CN103913300A - LED lighting product accelerated attenuation test method based on multilevel stepping stress - Google Patents

Abstract

The invention discloses an LED lighting product accelerated attenuation test method based on multilevel stepping stress. The method includes the following steps that first, the value of confidence and the number of samples are set; second, constant humidity stress, multiple stepping temperature stress levels and the total time of a test are set, and the step length time of all the temperature stress and measurement time nodes are set; third, the initial values of the luminous flux of target samples are measured, and luminous flux attenuation trajectory data of the target samples are obtained; fourth, an attenuation mechanism index test is performed; fifth, the pseudo failure life of the target samples is calculated; sixth, a distribution test is performed on the pseudo failure life, and a distribution function is selected; seventh, the distribution parameter of the distribution function is solved, and in combination with an Arrhenius temperature acceleration model, the reliability distribution function, the average life and the median life of the target samples are obtained. Through the method, the flexibility and the universality of a stepping stress test method applied to LED lighting products can be improved, the acceleration time is further shortened, and the accuracy of service evaluation of the accelerated test is improved.

Description

LED illuminating product based on multilevel stepstress accelerates die-away test method

Technical field

The present invention relates to the acceleration die-away test technology of LED illuminating product, specifically a kind of LED illuminating product based on multilevel stepstress accelerates the method for die-away test.

Background technology

LED (Light Emitting Diode) because of its life-span long, pollution-free, the advantage such as energy-efficient and enjoy favor.In general illumination field, its widespread use will be another revolution after incandescent lamp.But present stage lacks practicable LED lighting reliability acceleration service life test method and standard, this becomes restriction LED and enters the bottleneck in general illumination field.

The star of american energy proposes test specimen to carry out the cold test of at least 6000 hours, and such requirement is difficult to carry out the enterprise of Efficient Development for trying hard to.For addressing this problem, the method that has proposed in the invention that patent publication No. is CN102890249A to use subsystem to decompose is carried out reliability assessment.The method that the subsystem of the LED lighting using based on this patent decomposes, patent publication No. is in CN103292982A invention, to propose to use 3 to enter temperature stress, the constant acceleration die-away test method at the LED of 85%RH light fixture of humidity step by step.

But, on the one hand, while LED product being carried out to accelerated test under too low stress level, portioned product, containing LED packaging and LED lighting product, can be in active period, can within a period of time, there is floating high phenomenon in optical parameter, be that the logical output of light is greater than initial value, thus test figure be not enough to the to extrapolate situation of life of product information after causing the accelerated test of three step stepstress to complete; In addition on the one hand, because the stress weatherability of different product is different, after the accelerated test that also may occur 3 step stepstress completes, product luminous flux sustainment rate does not decay to inefficacy threshold values L70, be that luminous flux attenuation is 70% of initial value, and then causing the prediction accuracy of life of product too low, degree of accuracy also cannot further be enhanced simultaneously.

Summary of the invention

The object of the invention is to provide for the deficiencies in the prior art a kind of LED illuminating product based on multilevel stepstress to accelerate the method for die-away test.This method can be improved flexibility and the versatility that step stress test method is applied on LED illuminating product, further shortens and accelerates duration and improve the accelerated test precision of prediction in assessment life-span.

The technical scheme that realizes the object of the invention is:

LED illuminating product based on multilevel stepstress accelerates die-away test method, comprises following steps:

1) set degree of confidence numerical value and sample size, and prepare target sample;

2) the highest nominal environment for use temperature based on product, sets constant humidity stress and multiple stepping temperature stress level and total time on test, sets step-length time and the Measuring Time node of each temperature stress;

3) the luminous flux initial value of measurement target sample, and under the environment of constant humidity stress and multilevel stepping temperature stress level, obtain the luminous flux attenuation track data of target sample;

4) utilize the performance degradation data under the each temperature stress of complex indexes decay locus model matching, carry out the kill mechanism index test of each stress level;

5) calculate the pseudo-burn-out life of target sample different temperatures stress level under constant high humility stress >=80%RH, and this puppet burn-out life is converted to the pseudo-burn-out life for target sample different temperatures stress level under the humidity stress of practical service environment;

6) the pseudo-burn-out life of different temperatures stress level under the constant high humility stress >=80%RH of target sample is carried out to normal distribution, Weibull distribution and lognormal distribution inspection, and select distribution function;

7) obtain the distribution parameter of distribution function, and in conjunction with Arrhenius (Arrhenius) temperature acceleration model, obtain fiduciary level distribution function, mean lifetime and the median life of target sample target sample under practical service environment stress level.

