Manual Microwind

MICROWIND APPLICATION NOTE

Simulating Process Variations

Introducing Process Variation in Microwind35
Etienne SICARD Professor INSA-Dgei, 135 Av de Rangueil 31077 Toulouse – France www.microwind.org email: Etienne.sicard@insa-toulouse.fr

This application note describes the effects of process variations, and how it can simulated using Microwind35. The DC simulation exhibits ION/IOFFvariations, and the time-domain simulation shows up to 100% variations of the switching performances depending on the MIN/TYP/MAX mode.

1. MOS Variability
One important challenge in nano-CMOS technology is process variability. The fabrication of millions of MOS devices at nano-scale induces a spreading in switching performances in the same IC. Fig. 1 illustrates the huge difference between theinitial drawing (Fig. 1-a) as it appears in Microwind layout window and the real-case aspect of a selection of MOS devices (Fig. 1-b,c,d). The most important parameters affected by process variability are the threshold voltage, the carrier mobility and the effective channel length. Variations are handled in Microwind using random values in a Gaussian distribution, which is expressed by theformulation of Equ. 2. An illustration of the variation in electron mobility (parameter U0) for nearly a thousand MOS device samples is reported in Fig. 2.

(a) Initial drawing

(b) MOS device 1

(b) MOS device 2

(c) MOS device 3

Page 1/8

etienne.sicard@insa-toulouse.fr

31/08/10

MICROWIND APPLICATION NOTE

Simulating Process Variations

Figure 1: Initial drawing (a) and finalaspect of the MOS device depending on the location and immediate environment (b,c,d)
N-Channel MOS mobility (x 102 V. cm -2)

Value

Deviation

U0=320 V.cm-2
4.0

Mean value
3.0 U0=320 V.cm-2

Probability
2.0

1.0

0.0

MOS sample number

Figure 2 : The extracted mobility of 1000 MOS samples shows an important variability around 320 V.cm-2
( x−m) 2 2σ 2

f =
where

1 2π σe

−

(Equ. 2)

f = probability density of the random variable (x axis in Fig. 11) x = variable (y axis in Fig. 11) σ = deviation m = mean value

2. DC Simulation of process variability
In Microwind, use the command Simulate → Mos characteristics, and click “Ioff vs. Ion”. The software computes 100 DC characteristics Ids vs. Vds for varying VT, U0 and LD. For each Id/Vd curve, ION andIOFF are extracted and added to the drawing.

Page 2/8

etienne.sicard@insa-toulouse.fr

31/08/10

MICROWIND APPLICATION NOTE

Simulating Process Variations

Figure 3: Access to process variability - DC characteristics

Figure 4: Ioff/Ion calculated by Microwind on 100 samples of n-channel MOS with random distribution of VT, U0, and LINT with a Gaussian distribution around the nominalvalue

Page 3/8

etienne.sicard@insa-toulouse.fr

31/08/10

MICROWIND APPLICATION NOTE

Simulating Process Variations

Very high Ioff current

High speed

Low leakage Average Ioff current Average trend in this 32-nm technology Very low Ioff current Very low Ion current

Average Ion current

Very high Ion current

Figure 5: Finding compromises between high current drive andhigh leakage current The effect of process variability on the MOS Ioff/Ion characteristics is plotted using the menu “Ioff vs. Ion” under the “MOS I/V curve” menu (Fig. 3). It can be seen that the MOS devices have a wide variability in performances. The 3 MOS types (low leakage, high speed, high voltage) [1] are situated in well defined space in the Ioff/Ion domain. The low leakage is in themiddle (medium Ion, low Ioff), the high speed on the upper right corner (high Ion, high Ioff), and the high voltage is at the lower left side of the graphics (low Ion, very low Ioff). Note that the exact locations of the dots will change for each MOS characteristics plotted because it is a random process.

3. Time-domain Simulation of process variability
Single simulation
In Microwind, use the...