CN104748885B - The method that LED junction temperature is measured based on I V characteristic curves - Google Patents

Abstract

The method that LED junction temperature is measured based on I V characteristic curves, is related to the junction temperature measurement of LED.LED junction temperature device is measured based on I V characteristic curves using one kind, the device is provided with current source, test LED samples, temperature control is heat sink and temperature control instrument, computer.Measurement saturation current I_sWith non-lit up LED junction temperature T_j, and it is fitted I_s‑T_jCurve map；Measure the VA characteristic curve of LED；LED junction temperature is calculated using MATLAB programming softwares.It is that can obtain LED junction temperature only to need simple voltage source, current source test reverse saturation current and forward direction I V characteristic curves, improve LED junction temperature test process, avoid forward voltage method that the process of test low current is switched to from the high current of heating, and can realize being switched fast without instrument and the highly difficult test of gathered data in time, simplify the junction temperature measurement method of LED.

Description

Method based on I-V characteristic curved measurement LED junction temperature

Technical field

The present invention relates to the junction temperature measurement of LED, more particularly, to a kind of side based on I-V characteristic curved measurement LED junction temperature Method.

Background technology

LED illumination is 21 century lighting engineering most with prospects.LED has brightness high, long lifespan, volume because of it The features such as small, safety and environmental protection many fields such as display, backlight, illumination instead of conventional light source [M.H.Crawford, “LEDs for solid-state lighting:Performance challenges and recent advances,” IEEE J.Sel.Topics Quantum Electron.,2009.15(4)：1028–1040].From unlike conventional light source, LED is a kind of semiconductor devices.Its thermal losses mechanism depends critically upon heat transfer, and non-radiative and convection current.The basic knot of LED Structure is a P-N junction for semiconductor.When there is electric current to flow through LED component, P-N junction active area free electron produces multiple with hole Close, and discharge luminous energy and heat energy [E.F.Schubert, Light-Emitting Diodes.Cambridge, U.K: Cambridge Univ.Press,2006.].When heat is constantly assembled, the temperature in PN junction interface will rise.Say in the narrow sense, The temperature in PN junction interface is defined as the junction temperature of LED.But the chip of usual LED component is all smaller, so as to the temperature of LED chip Degree is considered as the junction temperature of LED.It was verified that the change of LED junction temperature can influence its luminous flux, light extraction efficiency, color, wavelength and just To the luminosity such as voltage, colourity and electric parameter, life and reliability [N.Narendran the and Y.Gu, " Life of device are influenceed of LED-Based White Light Sources"IEEE Journal of Display Technology,No 1,pg 167-(sept 2005).].However, because heat dissipation problem is not settled properly so that LED easily produces light decay, color A series of problems, such as declining, causes the performance of LED to be had a greatly reduced quality.The research of junction temperature is related to the evaluation of LED light source, thermal design and The prediction in life-span, therefore, the junction temperature of LED quickly and accurately measure just seem significant.

At present, the measuring method of conventional LED junction temperature includes：Forward voltage method, heat resistance method, peak wavelength method, blue white ratio Method, infrared photography method, relative radiation degree method and Finite Element etc..Current these method of testings are very numerous and jumbled, without unification Testing standard.Some methods need a whole set of test system, and method is cumbersome；Some methods need some high testers Device, it is expensive.Wherein, forward voltage method is presently considered to be with the most extensive, test method of testing the most accurate.Here Advantage and disadvantage only to positive voltage method are described, and it will not go into details for remaining.Forward voltage method is using the PN junction electronic transport of LED Temperature effect, junction temperature [the big base in field, Guan Rongfeng, king's apricot high power white lights are calculated by measuring the forward voltage under operating current Junction temperature measurement [J] the electronics of LED and encapsulation, 8 (12):10-14].Specific practice is：Measurement LED is (general under smaller current Take 10mA) linear relationship (k-factor) of forward voltage and junction temperature；Put down when LED reaches heat in normal operating conditions (high current) During weighing apparatus state, high current now is switched into low current, the voltage at now LED two ends is measured with most fast speed, according to electricity The corresponding k-factor of variable quantity and junction temperature of pressure calculates junction temperature.Forward voltage method has certainty of measurement high, method of testing letter Single, transient response is good and the advantages of non-destructive.But it is required that test system can be switched to test moment from the high current of heating Low current, noise and damped oscillation are readily incorporated during this, and require that switching time is more short better.Due to noise and The switching time interval of the presence of damped oscillation, even very little is inconsistent all can be very big to the generation of the accuracy of measurement result Influence.

