The equations are the easy part. The hard part is determining the correct activation energy. Any recommendations for books/training on how determine the correct activation energy?
Hi Don, The only accurate ways to really determine activation energy is to hope the supplier of your material has this data, otherwise, you will need to run a temperature experiment yourself and note the point where a change or reaction occurred. There's general equations for different types of materials that you can find in textbooks and Google searches, but it is always best to get this data directly from the supplier or characterize the data yourself.
Great video. Thank you. I would like to know how to find a stress factor. The physics model like Arrhenius assumes mechanisms of failure known beforehand. But for some problems we do not know it.
Hi Najoua, Thank you for the compliment. It depends on the trend of your data to determine the distribution that provides the best curve fit. Does it have a constant failure rate? If yes, you might be able to use an exponential distribution. Typically with electronics and materials, you will use some variation of a Weibull distribution. Check out my video on Weibull Analysis to learn more: ua-cam.com/video/ustgf9D7d5Q/v-deo.html Cheers, Tom
hey please tell me source of finding the suitable value for the constants like activation energy, power to the humidity ratio, coffin mission const, and power to the vibration level
My recommendation is to get material properties for you supplier. Using boiler plate values from the internet can result in a inaccurate AF. For vibe, you need to know your mission profile for your vibration loads and accelerate to your desired time while staying within the max recommended threshold for performing the time compression.
In the Random Vibration Accerelated Testing, is "a" the Fatigue strength at a specific stress cycle count? How does one determine "a" if a material and number of stress cycles is known? Seems a little washy to describe a factor in an expression as a "representative constant from a curve"... Is "a" a specific factor in the S-N curve or the result of the S-N curve or something else?
how to identify activation energy? If I consider example of PCB enclosure along with enclosure and connector. Each might have different activation energy based on their failure mechanism in temperature environment (Suppose temperature only). I have read that 0.7eV is typical one. could you please guide?
for the stress exaggeration factor, regarding the power model, do you use 2 X as a limit, or 4 X as a limit, makes a big difference, as another person commented, seems that there isn't much "science" when it comes to deciding how much of a stress exaggeration factor to use.
Q1: Is there anything that will change in the equations when testing one material against another, such as plastic, aluminum, etc.? Q2: If, for example, I find from the equations that the life acceleration test is 3 months versus 10 years, can I apply these 3 months intermittently? Due to a power outage to the simulator, would this be harmful? Finally, I would like to thank you personally for the interest you are sharing. Thank you very much...great job
Hi Louie, Great question! Not quite. ELFR and HTOL are single stress input tests where temperature is typically increased to an accelerated level. HASS is a screening test that typically combines rapid temperature cycling, vibration, and worst case operating parameters for electrical/electronic devices.
Depending on what type of stress you are trying to accelerate and what you are trying to test (component, sub assembly, system, etc) there are many standards to look at including JDEC and MIL-STDs as examples.
im doing my masters degree thesis or whatever the right term in english is so i need all the help i can get! LOVE THE VIDEO AND THANK YOUUU
Amazing videos with nice examples and really helpful for interview preparation
The equations are the easy part. The hard part is determining the correct activation energy. Any recommendations for books/training on how determine the correct activation energy?
Hi Don,
The only accurate ways to really determine activation energy is to hope the supplier of your material has this data, otherwise, you will need to run a temperature experiment yourself and note the point where a change or reaction occurred.
There's general equations for different types of materials that you can find in textbooks and Google searches, but it is always best to get this data directly from the supplier or characterize the data yourself.
Great video. Thank you. I would like to know how to find a stress factor. The physics model like Arrhenius assumes mechanisms of failure known beforehand. But for some problems we do not know it.
Regarding the dust and sunlight tests, can you tell us how to apply the acceleration test to them... Thank you very much, our esteemed professor.
Thank you ! Great video and very helpful content...How do we choose the statistical
distribution of lifetimes ?
Hi Najoua,
Thank you for the compliment. It depends on the trend of your data to determine the distribution that provides the best curve fit. Does it have a constant failure rate? If yes, you might be able to use an exponential distribution.
Typically with electronics and materials, you will use some variation of a Weibull distribution. Check out my video on Weibull Analysis to learn more: ua-cam.com/video/ustgf9D7d5Q/v-deo.html
Cheers,
Tom
hey please tell me source of finding the suitable value for the constants like activation energy, power to the humidity ratio, coffin mission const, and power to the vibration level
My recommendation is to get material properties for you supplier. Using boiler plate values from the internet can result in a inaccurate AF.
For vibe, you need to know your mission profile for your vibration loads and accelerate to your desired time while staying within the max recommended threshold for performing the time compression.
In the Random Vibration Accerelated Testing, is "a" the Fatigue strength at a specific stress cycle count? How does one determine "a" if a material and number of stress cycles is known? Seems a little washy to describe a factor in an expression as a "representative constant from a curve"... Is "a" a specific factor in the S-N curve or the result of the S-N curve or something else?
how to identify activation energy? If I consider example of PCB enclosure along with enclosure and connector. Each might have different activation energy based on their failure mechanism in temperature environment (Suppose temperature only). I have read that 0.7eV is typical one. could you please guide?
Check with your supplier to see if they have this data available, otherwise you will need to run experiments to identify your eV.
for the stress exaggeration factor, regarding the power model, do you use 2 X as a limit, or 4 X as a limit, makes a big difference, as another person commented, seems that there isn't much "science" when it comes to deciding how much of a stress exaggeration factor to use.
great video man!
Q1: Is there anything that will change in the equations when testing one material against another, such as plastic, aluminum, etc.?
Q2: If, for example, I find from the equations that the life acceleration test is 3 months versus 10 years, can I apply these 3 months intermittently? Due to a power outage to the simulator, would this be harmful?
Finally, I would like to thank you personally for the interest you are sharing. Thank you very much...great job
Helpful, thanks a lot!
Hey Tom, when you mentioned HASS as a screening method, I immediately thought of ELFR or HTOL burn-in. Are they the same thing?
Hi Louie,
Great question!
Not quite. ELFR and HTOL are single stress input tests where temperature is typically increased to an accelerated level.
HASS is a screening test that typically combines rapid temperature cycling, vibration, and worst case operating parameters for electrical/electronic devices.
Is there some way or law to speed up the sun test time?
is there any standard describing Accelerated life testing
Depending on what type of stress you are trying to accelerate and what you are trying to test (component, sub assembly, system, etc) there are many standards to look at including JDEC and MIL-STDs as examples.
FANTASTIC
Thank you!
応援してます🌏
Thank you!
Can you answer my previous questions, Mr. Tom?
keep rocking
Thank you!
great stuff
Thank you!
Amazing
Thank you!
nice video
Thank you!
keep going
Thank you!
Nice
Thank you!
wow liked
Thank you!
hmm ok I liked it
Thank you!
wooow
Thank you!
dsf
bh
Heimaa..
ytdkidu