Step 2) in, three step basic tests of a constant high humility stress level and three temperature stresses are set, the test duration of allowing according to maximum, arranges a stress level for subsequent use as alternative test simultaneously.

In step 4), utilize the luminous flux attenuation track of complex indexes decay locus model matching target sample, in data fitting process, by checking the kill mechanism index parameter α magnitude difference under each stress level to be less than setting value, the approximate kill mechanism of judging target sample does not change.

In process of the test, temperature stress level order from low to high loads continuously, and each temperature stress level is used same humidity stress simultaneously.

The described LED illuminating product based on multilevel stepstress accelerates die-away test method based on following 5 hypothesis:

Suppose one: the performance degradation of target sample experience is irreversible, and performance degradation process has monotonicity;

Suppose two: failure mechanism and the failure mode of accelerating target sample under stress level at each all remain unchanged;

Suppose three: the acceleration attenuation data of target sample has identical distribution form under different stress levels, the pseudo-burn-out life of target sample that utility attenuation data obtains simultaneously should be obeyed same distribution pattern under different stress levels;

Suppose four: the residual life of target sample and the mode of accumulation are irrelevant, only depend on the stress level of loading and cumulative failure part;

Suppose five: the performance degradation process of target sample can be described with the complex indexes locus model of decaying.

The application of the method for the described LED illuminating product acceleration die-away test based on multilevel step-down-stress in the light source subsystem of LED packaging or LED lighting product.

The advantage of this method is: applied multiple temperature stress level orders from low to high and loaded continuously, in the scope of allowing at test acceleration duration, flexibility and versatility that step stress test method is applied on LED illuminating product are improved, can effectively shorten the accelerated test assessment cycle of LED illuminating product, further shorten acceleration duration and improved accelerated test assessment life prediction precision, reduced experimentation cost.So make to accelerate die-away test method and can have more specific aim, the acceleration die-away test method that is suitable for LED illuminating product for further exploring has very important significance.

Brief description of the drawings

Fig. 1 is that in embodiment, the LED illuminating product based on multilevel stepstress accelerates die-away test method flow block diagram;

Fig. 2 is the relation curve schematic diagram of sample size and degree of confidence in embodiment;

Fig. 3 is that in embodiment, step rises the lumen depreciation variation track schematic diagram that accelerates die-away test;

Fig. 4 is the theoretical die-away curve schematic diagram of converting in embodiment after single target specimen test data;

Fig. 5 is the matching residual error schematic diagram of converting in embodiment after all target sample test figures;

Fig. 6 is the kill mechanism index alpha parameter schematic diagram of each stress level in embodiment;

Fig. 7 is the fiduciary level curve synoptic diagram under the environmental stress that in embodiment, target sample LED spotlight light source subsystem uses in reality;

Fig. 8 is LED projection lamp light fitting system dependability curve synoptic diagram in embodiment;

Fig. 9 is the activation energy Ea matching schematic diagram of product in embodiment.

Embodiment

Below in conjunction with drawings and Examples, content of the present invention is further elaborated, but is not limitation of the invention.

Embodiment:

With reference to Fig. 1-Fig. 9: the LED illuminating product based on multilevel stepstress accelerates the method for die-away test, comprises the steps:

1) set degree of confidence P _anumerical value and sample size n, and prepare target sample;

2) the highest nominal environment for use temperature based on product, sets constant humidity stress and multiple stepping temperature stress level and total time on test, sets step-length time and the Measuring Time node of each temperature stress;

3) the luminous flux initial value of measurement target sample, and under the environment of constant humidity stress and multilevel stepping temperature stress level, obtain the luminous flux attenuation track data of target sample;

4) utilize the performance degradation data under the each temperature stress of complex indexes decay locus model matching, carry out the kill mechanism index test of each stress level;

5) calculate the pseudo-burn-out life of target sample different temperatures stress level under constant high humility stress >=80%RH, and this puppet burn-out life is converted to the pseudo-burn-out life for target sample different temperatures stress level under the humidity stress of practical service environment;

6) the pseudo-burn-out life of different temperatures stress level under the constant high humility stress >=80%RH of target sample is carried out to normal distribution, Weibull distribution and lognormal distribution inspection, and select distribution function;

7) obtain the distribution parameter of distribution function, and in conjunction with Arrhenius temperature acceleration model, obtain fiduciary level distribution function, mean lifetime and the median life of target sample under practical service environment stress level.

