RU2465612C2 - Method for comparative assessment of transistor batches by reliability - Google Patents

Abstract

FIELD: electricity.

SUBSTANCE: on identical samples from compared batches of transistors, intensity of noise is measured in different transitions after fourfold exposure to electrostatic discharges with a potential, starting with the permissible one, then increasing it by not more than twice with number of exposures equal to accordingly 5, 4, 2, 1 of different polarity. By difference of noise intensity values on transitions, the potential reliability of compared transistor batches is identified.

EFFECT: non-destructive check of transistor batches by reliability criterion.

2 tbl, 1 dwg

Description

The invention relates to microelectronics, in particular to ensuring the reliability of transistors, and can be used to comparatively evaluate the reliability of batches of transistors in terms of resistance with electrostatic discharges (ESD) at the stages of production and use.

It is known [1] that with increasing supply voltage of ESD to semiconductor products, including transistors, their electrical parameters change for the worse.

A known method [2] of a comparative assessment of the reliability of batches of transistors by the influence of the ESR value on the appearance of parametric and catastrophic failures. In this case, ESR potential is applied to the transistors, which is twice as large as permissible according to the technical specifications (TU) with its subsequent increase stepwise by 20-30 V.

The disadvantage of this method is the need for ESD exposure, starting with the doubled value of the discharge voltage permissible according to the technical specifications and continuing to increase the voltage until a parametric or catastrophic failure occurs.

измеряются на различных переходах до испытаний и после каждого последующего воздействия ЭСР. По разности значений интенсивности шума на переходах коллектор-эмиттер и коллектор-база судят о потенциальной надежности сравниваемых партий транзисторов.The invention is aimed at eliminating this drawback, namely, in the proposed method, the ESD is supplied separately to the emitter-base and collector-base transitions with a potential, starting from the permissible, and then increasing it, but not more than twice, with the number of effects, respectively, 5, 4 , 2, 1 of different polarity. Low-frequency noise values

measured at various transitions before testing and after each subsequent exposure to ESD. The difference in the values of the noise intensity at the collector-emitter and collector-base junctions judges the potential reliability of the compared batches of transistors.

при прямом рабочем токе 3 мА до и после воздействия ЭСР. Воздействие ЭСР осуществляли по модели тела человека с потенциалом 500 В (допустимое значение по ТУ), 650 В, 850 В, 1000 В с числом воздействий ЭСР соответственно 5, 4, 2, 1 различной полярности. Результаты измерений

, а также значения изменения величины

(разность напряжения шума эмиттерного и коллекторного перехода) после воздействий ЭСР приведены в таблицах 1 и 2.The method was tested on 10 transistors type KT838A (high-power bipolar transistors pnp conductivity), which measured the magnitude of the noise intensity

at a direct operating current of 3 mA before and after exposure to ESD. The impact of ESD was carried out according to the model of a human body with a potential of 500 V (permissible value according to TU), 650 V, 850 V, 1000 V with the number of ESR effects, respectively, 5, 4, 2, 1 of different polarity. Measurement results

, as well as the values of the change in value

(the difference between the noise voltage of the emitter and collector junction) after exposure to ESD are shown in tables 1 and 2.

при различных напряжениях с целью представления изменения низкочастотного шума с увеличением напряжения. Практически замеряют

начальное и после 1000 В. На второй партии замеры проведены только после воздействия ЭСР напряжением 1000 В (см. рис.).Tables 1 and 2 show the measurement results.

at various voltages in order to represent the change in low-frequency noise with increasing voltage. Practically measure

initial and after 1000 V. On the second batch, measurements were carried out only after exposure to an ESD of 1000 V (see Fig.).

Since in this method the effect of ESD is carried out with a voltage exceeding the permissible value, it can be used only for a selective assessment of transistor batches by potential reliability. It can be seen from the figure that party I is more reliable than party II.

Information sources

1. Gorlov M.I., Emelyanov A.V., Plebanovich V.I. Electrostatic charge in electronics // Mn .: Bel. Science, 2006 .-- 295 p.

2. RF patent No. 2226698, IPC G01R 31/26. Published April 10, 2004. Bull. No. 10.

Claims (1)

A method for comparatively evaluating the batches of transistors in terms of reliability, which consists in measuring the noise intensity at different transitions of transistors after exposure to electrostatic discharges, characterized in that the electrostatic discharges supply at least four voltages separately to the emitter-base and collector-base junctions: admissible and exceeding the permissible technical conditions twice and with two intermediate voltages - with the number of impacts, respectively, 5, 4, 2, 1, of different polarity.