The content of the invention

It is an object of the invention to the measurement for differing and existing for the standard for overcoming existing LED junction temperature measurement The problem of noise and damped oscillation can be introduced during accuracy problem, especially forward voltage method current switching, there is provided a kind of Method based on I-V characteristic curved measurement LED junction temperature.

This hair is inhaled and is based on I-V characteristic curved measurement LED junction temperature device using a kind of, and the device is provided with current source, test LED Sample, temperature control are heat sink and temperature control instrument, computer.

The present invention is comprised the following steps：

1) measurement saturation current I_sWith non-lit up LED junction temperature T_j, and it is fitted I_s-T_jCurve map；

In step 1) in, the measurement saturation current I_sWith non-lit up LED junction temperature T_j, and it is fitted I_s-T_jThe tool of curve map Body method can be：By LED samples be placed on temperature control it is heat sink on, LED two ends connect with current source two ends, and temperature control is heat sink and temperature controller Device connects.Heat sink temperature is set with temperature control instrument, such as 25 DEG C, crosses 5min when LED reaches thermal balance, it is defeated that current source is set Go out backward voltage (typically -5V), now LED does not light and do not generate heat, LED junction temperature is heat sink temperature, test LED two ends Reverse current I_s；Regulation temperature control instrument changes heat sink temperature, repeats above-mentioned test process, obtains under a series of different junction temperatures Reverse saturation current I_s；The saturation current I obtained with the fitting test of computer origin data processing softwares_sWith LED junction temperature T_j, obtain I_s-T_jCurve, fit equation is：

Wherein, I_sIt is reverse current, I_s0(T_j0) represent T_j0Reverse current at=25 DEG C, T_jAs LED junction temperature, β₁、β₂、 β₃It is fitting coefficient.

2) VA characteristic curve of LED is measured；

In step 2) in, the specific method of the VA characteristic curve of the measurement LED can be：LED samples are placed on control On warm is heavy, LED two ends connect with current source two ends, and temperature control is heat sink to connect with temperature control instrument.Set heat sink with temperature control instrument Temperature, makes heat sink temperature keep constant, such as 25 DEG C；Now, forward current I is provided to LED samples by current source, using four Line measuring method test forward voltage V；In experiment under the usual multiple forward currents of test (such as 1-500mA is spaced 1mA) just VA characteristic curve I-V (the heat sink temperature T of forward current I and forward voltage V are obtained to voltage V_sink1It is fixed, T_sink1=25 ℃)；Then change heat sink temperature, such as 35 DEG C, repeat same experimentation, obtain another group of I-V characteristic curve (heat sink temperature Degree T_sink2It is fixed, T_sink2=35 DEG C).

3) it is calculated LED junction temperature using MATLAB programming softwares.

In step 3) in, the specific method that the utilization MATLAB programming softwares are calculated LED junction temperature can be：Heat sink Temperature is T_sink1Under, according to equation：

It is T in heat sink temperature_sink2Under, according to equation：

And there is the junction temperature temperature difference to be equal to the heat sink temperature difference：T_j2-T_j1=T_sink2-T_sink1 (3)

Wherein I₁Represent heat sink temperature T_sink1The forward current of LED, V at=25 DEG C₁Represent heat sink temperature T_sink1=25 DEG C The forward voltage of lower LED, T_sink1It is on T_j1Function, can be according to step 2) in equation obtain, q is unit charge band Electricity, k is Boltzmann constant, and n represents ideality factor, is unknown, T_j1As require in heat sink temperature T_sink1At=25 DEG C Junction temperature when LED is lighted；Similarly, I₂Represent heat sink temperature T_sink2The forward current of LED, V at=25 DEG C₂Represent heat sink temperature T_sink2The forward voltage of LED, I at=25 DEG C_s2(T_j2) it is on T_j2Function, can be according to step 2) in equation obtain, T_j2As we require in heat sink temperature T_sink2Junction temperature when LED is lighted at=35 DEG C；By MATLAB program calculations, ask Obtain T_j1, T_j2And n.