Step 2) in, three step basic tests of a constant high humility stress level and three temperature stresses are set, the test duration of allowing according to maximum, arranges a stress level for subsequent use as alternative test simultaneously.

In step 4), utilize the luminous flux attenuation track of complex indexes decay locus model matching target sample, in data fitting process, by checking the kill mechanism index parameter α magnitude difference under each stress level to be less than setting value, the approximate kill mechanism of judging target sample does not change.

In process of the test, temperature stress level order from low to high loads continuously, and each temperature stress level is used same humidity stress simultaneously.

The described LED illuminating product based on multilevel stepstress accelerates die-away test method based on following 5 hypothesis:

Suppose one: the performance degradation of target sample experience is irreversible, and performance degradation process has monotonicity;

Suppose two: failure mechanism and the failure mode of accelerating target sample under stress level at each all remain unchanged;

Suppose three: the acceleration attenuation data of target sample has identical distribution form under different stress levels, the pseudo-burn-out life of target sample that utility attenuation data obtains simultaneously should be obeyed same distribution pattern under different stress levels;

Suppose four: the residual life of target sample and the mode of accumulation are irrelevant, only depend on the stress level of loading and cumulative failure part;

Suppose five: the performance degradation process of target sample can be described with the complex indexes locus model of decaying.

The application of the method for the described LED illuminating product acceleration die-away test based on multilevel step-down-stress in the light source subsystem of LED packaging or LED lighting product.

Particularly:

In step 1), in the time that LED life of product is obeyed Weibull distribution, for accelerating die-away test, at definite fiduciary level R (t), can draw degree of confidence P _awith the pass of sample size n be:

$n=\frac{\ln (1-P_a)}{\ln R\left(t\right)}---\left(1\right)$

The present embodiment scope of application is LED packaging and LED lighting product, if target sample is LED lighting product, the method that the subsystem proposing in need to the patented claim that be CN102890249A according to publication number decomposes is processed target sample, then using light source subsystem as target sample, in addition, the patented claim method that need be CN103292982A with reference to publication number of building of accelerated test platform is built.

Utilize formula (1), getting fiduciary level R (t) is 0.8, as Fig. 2 obtains degree of confidence P _arelation with sample size n.Choose degree of confidence P _abe 0.83, the number of sample size n is 8.Choose the LED shot-light of 12W, the high ambient temperature of its nominal is T _b=55 DEG C, the method that the subsystem proposing in the patented claim that is CN102890249A according to publication number decomposes, decomposing LED shot-light is the subsystem of three functional independences, comprises LED light source subsystem, driver sub-system and interface clamping device subsystem, and using LED light source subsystem as target sample.

Step 2) in, according to the minimum needs of life of product extrapolation process, three Buwen's degree stress levels are set as the basic test stage, in three step tests, ensure that target sample has identical theoretical damping capacity under each stress level.With reference to the given the highest nominal serviceability temperature of the product T of producer _b, by formula (2), the horizontal S of minimum temperature is set ₁.The horizontal S of temperature stress ₂and S ₃arrange respectively according to formula (3), (4).

S ₁=T _b+ Δ T ₁(Δ T ₁get 0 DEG C～5 DEG C) (2)

S ₂=S ₁+ Δ T ₂(Δ T ₂get 10 DEG C～20 DEG C) (3)

S ₃=S ₂+ Δ T ₃(Δ T ₃get 10 DEG C～15 DEG C) (4)

Because humidity is a key factor that affects LED product, accelerate to add humidity stress in die-away test process and will further shorten test period.When LED lamp light source subsystem is accelerated to die-away test, the constant humidity stress of employing is 85%RH ± 2.The time point that each stress starts and finishes is measured respectively target sample performance data one time; Meanwhile, for ensuring enough extrapolation experimental datas, at least reset two target sample performance data Measuring Time points in the process of stress loading, each stress level has four groups of measurement data at least.

