• State
  • Locked Locked
  • Replies 2 replies
  • Subscribers 46 subscribers
  • Views 1948 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Solder Joint Reliability acceleration Model

Yun FANG
Intellectual 310 points

Hi

In the application report SPRABY2.pdf

Norris-Landzberg Acceleration Factor Equation is used for the calculation. Could you please explain the meaning of each parameter? If possible, would you please give an example?

Thanks

  • 0
    TI__Guru 71526 points

    Hi Yun,

      Please disregard my previous response that CSC would contact you, I've instead gone direct to the APp Note author to ask for an explanation. 

    Regards,

    ~Leonard 

  • 0
    TI__Prodigy 460 points

    Hi Yun, thank you for your question. I am glad to assist. Let us walk through an example based on eutectic Sn/Pb solder, which is 63%Sn, 37%Pb. The model parameters for this structure have been well characterized for a long time.

    The first component of the Norris-Landzberg AF, (deltaT-stress / deltaT-use)^n is called the Coffin-Manson AF, with exponent "n" (Coffin-Manson exponent). In our example case of eutectic solder the exponent is 1.9. Let us suppose that the Temp Cycle accelerated stress condition is -55C to 125C and the application use temp cycle condition is 25C to 75C. Therefore the Coffin-Manson AF is (180/50)^1.9 = 11.4 

    The next AF component is the frequency factor, accounting for differences in the frequency of temp cycling between the acceleration stress test environment and the application usage. This accounts for the variation is strain rate. For eutectic solder the value of the exponent (usually denoted as "m") is about 1/3 or 0.33. A typical frequency range for accelerated temp cycle stress is 30-60 minutes. Let us use the more common 30 minutes in our example. Further, let's suppose an application is assumed to have 1 cycle per day. So our AF in this case is (48/1)^0.33 or 3.6.

    Lastly, we have the Arrhenius AF accounting for differences in the maximum temperature of the accelerated stress test compared to the application environment. For eutectic solder, Ea/k (activation energy/Boltzmann's constant) is equal to 1414 deg.K. In our example, Tmax-stress = 125C (398K) and Tmax-use is 75C (348). Therefore the Arrhenius AF is EXP(1414*((1/348)-(1/398))) =  1.67.

    The final Norris-Landzberg AF in our example is 11.4 * 3.6 * 1.67 =  68.5.

    For SAC (SnAgCu, Pb-free) solder, Pan et al. have published model parameters of n=2.65, m=0.136 and Ea/k = 2185. Refer to Proc. SMTA, 2005, pp. 876-883

    Finally, note that many authors invert the "stress" and "application use" delta T ratio and frequency ratio (with application or field in the numerator and stress test in the denominator). This means the ratios have exponents of -n and -m respectively. This is merely a matter of preference.