The present invention solves the know-why that is used of its technical problem：

Generally, the C-V characteristic of LED meets Shockley equations, is designated as：

Wherein, I₀It is forward current, I_sIt is saturation current, q represents the quantity of electric charge of electronics, and V represents the forward direction electricity at LED two ends Pressure, n is ideality factor, and k is Boltzmann constant, T_jThat is the junction temperature of LED；Generally under the normal working conditions of LED, Exponentially change, meetSo can be omitted " -1 ", equation can be write as：

When the backward voltage applied to LED close to but during no more than its breakdown voltage, LED operation in reverse saturation mode, Operating current now is referred to as reverse saturation current.Be positioned over LED on heat sink by the present invention, and temperature control is carried out by temperature control instrument, and Lead to a backward voltage at LED two ends, while test reverse saturation current now, then adjusts temperature control instrument and change heat sink temperature Degree (as junction temperature), can obtain the relation of reverse saturation current and junction temperature.Choose LED operation backward voltage be -5V when, this When, the electric current by LED is minimum (about~nA orders of magnitude), it is believed that LED does not produce Joule heat, therefore in this case The junction temperature of LED is equivalent to heat sink temperature.By the temperature change that temperature control instrument controlling is heat sink, corresponding reverse saturation current Also can change.Corresponding saturation current under different LED junction temperature (heat sink temperature) is recorded by current measurement instrument, then it is right The data for being gathered carry out nonlinear fitting, it can be deduced that I_sAnd T_jDependency relation.It is designated as：

I_s=I_s(T_j)

When LED operation is under conditions of high current when, the heat that LED is produced because of the power supply of external current source can not be neglected Omit, therefore the junction temperature of LED is just not equal to heat sink temperature；Now, the junction temperature of LED, electric current and both end voltage still meet Shockley equations, are write as：

In the case of the forward current identical for being maintained to flow through LED, change heat sink temperature (from T_sink1=25 DEG C change to T_sink2=35 DEG C), LED current I, voltage V and junction temperature T under different heat sink temperatures_jIt is satisfied by Shockley equations.

It is T in heat sink temperature_sink1Under, there is equation：

It is T in heat sink temperature_sink2Under, there is equation：

Keeping the constant I of LED two ends electric currents₁=I₂, in the case of only changing the heavy temperature of LED heat, forward voltage variable quantity is very Small (about a few mV), so in T_sink1And T_sink2LED heat power in the case of temperature control is equal, and its chip of same LEDs to knot The thermal resistance of temperature is also equal, so T_j1-T_sink1=T_j2-T_sink2, therefore there is the junction temperature temperature difference to be equal to the heat sink temperature difference：

T_j2-T_j1=T_sink2-T_sink1 (3)

For three above equation, wherein q and k is constant, and electric current I and voltage V can be by experiment measurement and obtain Known quantity, n, T_j1And T_j2It is three unknown quantitys, Simultaneous Equations (1), (2), (3) just can calculate LED in T_sink1Under Junction temperature T_j1With LED in T_sink2Under junction temperature T_j2.So, according to this LED junction temperature measurement side based on Shockley equations Method, just can respectively measure the junction temperature of different heat sink lower LED.

The beneficial effects of the invention are as follows：The present invention only need to simple voltage source, current source test reverse saturation current and Forward I-V characteristic curves are that can obtain LED junction temperature, improve LED junction temperature test process, it is to avoid forward voltage method is from heating High current is switched to the process of test low current, and need not require that instrument can realize being switched fast (usually 1 μ s) and in time The highly difficult test of gathered data, greatly simplifies the junction temperature measurement method of LED.

Brief description of the drawings

Fig. 1 is present invention measurement saturation current I_sWith LED junction temperature T_jThe circuit theory diagrams of curve.

