Cumulated damage

[IGuillaumin]

Hello,

I'm trying to find the equation to define cumulated damage ( AF1 then AF2 for example).
I can calculate the acclerations factors for each stress but I don't know how to cumulate them.

My question is how can I cumulate these different accelerations factors (one is a temperature AF1 then the other will be thermal cycling AF2 for example)

[tarik]

For more info about the time varying stress model and cumulative damage review the following material (this analysis can be done in the ALTA software).

http://www.weibull.com/AccelTestWeb/model_formulation.htm
http://www.weibull.com/AccelTestWeb/time-varying_stress_models_chap_.htm

[Pantelis]

Are you asking about a cumulative damage model, or how to combine the two acceleration factors. If the later, what are you measuring for thermal cycling (I am assuming not temperature since that is the other stress).

[sandrine]

Hello Pantelis ,

Thank you for your answer.
I'm asking about a cumulative damage model .

I have a test in high temperature (eg :ageing of product) , then I have a test to model crack in solder joints (material fatigue)due to repeated temperature cycling .

I have two questions :

Is it possible to combine the two accelerations factors and to find one factor acceleration ? (what is applied on the product successively)

Is it possible to use time-varying stress model (in Alta 6 pro)when we have 2 different stresses (cumulative damage ) in order to model this test plan?

Please help
Thank you in advance

[Pantelis]

Sandrine:

Good question. To make sure we are both on the same page let me rephrase the question, in general terms, and as I understand it, along with any assumptions. Then I’ll try to answer it the best I can.

You stress a product under stress A for time TA time and then Stress B for time TB sequentially. Furthermore, and since you are trying to combine them, the assumption is then that both A and B accelerate the same failure mode and the product can fail if A alone is applied or B alone is applied (since you want to find an equivalency between the stresses). Also, and since the stresses are applied sequentially time spent at A shortens subsequent life at B, thus theoretically a CD (Cumulative Damage) model would seem well suited in modeling this problem.

Now when you try to actually apply the model using CD, the problem that will arise is how to numerically quantify (using a single unit of measurement) the stress levels at A and B. In other words the model requires a common stress measurement, which in this case you do not have, since stress level A and B are quantified differently (e.g. A may be temperature and B may be cycles). Even though that is a problem it does not mean it is impossible. One can theoretically come up with some relationships between the two. However, and before getting into it I want to make sure that you understand what the resulting acceleration factor would be. In this case the acceleration factor would be a function in terms of the value of stress A, B and the time exposed at stress A before going to B at this level (assuming that A is first) versus the use level of A, B and the time they would normally be exposed to A at use conditions. Assuming that is what you want then the resulting AF would require that one quantifies the acceleration factor relationship of A alone (e.g. A0 to A1) and B alone as well as the AF relationship between A and B. Thus the combined one would be function of all these three factors (not just AFA and AFB). In other words, to be able to apply CD you will need to understand the relationship between A and B – or the acceleration factor between A and B.

Hope this helps.