Fatigue Modifying Factors

[SlipperyPete]

Can I ask a couple of questions about Fatigue Modifying factors when performing a fatigue analysis of an aluminium component:
1. Temperature Factor - I have seen it recommended that a temperature factor of 1 is applied for temperatures up to 450degC. Whereas I have also seen it proposed that the ratio of UTS at operating temperature/UTS at room temperature should be used. Obviously with aluminium there is a huge difference between these factors. What would members on this site recommend?

2. Reliability Factor - Generally a reliability factor of 0.753 is applied to achieve a reliability of 0.999 (3-sigma). However, if the specification requirement has applied a Scatter Factor of 6 to the required number of cycles is it also appropriate to apply a 3-sigma reliability factor, or is this simply applying factors on factors?

Any advise gratefully received, as I seem to end up with some very low material fatigue limits.

Peter

 

[rb1957]

"Generally a reliability factor of 0.753 is applied to achieve a reliability of 0.999 (3-sigma)." this is a factor on load (not a factor on life), still a pretty samll safe life factor ... 1/.753 = 4/3, (4/3)^4 = 256/81 = 3.

if the customer specifies a safe life factor of 6 (that's high !) then IMHO that should be enough.

Quando Omni Flunkus Moritati

 

[bxtsafe]

Hi SlipperyPete and welcome to the forum!

1) Regarding the temperature factor, use this equation to calculate the temperature factor for ALUMINUM ALLOYS:

For T<=50[sup]o[/sup]C use => C[sub]T[/sub] = 1.0
For T>50[sup]o[/sup]C use => C[sub]T[/sub] = 1- 0.0012.(T-50)[sup]2[/sup]

This specification is the one speficied by FKM-Guideline.


2) I din't undertand your second question.

 

[rb1957]

1) you're an aerospace guy ... use AR-MMPDS. i think the ^2 expression above is good up to about 80deg (when Ct = 0) ... i think there's something wrong with that expression, from AR-MMPDS Ct = 0.95 at 100degC ??

Quando Omni Flunkus Moritati

 

[SlipperyPete]

Yes, I was also a bit confused by the expression, particularly as our operating temperature is >90degC.
I've not seen fatigue temperature factors defined in MMPDS, can you provide a reference for the Ct=0.95 that you quoted?

Many thanks,

Peter

 

[rb1957]

sorry, thought you were looking for Ftu. i can't think of seeing high temperature fatigue for Al ... maybe run your own test ? maybe get a nearby university intereted in helping you ... it'd be an expensive rig!

Quando Omni Flunkus Moritati

 

[rb1957]

why use Al for high temps (like 450deg)? ... shouldn't you be using steel or Ti ?

Quando Omni Flunkus Moritati

 

[SlipperyPete]

We wouldn't use Al. for those temperatures, it's strength would be non-existant. Our application is about 95degC.

 

[rb1957]

useful piece of data ...
at 95C, i might use the ftu (or fty) ratio, looks like a 4% reduction (in the s/n curve). as a stress number, that's pretty small, well within lots of other uncertainities, i'd probably also use an increased safe life factor ... 10? (typically 4 or 5). depends how critical the item is.

Quando Omni Flunkus Moritati

 

[bxtsafe]

I've got this expression from the book Metal Fatigue Analysis Handbook (equation 4.41), and the book's authors say they got this equation from FKM-Guideline. I haven't realized this but yes, there's something wrong with the equation; I think the authors made a mistake when they wrote the book. I guess the correct expression is without the exponent:

For T<=50[sup]o[/sup]C use => C[sub]T[/sub] = 1.0
For T>50[sup]o[/sup]C use => C[sub]T[/sub] = 1 - 0.0012.(T-50)