mask #1.

 

 

CROSS SECTION OF PHYSICAL STRUCTURE

(SIDE VIEW)

MASK (TOP VIEW)

THINOXIDE MASK

 

Figure 4.5b. The processes of implementation of a silicon CMOS transistor, and the geometric structure of the corresponding VLSI masks: mask #2.

 

 

CROSS SECTION OF PHYSICAL STRUCTURE

(SIDE VIEW)

MASK (TOP VIEW)

POLYSILICON MASK

 

Figure 4.5c. The processes of implementation of a silicon CMOS transistor, and the geometric structure of corresponding VLSI masks: mask #3

 

 

 

CROSS SECTION OF PHYSICAL STRUCTURE

(SIDE VIEW)

MASK (TOP VIEW)

p-PLUS MASK (POSITIVE)

 

Figure 4.5d. The processes of implementation of a silicon CMOS transistor, and the geometric structure of corresponding VLSI masks: mask #4

 

 

CROSS SECTION OF PHYSICAL STRUCTURE

(SIDE VIEW)

MASK (TOP VIEW)

p-PLUS MASK (NEGATIVE)

 

 

Figure 4.5e. The processes of implementation of a silicon CMOS transistor, and the geometric structure of corresponding VLSI masks: mask #5

 

 

CROSS SECTION OF PHYSICAL STRUCTURE

(SIDE VIEW)

MASK (TOP VIEW)

CONTACT MASK

 

 

Figure 4.5f. The processes of implementation of a silicon CMOS transistor, and the geometric structure of corresponding VLSI masks: mask #6

 

 

CROSS SECTION OF PHYSICAL STRUCTURE

(SIDE VIEW)

MASK (TOP VIEW)

  

METAL MASK

 

 

 

Figure 4.5g. The processes of implementation of a silicon CMOS transistor, and the geometric structure of corresponding VLSI masks: mask #7

 

NO. MASK FEATURE DIMENSION

1 Thinox A1. Minimum thinox width 2l

A2. Minimum thinox spacing ( to , to ) 2l

A3. Minimum p-thinox to n-thinox spacing 8l

2 p-well B1. Minimum p-well width 4l

B2. Minimum p-well spacing (wells at same potential) 2l

B3. Minimum p-well spacing (wells at different potential) 6l

B4. Minimum distance to internal thinox 3l

B5. Minimum distance to external thinox 5l

3 Poly C1. Minimum poly width 2l

C2. Minimum poly spacing 2l

C3. Minimum poly to thinox spacing l

C4. Minimum poly gate extension 2l

C5. Minimum thinox source/drain extension 2l

4 p-plus D1. Minimum overlap of thinox 1,5–2l

D2. Minimum p-plus spacing 2l

D3. Minimum gate overlap or distance to gate edge 1,5–2l

D4. Minimum spacing to unrelated thinox 1,5–2l

5 Contact E1. Minimum contact area 2l ´ 2l

E2. Minimum contact to contact spacing 2l

E3. Minimum overlap of thinox or poly over contact l

E4. Minimum spacing to gate poly 2l

E5. source/drain contact

E6. source/drain contact

E7. contact See reference [WesEsh85]

E8. contact

E9. Split contact

E10. Split contact

6 Metal F1. Minimum metal width 2–3l

F2. Minimum metal spacing 3l

F3. Minimum metal overlap of contact l

 

 

Figure 4.6. The Lambda rules that have to be satisfied when forming the geometrical structures for various mask levels in the implementation of a silicon CMOS transistor.

 

Examples:

lambda = 3 microns for the first 32-bit Silicon RISC

lambda = 1 microns for the first 32-bit GaAs RISC

lambda < 1 microns for the modern submicron

superscalar/superpipelined RISC maschines

 

Advantages of the full custom approach:

Chip area (can be minimized by carefull design)

Speed (direct and indirect impacts)

Disipation (the application dependent issue)

Price of the fabrication (2n; for n between 2 and 3)

 

Drawbacks:

Engineering time (can be maximized by careful design)

Price of the design (culture dependant)

Long idea to market time (the 6/50 rule)

Unfavorable learning curve (interdisciplinary

knowledge required)

 

What is being designed with the FC method:

  1. Standard cells for the FC VLSI

    2. Semi- custom standard cells

    N-input one-bit logic gates

    One-bit flip-flops

    Macro-cell standard cells

    (De)multiplexers, (De)coders,…

    Adders, multipliers,…

  2. Standard cell extensions for the existing

    3. standard cell families

    Open CAD versus closed CAD

  3. Highly-repetitive on-chip structures

    4. Memory chips,…

    Systolic arrays,…

  5. Not so highly-repetitive structures,

    6. to be fabricated in enormous quantities

    Popular microprocessors

    Specialized microcontrollers

  6. None of the above, but simple

Glue logic

Small specialized logic

 

What is being designed with the SC method:

 

  1. Structures which are not "simple"

    2.  

  2. Structures which are not to be

fabricated in enormous quantities

 

Examples:

 

  1. Prototype of the first 32=bit GaAs RISC for DARPA
  2. Specialized I/O RISCs

 

Comments:

 

  1. The 95/5 without_FC rule
  2. The 95/5 per_project versus per_chip rule

 

Examples:

 

The TANNER RESEARCH Tools for MOSIS

Orcad (S-Edit) + GateSim + L-Edit + the SCMOS cell family

The CADDAS Tools of RCA

Valid + Logsim + MPD2D+12 standard cells families