For the amount of degradation that can be low stress level by the amount of degradation conversion of high stress level, the amount of degradation respectively accelerating under stress level need to be set as identical value.Thus, according to Arrhenius relationship speedup factor AF, have:

AF _T1×t ₁=AF _T2×t ₂=AF _T3×t ₃

The i.e. step-length time t of three steps ₁, t ₂, t ₃pass be:

$\exp \left(\frac{E_a}{k}(\frac{1}{T_u}-\frac{1}{T_1})\right)\×t_1=\exp \left(\frac{E_a}{k}(\frac{1}{T_u}-\frac{1}{T_2})\right)\×t_2=\exp \left(\frac{E_a}{k}(\frac{1}{T_u}-\frac{1}{T_3})\right)\×t_3$

In formula, T _u=298.15K, T ₁, T ₂, T ₃be respectively S ₁, S ₂, S ₃the kelvin rating (K) converting to, Ea is activation energy, k is Boltzmann constant (Boltzmann constant), and then draws three load time ratios that accelerate stress.

After completing the setting of three step basic tests, allow accumulation test period t according to the maximum that can carry out continuously _a, the 4th stress level is set as stress level S for subsequent use ₄, for ensureing continuity and the maximum acceleration effect of extrapolated data, stress S for subsequent use ₄arrange according to formula (5).

S ₄=S ₃+ Δ T ₄(Δ T ₄get 0 DEG C～15 DEG C) (5)

Stress S for subsequent use ₄accumulation load duration be as the criterion can obtain plural luminous flux measurement data point, i.e. the acceleration time t of stress level for subsequent use ₄>=t ₃/ 2, ensure the validity of stress accelerated test for subsequent use; According to 55 DEG C of the highest nominal serviceability temperatures of product, choose S1=65 DEG C ± 2 DEG C of three step basic test stress levels, S2=85 DEG C ± 2 DEG C, S3=95 DEG C ± 2 DEG C by formula (2)-(5), stress level S4=S3 for subsequent use, select constant humidity stress level is 85%RH ± 2 simultaneously.The present embodiment LED shot-light maximum allows that accumulation test period is 1500 hours, and the total testing time of setting three step basic tests is about 1258 hours.Can obtain t1 according to formula (4), t2, t3 is respectively: 538 hours, 384 hours, 336 hours.In the process of the test of each stress level, except starting and finishing the DATA REASONING of each a group, it is that each stress level has four groups of measurement data points at least that intermediate means is set two groups of DATA REASONING time points.Meanwhile, determine that the accumulation test duration of stress level S4 for subsequent use is 240 hours, DATA REASONING frequency is identical with stress level S3.

In step 3), testing evaluation object is LED lighting product, the method that the subsystem proposing in should the patented claim that be CN102890249A according to publication number decomposes is processed test assessment object, then using light source subsystem as target sample, in addition, the patented claim method that need be CN103292982A with reference to publication number of building of accelerated test platform is built, the initial luminous flux value of measurement products, and the test constant humidity that uses the present embodiment to arrange, step cooling degree stress level, the acceleration accumulation load time of each stress level, the test parameterss such as parameter measurement time point obtain the luminous flux attenuation track data of target sample, wherein, the temperature stress section of three step basic tests loads by the stress sequencing of (S2) low (S3) in high (S1), and under each stress level, use a constant humidity stress (85%RH ± 2).While executing three step basic tests, can carry out the follow-up life of product extrapolation process of step 4).Meanwhile, observe luminous flux sustainment rate and whether remain on more than 70%, if not, can abandon using stress level for subsequent use; If so, select to have continued stress accelerated test for subsequent use, and again carry out life-span extrapolation, obtain compared with three more accurate life-spans of step basic test.The present embodiment, complete three while entering basic test step by step, as shown in Figure 3, under 85%RH humidity environment, target sample 1-8 is the lumen depreciation decay track under 65 DEG C, 85 DEG C, 95 DEG C three step intensification degree stress levels respectively for the lumen depreciation track of target test.

Because three when entering basic test and completing step by step, the mean lumens sustainment rate of target sample is still greater than the inefficacy threshold values L70 of product, in carrying out life-span extrapolation, can select to enable in advance stress level S4 for subsequent use, until complete whole accelerated test plan.But, consider that on the one hand three enter the test parameters decay stability-of-path of basic test, the life of product information of enough having extrapolated step by step; In addition on the one hand, consider will be very limited at the test duration intrinsic parameter attenuation degree of stress for subsequent use, and consider experimentation cost problem, the implementation case is selected the accelerated test of abandoning enabling stress level for subsequent use.Enter basic test based on the implementation case three as follows step by step, carry out step 4).

