Component Fatigue Data Statistical Analysis

[bcats4life]

Hello,

I am looking for some statistical methods that may be applicable to a study that I am performing. I am currently performing some vibratory high cycle fatigue (HCF) testing on some parts in a 1F mode. I need to demonstrate that the parts have a certain amount of HCF strength, for example 10 ksi. So far I have tested 9 parts and all parts have demonstrated to “runout or pass” at 15 ksi or above. The issue that I have is that I have been unable to break the parts in the mode of interest; unfortunately changing the mode is not an option. If I were able to break the parts, that would tell me what the actual HCF strength is. Instead, once I get to levels between 20 ksi – 35 ksi, varies for each part, the part starts coupling and I can no longer run in the mode of interest.

Enough background, here is my statistical question. The two options that I think I have with the current data is to perform some type statistical analysis that analyzes the go no-go data (I have 9 parts that pass and 0 parts that do not, what does this tell me statistically) or I could do some type of analysis that uses the known minimum HCF levels that I have determined for each part (for example, I know that 4 of the parts that have a strength of at least 15 ksi, 3 parts have a strength of at least 20 ksi, 1 part has a strength of at least 25 ksi, and one part has a strength of at least 35 ksi). If anyone has any ideas on some type of statistical analysis that I can do to demonstrate that I have plenty of HCF margin (compared to my 10 ksi requirement) which would give me confidence when these parts start in production that I will always meet my 10 ksi limit, please feel free to share. If there are other types of options that may be applicable for my study, please also share those ideas.

I have also posted this in the Statistics forum.

Thank you.

 

[GregLocock]

Look at Weibull analysis. How are you turning no. of cycles into ksi?




Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[unclesyd]

Though I've never run this type fatigue test, I believe that you can find what you want at the Fatigue Calculator.

Courtesy of metengr

http://www.fatiguecalculator.com/probabilistic/index.htm

Allow Java Script and popups.

 

[CoryPad]

I concur with Greg - look to Weibull.

Regards,

Cory

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[bcats4life]

Thank you for suggesting to look into Weibull analysis. It appears that a Weibaye or Weibest analysis may work for my case. One question I have on this is can a Weibull analysis be used with stress on the x-axis rather than a time based x-axis? All plots and examples that I have seen have always used some type of time based unit on the x -axis. I have stress data that I can plot on the x-axis but am unsure if this is applicable for a Weibull analysis. Any help or reassurance would be great.

Thanks

 

[GregLocock]

weibull works on numbers of cycles. As I asked in my first reply, how are you turning cycles into ksi?



Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[bcats4life]

I am not exactly turning number of cycles into ksi. I am doing stress-step testing. I first test the part at 5 ksi for 10e7 cycles. If the part does not fail within the 10e7 cycles, I increase the stress to 10 ksi and again run to 10e7 cycles. If the part again does not fail, I again increase the stress by 5 ksi and run to 10e7 cycles. This process will continue until finding a stress level that the part fails at.

 

[GregLocock]

Ah, I don't know really enough about Weibull to say, but if your test sequence is 5, 10, 15 then the 5 and 10 are such a tiny part of the fatigue dose compared with the 15 (assuming you get a reasonable number of cycles at 15) that you can ignore them. Does the weibull site not discuss this?





Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[rb1957]

i'd suggest taking a look at MIL HDBK 5, they present fatigue data on materials and derive statistical best fits.

 

[TVP]

You may want to look into staircase methods for fatigue evaluation. ISO 3800 has a nice description of the theoretical background as well as the practical implementation for use with fatigue testing of threaded fasteners. The following presentation discusses the staircase method as well:

http://www.fatigue.org/Minutes/Spring-2005/Engler SAE 2005-01-0802.pdf

Google "staircase fatigue" for more info.

 

[IceStationZebra]

"what does this tell me statistically" They all pass and the HCF is >15ksi. Short of FEA or changing the test I don't know how you can determine the actual HCF level.

Are all your samples from the same batch? Multiple batches could give you more confidence and explain your scatter. I'm curious why there is such a spread in the level in which the parts couple?

Can you do some other mechanical test to help quantify the material? My first thought is some type of pull test.

ISZ

 

[IRstuff]

From my answer to your other posting:
Weibull's seminal paper doesn't really discuss in terms of time:

http://www.barringer1.com/aug01prb.htm

Lots of stuff on the Barringer website:

http://www.barringer1.com/wa.htm

http://www.barringer1.com/Papers.htm

TTFN

FAQ731-376