JP2970024B2 - How to measure junction temperature - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for measuring a junction temperature of a semiconductor device, and more particularly to a method for measuring a junction temperature of a Schottky diode.

[0002]

2. Description of the Related Art A junction temperature refers to a temperature of a junction (for example, a PN junction) provided in a semiconductor device and is given as a sum of an environmental temperature and a temperature rise due to Joule heat generated in the junction. The junction temperature is one of the ratings of the semiconductor device. If the junction temperature exceeds the rating, an increase in leakage current, a decrease in long-term reliability, or a breakdown occurs. Therefore, measurement of the junction temperature is essential when designing an IC or when actually constructing a system using the IC.

Conventionally, there is a diode method as a method for measuring a junction temperature of a semiconductor device. This is because the forward voltage of the diode is very sensitive to changes in junction temperature and changes regularly with good reproducibility. Taking advantage of this feature, some diodes are also used in temperature sensors.

Equation 1 below is an equation representing the forward current-voltage characteristics of a Schottky diode.

[0005]

(Equation 1)

In the above equation 1, _IF is the forward current, S is the junction area, A is the effective Richardson constant, T is the absolute temperature, q is the electron charge, Φ _B is the Schottky barrier height, k is Boltzmann's constant, V _F is the forward voltage,
n represents an ideal factor.

[0007] By transforming equation (1),

[0008]

(Equation 2)

## EQU1 ## In this equation 2, k, q, A
Is a constant, and Φ _B , S, and n are constants specific to individual diodes. Strictly speaking, Φ _B and n have temperature dependence, but from room temperature, which we usually need, to around 150 ° C.,
It can be considered almost constant. That is, in Equation 2, can be regarded as almost constant except absolute temperature T, it is possible to know the forward voltage V _F from the absolute temperature (junction temperature) T.

[0010]

In a compound semiconductor device such as GaAs, a Schottky barrier diode utilizing a potential barrier generated at a contact portion between a metal electrode and a semiconductor is widely used. However, the characteristics of the Schottky barrier diode greatly depend on the state of the interface between the metal electrode and the semiconductor, and the uniformity of the compound semiconductor substrate is not as good as that of silicon.

The state of this interface is expressed by the ideal factor n
Is reflected in In Equation 2 n is Kakari' to all terms, variations in the n is directly reflected in the forward voltage V _F. To make this easier to understand, differentiating Equation 2 with the absolute temperature T gives:

[0012]

(Equation 3)

## EQU1 ## Temperature coefficient as the forward voltage V _F is the change in ideality n in formulas 3 T. C. Is reflected in That status of the interface is a difference for any reason, the value of the ideal factor n is different for each diode, the temperature coefficients of the forward voltage V _F T. C. Will be different. And this is not unusual, it is a common occurrence.

[0014] Figure 4, for three Schottky diodes taken from the same wafer, a graph of the temperature dependence of the forward voltage V _F. Temperature dependence of the forward voltage V _F is slightly different in the diode 1-3, which is due to the variation described above. Thus the temperature dependence of the forward voltage V _F is mixed, by changing the ambient temperature each time measured from measuring the temperature dependency of the forward voltage V _F, the measurement of newly junction temperature T Had to do.

[0015] Accordingly, determined by temperature dependence measurement of the forward voltage V _F is very time consuming, which hinders the efficiency of junction temperature measurement.

Therefore, the present invention provides a method for promoting the efficiency of measuring the junction temperature.

[0017]

A method of measuring a junction temperature according to the present invention is to measure a current-voltage of a diode to be measured at room temperature, and to obtain a forward voltage and an ideal factor of the diode at room temperature. From a chart showing the relationship between the temperature coefficient indicating the change in the forward voltage with temperature and the ideality factor, a measurement preparation stage for reading the temperature coefficient corresponding to the value of the ideality factor, and actually want to measure the junction temperature of the diode Measure the forward voltage under the circumstances, the value of the forward voltage,
The method further comprises a main measurement step of obtaining a junction temperature from the value of the forward voltage obtained in the measurement preparation step and the temperature coefficient read in the measurement preparation step.

The above chart shows the correlation between the ideality factor obtained from the current-voltage characteristics of the diode measured at room temperature and the value obtained by linearly approximating the temperature coefficient of the forward voltage measured by actually changing the ambient temperature. It was created by taking

[0019]

In the measurement preparation stage, the ideal factor of the diode and the temperature coefficient corresponding thereto can be estimated, and based on these values, the junction temperature can be calculated in the main measurement stage. Therefore, the temperature dependency of the forward voltage can be predicted without actually measuring, and the junction temperature can be calculated.