In step 4), utilize the LED light fixture performance degradation data under the each temperature stress of complex indexes decay locus model matching, carry out the kill mechanism index test of the each stress level of target sample; Concrete steps are described as follows:

According to test hypothesis, the lumen depreciation decay track of target sample to be measured need to meet complex indexes decay requirement, just can carry out life prediction.But need to judge that target sample is in the time of each stress level accelerated test, whether kill mechanism changes, thus the validity of guarantee accelerated test.In data conversion analytic process, can judge whether degradation mechanism exceedes the maximum threshold values of setting by observing parameter alpha, the selected parameter alpha variable quantity between all stress levels of the present embodiment is no more than 0.2.

Accelerated test stress level loads with time relationship and can be expressed as follows:

$S=\left\{\begin{array}{cc}{S}_1& (t_1\≤t\≤t_2)\\ S_2& (t_2\≤t\≤t_3)\\ ...& ...\\ S_m& (t_m\≤t\≤t_{m+1})\end{array}\right.---\left(6\right)$

Make H ₀(t|S _j) Degradation path that arrives for test observation, L ₀(t|S _j) be track (j=1,2,3 of prediction ... m).The terminal of previous stress level is the starting point of a rear stress level in test, can make the lumen depreciation decay track under a certain stress level be:

L(t)=L(t-t _j+ω _j|S _j) （7）

Wherein: t ₁=ω ₁=0.

To two stress levels arbitrarily, generalized case has:

L(ω _i|S _i)=L(ω _i-1+t _i-t _i-1|S _i-1) （8）

Under each stress level, expression formula is as follows:

$L_0\left(t\right)=\left\{\begin{array}{cc}L(t+{\mathrm{\ω}}_1-t_1|S_1)& (t_1\≤t\≤t_2)\\ L(t+{\mathrm{\ω}}_2-t_2|S_2)& (t_2\≤t\≤t_3)\\ ...& ...\\ L(t+{\mathrm{\ω}}_m-t_m|S_m)& (t_m\≤t\≤t_{m+1})\end{array}\right.---\left(9\right)$

Wherein: t ₁=ω ₁=0, i=1,2,3 ... m

The Degradation path of the lumen depreciation under each stress meets:

L(t|S _i)=exp(-β _i·t ^α)（i=1,2,3…m） （10）

Can be obtained by formula (8) (10):

${\mathrm{\β}}_i={\mathrm{\β}}_{i-1}\·{\left(\frac{t_i-t_{i-1}+{\mathrm{\ω}}_{i-1}}{{\mathrm{\ω}}_i}\right)}^{\mathrm{\α}}---\left(11\right)$

By (10) (11) substitutions (9) and twice logarithm got in two ends can obtain simultaneously:

$\ln (-\ln \left(L_0\left(t\right)\right))=\left\{\begin{array}{cc}\ln {\mathrm{\β}}_1+\mathrm{\α}\·\left(\ln t\right)& (t_1\≤t\≤t_2)\\ \ln {\mathrm{\β}}_1+\mathrm{\α}\·\{\ln \left(\frac{t_2}{{\mathrm{\ω}}_2}\right)+\ln (t-t_2+{\mathrm{\ω}}_2)\}& (t_2\≤t\≤t_3)\\ ...& ...\\ \ln {\mathrm{\β}}_{m-1}+\mathrm{\α}\·\{\ln \left(\frac{t_m-t_{m-1}+{\mathrm{\ω}}_{m-1}}{{\mathrm{\ω}}_m}\right)+\ln (t-t_m+{\mathrm{\ω}}_m)\}& (t_m\≤t\≤t_{m+1})\end{array}\right.---\left(12\right)$

Can be obtained at stress S by formula (12) _iunder have: t ∈ [t _i, t _i+1) time

ln(-ln(L ₀(t)))=α·ln(t-t _i+ω _i)+A _i （13）

Wherein, $A_i=\ln {\mathrm{\β}}_{i-1}+\mathrm{\α}\·\{\ln \left(\frac{t_i-t_{i-1}+{\mathrm{\ω}}_{i-1}}{{\mathrm{\ω}}_i}\right)$