Fig. 2 is one group of I of present invention measurement_s-T_jCurve map.

Fig. 3 is the circuit theory diagrams of present invention measurement LED C-V characteristics.

Fig. 4 is the VA characteristic curve figure of present invention measurement.

In figures 1 and 3, each mark for：

1- current sources, voltage source (the SYSTEM SOURCE of KEITHLEY 2611)；

2- test samples LED；

3- is heat sink；

4- temperature controls instrument (the TEC SOURCE of KEITHLEY 2510)；

5- computers；

The 6- senses of current.

Specific embodiment

Implementation process of the invention is discussed in detail with reference to specific embodiment and accompanying drawing：

1st, measurement saturation current I_s, obtain I_sWith non-lit up LED junction temperature T_jRelation, and fit I_s-T_jCurve map.

Experimental equipment according to Fig. 1 connects the experimental provision of measurement LED reverse currents.Using current source 1 to tested The LED samples 2 of examination provide backward voltage (usually -5V), test reverse current I_s.Using temperature control instrument 4 by heat sink 3 temperature Control respectively at least four different temperature spots, in experiment select temperature spot be 25 DEG C, 35 DEG C, 45 DEG C, 55 DEG C, 65 DEG C, 75 DEG C, at least 2min will be waited after often changing heat sink temperature so that re-test reverse current when LED samples reach thermal balance I_s, now LED is not luminous does not generate heat, and heat sink temperature is non-lit up LED junction temperature.It is soft by computer origin data processings Part is by measured saturation current I_sWith non-lit up LED junction temperature T_jNonlinear fitting is carried out, I is obtained_s-T_jCurve, such as Fig. 2 institutes Show.Fit equation is：

Wherein, I_s0(T_j0) represent T_j0Reverse current at=25 DEG C, coefficient of determination R²=0.99, show that degree of fitting is approached In 1.It is therefore contemplated that it is very rational to fit the equation for coming.Table 1 provides 6 fitting coefficients of sample：

Table 1

LED samples
					Blue-ray LED 1 (B1)		36.40	1.80
Blue-ray LED 2 (B2)	-5.51×10	95.07	3.37
					Blue-ray LED 3 (B3)		61.71	5.10
Red-light LED 1 (R1)		-59.81	0.83
					Red-light LED 2 (R2)		-57.76	0.86
Red-light LED 3 (R3)		-58.60	0.69

2nd, LED current voltage response when lighting situation is measured.

Experimental equipment according to Fig. 3 connects the experimental provision of measurement LED sample I-V characteristic curves.Use temperature controller Device 4 controls heat sink 3 temperature, general to choose and room temperature temperature relatively, 25 DEG C of selection in experiment.Now pass through current source 1 provides forward current I to LED samples 2, waits at least 3min when LED reaches thermal equilbrium state, is tested using four-wire measurement method Forward voltage V.Change the experimental data that electric current I (generally selecting 1~500mA intervals 1mA) test obtains a series of I and V, from And the VA characteristic curve I-V (heat sink temperature is maintained at 25 DEG C) of forward current I and forward voltage V is obtained, as shown in Figure 4.Connect Get off the temperature for changing heat sink 3, the temperature now chosen is most appropriate at 10 DEG C or so with temperature difference before, selected in experiment Take 35 DEG C, repeat same experimentation, obtain another group of I-V characteristic curve (heat sink temperature is maintained at 35 DEG C).

3rd, with reference to the i-v curve and I of LED_s-T_jThe fit equation of curve, is calculated by MATLAB programming softwares LED junction temperature.

It is T in heat sink temperature_sink1Under, there is equation：

It is T in heat sink temperature_sink2Under, there is equation：

And there is the junction temperature temperature difference to be equal to the heat sink temperature difference：T_j2-T_j1=T_sink2-T_sink1 (3)

With reference to：

T_j2-T_j1=35-25=10

Calculating is programmed by MATLAB softwares, just can be in the hope of the LED junction temperature T at 25 DEG C_j1。

The data of 4 and Raman methods measurement LED junction temperature are contrasted, and as shown in table 2, checking is based on Shockley equations Measure the accuracy of the new method of LED junction temperature.

