Robust Analog Design for Automotive Applications by Design Centering with Safe Operating Areas

Abstract

The effects of random variations during the manufacturing process on devices can be simulated as a variation of transistor parameters. Device degradation, due to temperature or voltage stress, causes a shift of device parameters, for example threshold voltage V th , which can also be modeled as a degradation of transistor parameters. Therefore, in order to design circuits, which are robust and reliable, analysis and optimization of their sensitivity to variations in model parameters is important. Furthermore, constraints on the operating regions and voltage differences of transistors are used in order to keep operating points stable over a large temperature range. In this work, using two circuits for automotive applications and current process development kits (PDK), we show how design centering software can be used to consider both sensitivity reduction towards model parameter variation and constraints to control safe operating areas (SOA). Beyond that a comparison of the constraint matrix method with two established methods of SOA checking is done.

Lifetime Yield Optimization of Analog Circuits Considering Process Variations and Parameter Degradations

“…Taking the process variation as second order effects on the sensitivity towards degradation, C(t) is assumed to be constant, i.e., C(t)=C (Sobe et al, 2009). …”

“…which differs from (5) in (Sobe et al, 2009). From (37) it is clear that the evaluation of the worst-case distance degradation rate for a performance upper bound involves only multiple sensitivity evaluations which can be carried out efficiently.…”

A Cross-Layer Approach to Measure the Robustness of Integrated Circuits

“…It can model robustness across all levels of abstraction from transistor to system level, both in the analog and digital domain. This broad applicability is the main difference compared to other approaches to measure robustness in fields like computer science [Higham 1996;Eslamnour and Ali 2009], automotive electronics [SAE 2008], or analog circuits [Sobe et al 2008] and underlines its cross-layer capabilities.…”

Robustness validation of integrated circuits and systems

“…Examples are the robust ness of numerical algorithms [2] or the use of the term in control theory [3] or temporal logic [4]. Also, the robust ness of workstation clusters [5] and even of analog circuits [6] have been discussed before.…”