Skip to main navigation Skip to search Skip to main content

Direct Extraction of InP/GaAsSb/InP DHBT Equivalent-Circuit Elements From S-Parameters Measured at Cut-Off and Normal Bias Conditions

    Research output: Contribution to journalJournal articleResearchpeer-review

    1 Downloads (Orbit)

    Abstract

    A unique direct parameter extraction method for the small-signal equivalent-circuit model of InP/GaAsSb/InP double heterojunction bipolar transistors (DHBTs) is presented. $S$-parameters measured at cut-off bias are used, at first, to extract the distribution factor $X_{0}$ for the base-collector capacitance at zero collector current and the collector-to-emitter overlap capacitance $C_{ceo}$ present in InP DHBT devices. Low-frequency $S$-parameters measured at normal bias conditions then allows the extraction of the external access resistances $R_{bx}$, $R_{e}$, and $R_{cx}$ as well as the intrinsic HBT elements of the device. The terminal inductances of the device are extracted from high frequency $S$-parameters by employing the intrinsic HBT elements extracted at low-frequency. Compared to other published direct parameter extraction techniques the proposed method is developed specifically for III-V based HBTs and avoids $S$ -parameters measured at the critical open-collector bias condition. The method is applied to an $1.5~mu{rm m}$ emitter width InP/GaAsSb/InP DHBT device and leads to excellent prediction of the measured $S$ -parameters in the 250 MHz – 65 GHz frequency range.
    Original languageEnglish
    JournalI E E E Transactions on Microwave Theory and Techniques
    Volume64
    Issue number1
    Pages (from-to)115-124
    ISSN0018-9480
    DOIs
    Publication statusPublished - 2016

    Keywords

    • Direct parameter extraction
    • GaAsSb
    • Heterojunction bipolar transistor (HBT)
    • InP
    • Small-signal equivalentcircuit model

    Fingerprint

    Dive into the research topics of 'Direct Extraction of InP/GaAsSb/InP DHBT Equivalent-Circuit Elements From S-Parameters Measured at Cut-Off and Normal Bias Conditions'. Together they form a unique fingerprint.

    Cite this