As t ∈ [t _i, t _i+1) time, test measurements is data fitting residual error is:

${{\mathrm{SSE}}_i}^2=\underset{t=t_i}{\overset{t_{i+1}}{\mathrm{\Σ}}}{\{\ln (-\ln \left(H_0\left(t\right)\right))-\mathrm{\α}\left(i\right)\·[\ln (t-t_i+{\mathrm{\ω}}_i)]-A_i\}}^2---\left(14\right)$

Pass through $\left\{\begin{array}{c}\frac{\∂\left({{\mathrm{SSE}}_i}^2\right)}{\∂{\mathrm{\α}}_i}=0\\ \frac{\∂\left({{\mathrm{SSE}}_i}^2\right)}{\∂{\mathrm{\ω}}_i}=0\end{array}\right.$ Combinatorial formula (11) is tried to achieve SSE _i ²β when minimalization _i, α _i, ω _i(i=1,2 ... m), value

If make can further judge that lumen depreciation decay track need to meet complex indexes attenuation law.Meanwhile, to i ≠ j(1≤i arbitrarily, j≤m) have and meet δ≤0.2 o'clock, can be similar to judge target sample in attenuation process between different stress levels kill mechanism do not change.If judge that change has occurred the kill mechanism of target sample, proves that this accelerated test result is unavailable.Utilize the LED light fixture Performance Degradation Data under the each temperature stress of complex indexes decay locus model matching, if Fig. 4 is the theoretical die-away curve schematic diagram after single target sample experimental data is converted, same translation method, converts and matching the data of all target sample.Fig. 5 has provided the matching residual error of all target sample under each stress level, wherein uses residual error SSE threshold values to get 0.05, and the SSE of target sample 1-8 all, within the scope of threshold values, shows that complex indexes attenuation model is applicable to the lumen attenuation data of the present embodiment thus.In data fitting process, as shown in Figure 6, under three temperature stress levels, kill mechanism criterion parameter alpha is respectively 0.552,0.544 and 0.519, kill mechanism index parameter α magnitude difference under each stress level is less than 0.2, approximate thinks that the kill mechanism of target sample does not change under each acceleration stress.

In step 5), the fitting parameter calculating according to process d try to achieve each target sample at constant high humility stress (H by formula (10) _s) the pseudo-burn-out life LF of lower different temperatures stress level _ik(wherein k=1,2 ..., k is sample number).Utilize formula (15) model, the pseudo-burn-out life LF of conversion _ikfor target sample is at the humidity (H of practical service environment _u) the pseudo-burn-out life LF' of lower different temperatures stress level _ik;

LF' _ik=LF _ik·AF _RH （15）

Wherein ${\mathrm{AF}}_{\mathrm{RH}}={\left(\frac{H_s}{H_u}\right)}^{\mathrm{\ϵ}}---\left(16\right)$

ε is that fitting parameter the present embodiment of relative humidity is got ε=0.8.Theoretical die-away curve by each target sample under each stress level, obtains target sample at high humility stress (H _s) the pseudo-burn-out life LF of lower different temperatures stress level _ik, and calculated the humidity (H of practical service environment by formula (14)-(15) _u) the pseudo-burn-out life LF' of lower each temperature stress level _ik, wherein Hs is taken as 85%RH, and Hu is taken as 65%RH, and β gets 0.8.

In step 6), under the prerequisite that setting fiducial interval is 95%, in conjunction with Arrhenius acceleration model, utilize normal distribution, Weibull distribution and lognormal distribution to the pseudo-burn-out life (LF' of the target sample under each stress level _ik) carry out distributional assumption inspection, the lognormal distribution function that selection check goodness P value is greater than 0.9, and the distribution parameter of Computation distribution function.

In step 7), in conjunction with Arrhenius temperature acceleration model, simulate the pseudo-burn-out life (LF' of target sample _ik) distribution parameter under each stress level.And then obtain the reliability characteristic data such as Reliability Function, failure probability, mean lifetime and median life.Because the target sample in the implementation case is the light source subsystem of LED lamp system, the light source subsystem in the patented claim that is CN102890249A according to publication number, to the reliability synthetic method of whole lamp system, obtains the reliability characteristic data of whole lamp.In the time calculating the fiduciary level of lamp system, supposed the failure probability function of driving and mechanical part subsystem, and the comprehensive MTTF of driving and mechanical part subsystem is about 35000 hours.Can be obtained by Fig. 7 and Fig. 8, the MTTF of target light source subsystem and lamp system is respectively: 45089 hours, and 24684 hours.Meanwhile, as shown in Figure 9, the activation energy of test specimen is 0.336eV, and the speedup factor of product under three temperature stresses is respectively 5.82,11.08,14.89.