Table 2

Note：B1：Blue-ray LED 1, R1：Red-light LED 1.

The data of LED junction temperature are measured by contrasting Raman methods, " based on Shockley equations measurement LED junction temperature The LED junction temperature error that new method " is measured is less than 1%, and this small difference belongs within experimental error allowed band, assistant The feasibility and accuracy of this measuring method are demonstrate,proved.

Claims (1)

1. the method for I-V characteristic curved measurement LED junction temperature is based on, it is characterised in that comprised the following steps：

1) measurement saturation current I_sWith non-lit up LED junction temperature T_j, and it is fitted I_s-T_jCurve map；The measurement saturation current I_sWith Non-lit up LED junction temperature T_j, and it is fitted I_s-T_jThe specific method of curve map is：By LED samples be placed on temperature control it is heat sink on, LED Two ends connect with current source two ends, and temperature control is heat sink to connect with temperature control instrument；Heat sink temperature is set with temperature control instrument, when LED reaches During to thermal balance, current source output backward voltage is set, now LED does not light and do not generate heat, and LED junction temperature is heat sink temperature, The saturation current I at test LED two ends_s；Regulation temperature control instrument changes heat sink temperature, repeats above-mentioned test process, and obtaining one is Arrange the saturation current I under different junction temperatures_s；The saturation current I obtained with the fitting test of computer origin data processing softwares_s With LED junction temperature T_j, obtain I_s-T_jCurve, fit equation is：

$I_s=I_{s0}\left(T_{j0}\right)\·{\mathrm{\β}}_1\·\exp \left(\frac{-T_j}{{\mathrm{\β}}_2}\right)+{\mathrm{\β}}_3$

Wherein, I_sIt is saturation current, I_s0(T_j0) represent T_j0Saturation current at=25 DEG C, T_jAs LED junction temperature, β₁、β₂、β₃For Fitting coefficient；

2) VA characteristic curve of LED is measured；The specific method of the VA characteristic curve of the measurement LED is：LED samples are put Put on temperature control is heat sink, LED two ends connect with current source two ends, temperature control is heat sink to connect with temperature control instrument；Set with temperature control instrument Heat sink temperature, makes heat sink temperature keep constant；Now, forward current I is provided to LED samples by current source, using four lines Measuring method test forward voltage V；The forward voltage V tested in experiment under multiple forward currents obtains forward current I and forward direction The VA characteristic curve I-V of voltage V；Then change heat sink temperature, repeat same experimentation, obtain another group of I-V characteristic Curve；

3) LED junction temperature is calculated using MATLAB programming softwares, the utilization MATLAB programming softwares are calculated LED junction temperature Specific method be：It is T in heat sink temperature_sink1Under, there is equation below：

$I_1=I_{s1}\left(T_{j1}\right)\·\exp \left(\frac{{\mathrm{qV}}_1}{{\mathrm{nkT}}_{j1}}\right)---\left(1\right)$

It is T in heat sink temperature_sink2Under, according to equation：

And there is the junction temperature temperature difference to be equal to the heat sink temperature difference：T_j2-T_j1=T_sink2-T_sink1 (3)

Wherein I₁Represent heat sink temperature T_sink1The forward current of LED, V at=25 DEG C₁Represent heat sink temperature T_sink1LED at=25 DEG C Forward voltage, I_s1(T_j1) it is on T_j1Function, according to step 1) in equation obtain, q is unit charge carried charge, and k is Boltzmann constant, n represents ideality factor, is unknown, T_j1As require in heat sink temperature T_sink1LED is lighted at=25 DEG C When junction temperature；Similarly, I₂Represent heat sink temperature T_sink2The forward current of LED, V at=25 DEG C₂Represent heat sink temperature T_sink2=25 The forward voltage of LED, I at DEG C_s2(T_j2) it is on T_j2Function, according to step 1) in equation obtain, T_j2As require Heat sink temperature T_sink2Junction temperature when LED is lighted at=35 DEG C；According to equation (1), (2) and (3) and by MATLAB program calculations, Try to achieve T_j1, T_j2And n.