Further, by using a chart created by correlating the ideal factor with the temperature coefficient, it is possible to estimate the temperature coefficient for each diode without actually measuring it.

[0021]

FIG. 1 is a flowchart of a method for measuring a junction temperature according to the present invention. First to perform current-voltage measurements at room temperature of the diode to be measured, obtaining the ideality factor n and the forward voltage V _F at room temperature. Next, the ideality factor n and the temperature coefficient T.T. C. (FIG. 2)
(Shown), the temperature coefficient T.C. corresponding to the value of the ideality factor n just obtained. C. Read. The above is the measurement preparation stage.

Next, a main measurement stage is entered. Indeed placed under conditions to be measured the junction temperature diode, measuring the forward voltage V _F. The value of the forward voltage V _F of this time, in the previous measurement preparation stage, the forward voltage V _F obtained at room temperature
And the estimated temperature coefficient T. C. From Equation (3), the target junction temperature can be obtained by Expression 3.

As a result of a detailed study of Equation (3), the change in the parentheses in Equation (3) is small from room temperature, which is a temperature range to be measured, to about 150 ° C., and a large error does not occur even if it is regarded as almost constant. Furthermore, if the value in parentheses can be regarded as constant, the temperature coefficient T.E. C. Is proportional to the ideality factor n. By measuring this ideality factor n at room temperature, the temperature coefficient T. C. Can be estimated. Therefore, the ideal factor n actually measured at room temperature and the ambient temperature actually
° C. After varied from 0.99 ° C. The temperature coefficient of the forward voltage V _F T. C. After investigating the correlation between the values obtained by linear approximation of
As shown in FIG. 2, it was confirmed that there was a strong correlation with the correlation coefficient of 0.988. This correlation diagram is the aforementioned chart.

Then, another diode is used to determine the value of the ideality factor n at room temperature, and the temperature coefficient T.T. C. Was estimated. Next, by actually changing the ambient temperature of the diode, the forward voltage V
The temperature dependence of _F was measured. Fig. 3 shows the results.
It is.

In FIG. 3, the solid line represents the ideality factor n at room temperature.
_Is the temperature dependence of the forward voltage V _F estimated from the value of
The plot is the temperature dependency measured by actually changing the ambient temperature. At this time, the difference between the estimated value and the actually measured value was within ± 1 ° C. at the maximum.

[0026]

According to the method of measuring the junction temperature according to the present invention, the temperature dependency of the forward voltage can be predicted from the very simple measurement result of the ideality factor at room temperature. As a result, the measurement of the temperature dependency of the forward voltage, which is conventionally required, becomes unnecessary, and the junction temperature measurement can be efficiently performed in a short time.

In addition, it is not necessary to obtain a temperature coefficient or the like for each diode by actual measurement, so that a more efficient junction temperature measurement can be performed.

[Brief description of the drawings]

FIG. 1 is a flowchart of a joint temperature measuring method according to the present invention.

FIG. 2 is a diagram showing a correlation between an ideal factor at room temperature and an actually measured temperature coefficient.

FIG. 3 is a diagram showing the coincidence of the temperature dependence of the forward voltage obtained by the junction temperature measurement method according to the present invention with the actually measured temperature dependence.

FIG. 4 is a diagram showing that the temperature dependency of a forward voltage varies due to a variation in an ideal factor in a conventional measurement method.

Claims (2)

(57) [Claims] 1. A temperature coefficient and an ideality factor indicating a change in a forward voltage with temperature after a current-voltage measurement of a diode to be measured is performed at room temperature to determine a forward voltage and an ideality factor of the diode at room temperature. From the chart showing the relationship with
A measurement preparation step for reading a temperature coefficient corresponding to the value of the ideality factor, and measuring a forward voltage under a situation where the junction temperature of the diode is to be actually measured, and a value of the forward voltage.
A method of measuring the junction temperature, comprising: a main measurement step of determining the junction temperature from the value of the forward voltage obtained in the measurement preparation step and the temperature coefficient read in the measurement preparation step. . 2. The chart shows a correlation between an ideal factor obtained from a current-voltage characteristic of the diode measured at room temperature and a value obtained by linearly approximating a temperature coefficient of a forward voltage measured by actually changing an ambient temperature. 2. The method for measuring a junction temperature according to claim 1, wherein the method is performed by using the following method.