The described LED illuminating product based on multilevel stepstress accelerates die-away test method based on following 5 hypothesis:

Suppose one: the performance degradation of target sample experience is irreversible, and performance degradation process has monotonicity;

Suppose two: failure mechanism and the failure mode of accelerating target sample under stress level at each all remain unchanged;

Suppose three: the acceleration attenuation data of target sample has identical distribution form under different stress levels, the pseudo-burn-out life of target sample that utility attenuation data obtains simultaneously should be obeyed same distribution pattern under different stress levels;

Suppose four: the residual life of target sample and the mode of accumulation are irrelevant, only depend on the stress level of loading and cumulative failure part;

Suppose five: the performance degradation process of target sample can be described with the complex indexes locus model of decaying.

Claims (6)

1. the LED illuminating product based on multilevel stepstress accelerates die-away test method, it is characterized in that, comprises following steps:

1) set degree of confidence numerical value and sample size, and prepare target sample;

2) the highest nominal environment for use temperature based on product, sets constant humidity stress and multiple stepping temperature stress level and total time on test, sets step-length time and the Measuring Time node of each temperature stress;

3) the luminous flux initial value of measurement target sample, and under the environment of constant humidity stress and multilevel stepping temperature stress level, obtain the luminous flux attenuation track data of target sample;

4) utilize the performance degradation data under the each temperature stress of complex indexes decay locus model matching, carry out the kill mechanism index test of each stress level;

5) calculate the pseudo-burn-out life of target sample different temperatures stress level under constant high humility stress >=80%RH, and this puppet burn-out life is converted to the pseudo-burn-out life for target sample different temperatures stress level under the humidity stress of practical service environment;

6) the pseudo-burn-out life of different temperatures stress level under the constant high humility stress >=80%RH of target sample is carried out to normal distribution, Weibull distribution and lognormal distribution inspection, and select distribution function;

7) obtain the distribution parameter of distribution function, and in conjunction with Arrhenius temperature acceleration model, obtain fiduciary level distribution function, mean lifetime and the median life of target sample under practical service environment stress level.

2. the LED illuminating product based on multilevel stepstress according to claim 1 accelerates die-away test method, it is characterized in that, step 2) in, three step basic tests of a constant high humility stress level and three temperature stresses are set, the test duration of allowing according to maximum, arranges a stress level for subsequent use as alternative test simultaneously.

3. the LED illuminating product based on multilevel stepstress according to claim 1 accelerates die-away test method, it is characterized in that, in step 4), utilize the luminous flux attenuation track of complex indexes decay locus model matching target sample, in data fitting process, by checking the kill mechanism index parameter α magnitude difference under each stress level to be less than setting value, the approximate kill mechanism of judging target sample does not change.

4. the LED illuminating product based on multilevel stepstress according to claim 1 accelerates die-away test method, it is characterized in that, in process of the test, temperature stress level order from low to high loads continuously, and each temperature stress level is used same humidity stress simultaneously.

5. LED illuminating product based on multilevel stepstress according to claim 1 accelerates die-away test method, it is characterized in that, the described LED illuminating product based on multilevel stepstress accelerates die-away test method based on following 5 hypothesis:

Suppose one: the performance degradation of target sample experience is irreversible, and performance degradation process has monotonicity;

Suppose two: failure mechanism and the failure mode of accelerating target sample under stress level at each all remain unchanged;

Suppose three: the acceleration attenuation data of target sample has identical distribution form under different stress levels, the pseudo-burn-out life of target sample that utility attenuation data obtains simultaneously should be obeyed same distribution pattern under different stress levels;

Suppose four: the residual life of target sample and the mode of accumulation are irrelevant, only depend on the stress level of loading and cumulative failure part;

Suppose five: the performance degradation process of target sample can be described with the complex indexes locus model of decaying.

6. the application of the method for the LED illuminating product acceleration die-away test based on multilevel step-down-stress described in claim 1-5 any one in the light source subsystem of LED packaging or LED lighting product.